Patent Application
20160344194
At least some embodiments of this disclosure relate to photovoltaic systems in general, and more particularly but not limited to, improving the energy production performance of photovoltaic systems.
A traditional maximum power point tracking (MPPT) algorithm sees a solar array as if it were a single solar module. MPPT can pull and push current on all strings and solar modules in a solar array in an equivalent fashion. As such, if solar modules in the solar array operate at different working points on the I-V curve, due to differences in installation, fabrication, or degradation over time, an MPPT algorithm may not be able to find the maximum power point (MPP) for the solar array.
U.S. Pat. No. 7,602,080, issued on Oct. 13, 2009 and entitled “Systems and Methods to Balance Solar Panels in a Multi-Panel System,” discloses a local management unit for a solar module. The local management unit has a controller for controlling the operation of the solar module and a link module unit to provide connectivity to a power bus for energy delivery and/or for data communications. The entire disclosure of U.S. Pat. No. 7,602,080 is incorporated herein by reference.
Systems and methods to control power generators individually with reduced cost are described herein. Some embodiments are summarized in this section.
In one of many embodiments disclosed in the disclosure, an apparatus includes a plurality of direct current converters with each converter connected to a respective solar panel of a plurality of solar panels and configured to receive direct current power from the panel connected thereto. The direct current outputs of the direct current converters are connected in series to provide power generated by the solar panels.
In one embodiment, the apparatus includes a control circuitry including a communications unit to receive control signals from a location remote from the control circuit, and a controller coupled with the communications unit to control the direct current converters according to the control signals. The control signals can include duty cycle signals used to operate each of the respective converters connected to the controller. In one embodiment, the apparatus can further include sensors configured to measure operating states of the solar panels; and the communications unit is configured to transmit measurements generated by the sensors to a remote management unit from which the control signals are received.
In one of many embodiments disclosed in the disclosure, the apparatus is configured to be mounted on one solar panel of the plurality of solar panels connected thereto, and remaining solar panels of the plurality of solar panels are connected to the apparatus via wires that provide input to respective direct current converters. In one embodiment, the solar panel to which the apparatus is secured is the panel physically located in the center position in the layout of the plurality of solar panels connected to the apparatus.
In one embodiment, an apparatus includes a plurality of direct current converters, with each converter of the direct current converters having an input configured to be connected to a separate photovoltaic panel of a plurality of photovoltaic panels, and a direct current output to provide electrical power generated by the separate photovoltaic panel. Outputs of the direct current converters are connected in series in the apparatus, and a single communications unit is coupled between the direct current converters and a system management unit. The system management unit is configured to control operations of the direct current converters via communicating with the communications unit.
In one embodiment, a plurality of controllers are coupled with the plurality of direct current converters respectively to control duty cycles of the direct current converters.
In one embodiment of the disclosure, systems include a plurality of solar panels, and a combined local management unit (CLMU) connected between the plurality of solar panels and a first output that is configured to provide electrical power generated by the plurality of solar panels. The CLMU includes a plurality of converters each connected to a respective solar panel of the plurality of solar panels to receive direct current input and to provide a second direct current output. The second direct current outputs from the converters are connected in series to provide the first output.
In one embodiment, the CLMU further includes a communications unit configured to communicate with a remote management unit separate from the CLMU. A controller of the CLMU is coupled with the communications unit to control operations of the converters in accordance with communications with the remote management unit.
In one embodiment, the CLMU is secured to a single solar panel of the plurality of solar panels. The CLMU can be secured to the solar panel physically located at a middle position of the plurality of solar panels. In one embodiment, the CLMU can include a current sensor, a plurality of voltage sensors, and a remote management unit. The current sensor can be configured to measure a direct current provided by the first output. The plurality of voltage sensors can be configured to measure voltages provided by the plurality of solar panels. The remote management unit can be configured to communicate with the communications unit of the CLMU to control operations of the plurality of converters for the plurality of respective solar panels. In one embodiment, the remote management unit can be further configured to receive measurements indicating operating states of the solar panels, determine duty cycles of the converters, and communicate data identifying the duty cycles to the CLMU's communications unit. The controller can be configured to operator the converters in accordance with the data identifying the duty cycles.
In one embodiment, the communications unit is configured to communicate with the remote management unit via a wired connection for providing the first output. Alternatively or in combination, the communications unit is configured to communicate with the remote management unit through a wireless connection.
In one embodiment, the remote management unit can be configured to instruct the controller, via the communications unit, to operate the converters in states to maximize power generated by the plurality of solar panels.
In one embodiment, the CLMU connects the plurality of solar panels to form a first string of solar panels. The system can have a plurality of strings, including the first string. The embodiment can further include a parallel direct current bus configured to receive outputs from a plurality of strings in parallel.
In one embodiment, the system includes a combiner box housing the parallel direct current bus. The combiner box can include one or more receptacles configured to receive a removable modular unit having a plurality of first terminals and a second terminal. When a first string is connected to a receptacle in the combiner box, the removable modular unit is configured to receive first power from the first string via the plurality of first terminals, and feed the first power to the parallel direct current bus through the second terminal. In one embodiment, the removable modular unit is further configured to measure power input received at the respective first terminals, selectively disconnect the first string from the parallel direct current bus based on communications with the remote management unit, or convert a voltage of the first string for outputting to the parallel direct current bus.
The disclosure includes methods and apparatuses which perform these methods, including data processing systems which perform these methods, and computer readable media containing instructions which when executed on data processing systems cause the systems to perform these methods.
Other features will be apparent from the accompanying drawings and from the detailed description which follows.
The embodiments are illustrated by way of example and not limitation in the figures of the accompanying drawings in which like references indicate similar elements.
The following description and drawings are illustrative and are not to be construed as limiting. Numerous specific details are described to provide a thorough understanding. However, in certain instances, well known or conventional details are not described in order to avoid obscuring the description. References to one or an embodiment in the present disclosure are not necessarily references to the same embodiment; and, such references mean at least one.
Reference in this specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosure. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Moreover, various features are described which may be exhibited by some embodiments and not by others. Similarly, various requirements are described which may be requirements for some embodiments but not other embodiments. As a result, this specification represents a disclosure of all possible combinations of features described herein, except that certain combinations are excluded by reasons of mutually exclusive relationships in features, where the mutual exclusiveness is either explicitly identified in this specification or is apparent from the description of the respective features.
When solar modules are connected in series or mesh configuration, there can be a problem in which weaker modules not only produce less energy but also affect other modules in the same string or wiring section. By measuring one can determine that a few modules are weaker than the others in most commercially installed strings. Thus, the string is generating less power than the sum available at each module if modules were operated separately.
At least one embodiment of the present disclosure provides methods and systems to switch on and off weak modules in the string in a way that the current on the string bus from the good modules won't be affected by the weak modules.
The present invention allows transmission of data from solar modules to a central (or system controller management) unit and other local management units in an energy production or photovoltaic system without adding significant cost. One embodiment of the present invention involves using the typically undesired electrical noise produced when operating local management units (sometimes referred to as “controllers” or “converters”) to act as a carrier system for data to be transferred. As there are a multitude of solar modules, each can be run on a slightly different frequency. Such an approach allows a receiver in the energy production or photovoltaic system to identify the carrier signal of each local management unit separately. This approach has the added benefit of reducing the overall system noise, because each local management unit sends “noise energy” in a different part of the spectrum, thus helping to avoid peaks.
In
The local management unit (LMU) (101) can include a solar module controller to control the operation of the solar module (102) and/or a link module unit to provide connectivity to the serial power bus (103) for energy delivery and/or for data communications.
In one embodiment, the command to control the operation of the switch Q1 (106) is sent to the local management unit (101) over the photovoltaic (PV) string bus (power line) (103). Alternatively, separate network connections can be used to transmit the data and/or commands to/from the local management unit (101).
In
In
In
In one embodiment, the controller (109) receives the parameters (104a, 104b, 104c) from a remote management unit via the serial power bus (103) or a separate data communication connection (e.g., a separate data bus or a wireless connection). In some embodiment, the controller (109) can communicate with other local management units connected on the serial power bus (103) to obtain operating parameters of the solar modules attached to the serial power bus (103) and thus compute the parameters (e.g., 104a and 104b) based on the received operating parameters. In some embodiment, the controller (109) can determine the parameter (e.g., 104a and 104b) based on the operating parameters of the solar module (102) and/or measurements obtained by the controller (109), without communicating with other local management units of other solar modules, or a remote system management unit.
In
In
In
In
In some cases, a filter (not shown), including a serial coil and a parallel capacitor, is also used. The filter can be placed at the local management unit or placed just before the fuse box or inverter, or be part of either one of those.
In
In one embodiment, the controller (109) is coupled to the solar module (102) in parallel to obtain power for processing; and the controller (109) is coupled to the serial power bus (103) to obtain signals transmitted from other management units coupled to the serial power bus (103).
By switching the module (102) (or groups of cells, or a cell) on and off to the string periodically, the local management unit (101) can lower the voltage reflected to the string bus (103) (e.g., a lower average voltage contributed to the string bus) and can cause the current reflected to the string bus (103) to be higher, nearer the level it would be if the module was not weak, generating a higher total power output.
In one embodiment, it is preferable to use different phases to operate the switches in different local management units on a string to minimize voltage variance on the string.
In
In
In
In
In one embodiment, the controller (109) is connected (not shown in
In one embodiment, the system management unit (204) is part of the inverter (203), the combiner box (206), a local management unit, or a stand-alone unit. The solar modules (201a, 201b, . . . , 201n) are connected in parallel to the local management units (202a, 202b, . . . , 202n) respectively, which are connected in series to form a string bus (205), which eventually is connected to an inverter (203) and the management unit (204). The solar module (201a), for example, is connected to the local management unit (202a) by the terminals (182, 184, 186) (
In
In one embodiment, beyond the module connection the local management units can have the signal inputs, including but not limited to duty cycle (104a), phase (104b) and synchronization pulse (104c) (e.g., to keep the local management units synchronized). In one embodiment, the phase (104b) and the synchronization pulse (104c) are used to further improve performance, but the local management unit (101) can work without them.
In one embodiment, the local management unit can provide output signals. For example, the local management unit (101) can measure current and voltage at the module side and optionally measure current and voltage in the string side. The local management unit (101) can provide other suitable signals, including but not limited to measurements of light, temperature (both ambient and module), etc.
In one embodiment, the output signals from the local management unit (101) are transmitted over the power line (e.g., via power line communication (PLC)), or transmitted wirelessly.
In one embodiment, the system management unit (204) receives sensor inputs from light sensor(s), temperature sensor(s), one or more each for ambient, solar module or both, to control the photovoltaic system (200). In one embodiment, the signals can also include synchronization signals. For example, a management unit can send synchronization signals periodically to set the timing values, etc.
Using the described methods the local management unit can be a very non-expensive and reliable device that can easily increase the throughput of a photovoltaic solar system by a few (e.g., signal or low double digits) percentage points. These varied controls also allow installers using this kind of system to control the VOC (open circuit voltage) by, for example by shutting off some or all modules. For example, by using the local management units of the system, a few modules can be disconnected from a string if a string is getting to the regulatory voltage limit, thus more modules can be installed in a string.
In some embodiments, local management units can also be used within the solar panel to control the connection of solar cells attached to strings of cells within the solar panel.
In
Some embodiments of the disclosure include methods to determine the duty cycles and/or phases for local management units connected to a string or mesh of solar modules.
In some embodiments, the duty cycle of all local management units in a string or mesh can be changed, to increase or decrease the string voltage. The duty cycles can be adjusted to avoid exceeding the maximum voltage allowed. For example, the maximum voltage can be limited by the combiner box (206), the inverter (203), or any other load connected to the string bus (205), or limited by any regulations applicable to that system. In some embodiments, the duty cycles are adjusted to align the voltage of multiple strings.
In some embodiments, the duty cycle of one local management unit (101) in a string can be changed to cause higher current in that local management unit (101) and overall higher power harvesting.
In one embodiment, the duty cycles are computed for the solar modules that are connected to a string via the corresponding local management units. The duty cycles can be calculated based on the measured current and voltages of the solar modules and/or the temperatures.
After an initial set of duty cycles is applied to the solar modules, the duty cycles can be further fine tuned and/or re-adjusted to changes, such as shifting shading etc., one step a time, to improve power performance (e.g., to increase power output, to increase voltage, to increase current, etc.). In one embodiment, target voltages are computed for the solar modules, and the duty cycles are adjusted to drive the module voltage towards the target voltages.
The methods to compute the duty cycles of the solar modules can also be used to compute the duty cycles of the groups of solar cells within a solar module.
In
For example, the duty cycle can be adjusted to increase the current in the string and/or the solar energy production unit, to increase the output power of the string and/or the solar energy production unit, to increase the voltage of the solar energy production unit, etc.
In
In
The duty cycle of the first solar panel is adjusted (427) to improve the performance of the first solar energy production unit and/or the string, until a decrease in the operating voltage of the second solar panel is detected. For example, the duty cycle of the first solar panel can be adjusted to increase the total output power of the string, to increase the current of the string, to increase the current of the first solar panel, to drive the voltage of the first solar panel towards a target voltage, such as its maximum power point voltage estimated based on its current operating parameters, such as temperature or a voltage calculated using its estimated maximum power point voltage.
In
In
In one embodiment, the duty cycle of the second solar panel is repeatedly decreased (429) until it is determined (431) that the decrease (429) in the duty cycle of the second solar panel cannot increase the voltage of the second solar panel.
In
Alternatively, the target voltage can be set as the first maximum power point voltage of the first solar panel.
In one embodiment, to adjust voltage a same factor is applied to all modules in that string. For example, in a case of a first module A1 that is producing only 80%, and the voltage of the whole string needs to be 5% lower, the duty cycle of A1 is 80% multiplied the duty cycle applied to the whole string (which is Y in this example) so module A1 then has Y×0.8 as duty cycle.
In some embodiments, the system management unit (204) and/or the local management units (e.g., 202a, 202b, . . . , 202n) are used solely or in combination to determine the parameters to control the operations of the switches.
For example, in one embodiment, a system management unit (204) is the “brain” of the system, which decides on the duty cycle and phase parameters.
For example, in another embodiment, each local management unit broadcasts information to the other local management units on the string to allow the individual local management units to decide their own duty cycle and phase parameters.
In some embodiment, a local management unit can instruct one or more other local management units to adjust duty cycle and phase parameters. For example, the local management units on a string bus (205) can elect one local management unit to compute the duty cycle and phase parameters for other local management units on the string.
For example, in some embodiment, the system management unit (204) can determine one or more global parameters (e.g., a global duty cycle, the maximum power on the string, the maximum voltage on the string, etc.), based on which individual local management units adjust their own duty cycles.
In some embodiments, a local management unit can effectively self manage and determine its own duty cycles without relying upon communicating with other management units. For example, the local management unit can adjust its duty cycle for connecting its solar module to the string to operate the solar module at the maximum power point. No local management unit is in control over the system, and each adjusts its own duty cycle (and thus, its power and voltage.)
In one embodiment, module voltages are measured by the local management units in the same string at substantially/approximately the same time and used to identify the strongest solar module. A strongest solar module provides the most power in the string. Since the modules are connected in series, the solar module having the highest module voltage in the string can be identified as the strongest solar module. In some embodiment, the operating voltage and current of the solar module are measured to determine the power of the solar module.
Additional approaches can be implemented to control the voltage, power output, or the efficiency of one or more strings of solar module controllers as described above. In some embodiments, a system controller management unit controls the operation of a plurality of local management units in one or more strings. In some embodiments, one or more local management units control the operation of a plurality of local management units in one or more strings. In some embodiments, the local management unit can only control its own operation, or can control the operation of itself and other local management units in the same string.
One or more local management units in a string can have the capability to control the operation of other local management units in the same string. In one embodiment, a single local management unit can be selected to be a controlling local management unit to control a plurality of panels in a string. The controlling local management unit in a string can be selected using any suitable protocol. In one embodiment, in a string of local management units, the first local management unit that announces its intent to take control of other modules in the string could become the controlling local management unit.
In one embodiment, to improve power output by a string, one or more local management units can each receive module voltage from all local management units in the same string and identify the strongest local management unit (i.e., the one with the maximum power and voltage). Each local management unit can then set its own duty cycle as a function of the received voltage.
In one embodiment, after the highest module voltage Vm in the string is identified, the duty cycle for each module can be computed as a function of a ratio between the module voltage V of the module and the highest module voltage Vm. For example, the duty cycle for a module can be computed as 1−((Vm−V)/Vm)=V/Vm. In one embodiment, a particular local management unit receives the voltages of all other local management units at the same time or substantially same time (e.g., all voltages are received within an interval of less than one second.)
In one embodiment, the system management unit (204) can identify the highest module voltage from the module voltages received from the local management units (202a, 202b, . . . , 202n), and compute the duty cycles for the corresponding local management units (202a, 202b, . . . , 202n).
In one embodiment, the local management units (202a, 202b, . . . , 202n) can report their module voltages on the string bus (205) to allow individual local management units (202a, 202b, . . . , 202n) to identify the highest module voltage and compute the duty cycles, without relying upon the system management unit (204).
In one embodiment, one of the local management units (202a, 202b, 202n) can identify the highest module voltage and/or compute the duty cycles for the other local management units (202a, 202b, . . . , 202n).
In one embodiment, the duty cycles are determined and/or adjusted periodically (e.g., every 30 seconds). The intervals can take into account various environmental factors (e.g., where shadows on a solar panel are cast on different parts of the panel over the course of a day).
In one embodiment, after the duty cycles for the solar modules on the string are set based on the module voltage ratio relative to the highest module voltage in the string, the duty cycles can be fine tuned to increase the power performance. The duty cycles can be fine tuned one step a time, until a decrease of voltage of the module with the highest power is detected. In response to the detected decrease, the last change that caused the decrease can be reversed (undone). The fine tuning of the duty cycles can be used to reach the peak performance point (e.g., for maximum power point tracking).
In one embodiment, after the strongest module is identified, the duty cycles of the solar modules on the string are adjusted until the module with the highest power in the string decrease its voltage. Since decreasing the duty cycle of a solar module decreases the time period the module is connected to the string and thus increases its voltage, the duty cycle of the module with the highest power in the string can be decreased to increase its voltage, in response to the decrease in its voltage caused by the adjustment to the duty cycles of other solar modules on the string. For example, the duty cycle of the module with the highest power in the string can be decreased until its voltage is maximized.
The performance of solar modules can vary significantly with temperature. A system capable of measuring temperature can implement methods for controlling the voltage, power output, or the efficiency of one or more strings of solar module controllers using module temperature as a factor. In one embodiment, the local management unit measures module and ambient temperatures for some methods to determine the duty cycles. For example, the operating parameters measured at the local management units (e.g., 202a, 202b, . . . , 202n), such as module temperature, can be used to compute the estimated voltages of the solar modules at their maximum power points. For example, a formula presented by Nalin K. Gautam and N. D. Kaushika in “An efficient algorithm to simulate the electrical performance of solar photovoltaic arrays,” Energy, Volume 27, Issue 4, April 2002, pages 347-261, can be used to compute the voltage Vmp of a solar module at the maximum power point. Other formulae can also be used. Once the maximum power point voltage Vmp of a solar module is computed or estimated, the duty cycle of the solar module connected to a string can be adjusted to drive the module voltage to the computed/estimated maximum power point voltage Vmp, since decreasing the duty cycle of a solar module normally increases its voltage.
In one embodiment, a local management unit can adjust the duty cycle of the solar module connected to the local management unit to change the module voltage to the computed/estimated maximum power point voltage Vmp, without having to communicate with other management units.
In one embodiment, a local management unit (or a system management unit) can adjust the duty cycle of the solar module connected to the local management unit to perform maximum power point tracking.
In one embodiment, after identifying the strongest module and computing/estimating the maximum power point voltage Vmpm of the strongest module, the duty cycle for each module on a string can be computed as a function of a ratio between the maximum power point voltage Vmp of the module and the maximum power point voltage Vmpm of the strongest module. For example, the duty cycle for a module can be computed as 1−((Vmpm−Vmp)/Vmpm)=Vmp/Vmpm. The duty cycle can be periodically updated, based on the current operating parameters measured, and/or fine tuned until a decrease in the voltage of the strongest module is detected.
Alternatively, a target voltage for each module on the string can be computed as a function of a ratio between the maximum power point voltage Vmp of the module and the maximum power point voltage Vmpm of the strongest module. For example, the target voltage for a module can be computed as Vm×Vmp/Vmpm, where Vm is the measured voltage of the strongest module. The duty cycle of the module can be changed to drive the module voltage of the module towards the target voltage.
In one embodiment, after identifying the strongest module and computing/estimating the maximum power point power Pmpm of the strongest module, the duty cycle for each module on a string can be computed as a function of a ratio between the maximum power point power Pmp of the module and the maximum power point power Pmpm of the strongest module. For example, the duty cycle for a module can be computed as 1−((Pmpm−Pmp)/Pmpm)=Pmp/Pmpm. The duty cycle can be periodically updated, based on the current operating parameters measured, and/or fine tuned until a decrease in the voltage of the strongest module is detected, since decreasing the duty cycle normally increases the module voltage.
In one embodiment, a target voltage for each module on the string can be computed as a function of a ratio between the maximum power point power Pmp of the module and the maximum power point power Pmpm of the strongest module. For example, the target voltage for a module can be computed as Vm×Pmp/Pmpm, where Vm is the measured voltage of the strongest module. The duty cycle of the module can be changed to drive the module voltage of the module towards the target voltage, since decreasing the duty cycle normally increases the module voltage.
In one embodiment, the duty cycle for each local management unit is changed to increase the current of the solar module attached to the local management unit (e.g., based on the measurement of the voltage and current of the solar module), until the maximum current is achieved. This method assumes that string maximum power can be achieved with some accuracy by driving each local management unit to maximum current. In one embodiment, the voltages and currents of the solar modules are measured for tuning the duty cycles for maximum power point tracking for the string. The measurements of the voltages and currents of the solar modules also enable the local management units to additionally serve as a module level monitoring system.
The duty cycles can be adjusted by the system management unit (e.g., 204) based on the measurements reported by the local management units (e.g., 202a, 202b, . . . , 202n), or adjusted directly by the corresponding local management units (e.g., 202a, 202b, . . . , 202n).
In one embodiment, during the process of setting and/or tuning the duty cycles, the maximum power point tracking operation by the inverter (203) is frozen (temporarily stopped). Light intensity at the solar modules is monitored for changes. When the light intensity at the solar modules stabilizes, the voltage and current of the solar modules are measured for the determination of the duty cycles. Then normal operation resumes (e.g., unfreezing of maximum power point tracking operation).
In one embodiment, the local management units measure the voltages and currents of the solar modules to determine the power of the solar modules. After identifying the highest power Pm of the solar module on the string, the duty cycles of the solar modules on the string are determined by the power ratio relative to the highest power Pm. For example, if a module produces 20 percent less power, it will be disconnected from the string bus about 20 percent of the time. For example, if a module produces power P, its duty cycle can be set to 1−((Pm−P)/Pm)=P/Pm.
In one embodiment, a predetermined threshold is used to select the weak modules to apply duty cycles. For example, in one embodiment, when a module produces power less than a predetermine percent of highest power Pm, a duty cycle is calculated and applied to the solar module. If the module is above the threshold, the module is not disconnected (and thus having a duty cycle of 100%). The threshold can be based on the power, or based on the module voltage.
In one embodiment, the system management unit (204) finds the duty cycles for the local management units (202a, 202b, . . . , 202n) and transmits data and/or signals representing the duty cycles to the local management units (202a, 202b, . . . , 202n) via wires or wireless connections. Alternatively, the local management units (202a, 202b, . . . , 202n) can communicate with each other to obtain the parameters to calculate the duty cycles.
In one embodiment, the system management unit (204) knows all the different duty cycles indicated for the local management units (202a, 202b, . . . , 202n).
In one embodiment, during power fine tuning, the system management unit (204) sends the appropriate data/signal to the appropriate local management units (202a, 202b, . . . , 202n), and then the system management unit (204) calculates the total power of the string and corrects the duty cycle to produce maximum power. Once maximum power is achieved, the duty cycles for the local management units (202a, 202b, . . . , 202n) can be saved in a database and serve as a starting point for the corresponding local management units (202a, 202b, . . . , 202n) at the same time of day on the next day. Alternatively, a local management can store the duty cycle in its memory for the next day.
The stored duty cycles can be used when there is a fixed shade on the modules, such as a chimney, a tree, etc., which will be the same shade on any day at the same time. Alternatively, historical data may not be saved, but can be recalculated from scratch on each run, for example every 30 minutes.
In one embodiment, the light intensity at the solar modules is monitored for changes. The duty cycles are calculated when the light intensity does not change significantly. If there are changes in sun light radiation at the solar modules, the system will wait until the environment stabilizes before applying or adjusting the duty cycles.
In one embodiment, the system management unit (204) can communicate with the inverter as well. When the environment is not stable (e.g., when the sun light radiation is changing), the inverter can stop maximum power point tracking. In such a situation, the inverter can be set up for its load, instead of tracking for maximum power point. Instead of using the inverter to perform maximum power point tracking, the system management unit (204) and the local management units (202a, 202b, . . . , 202n) are used to set the operating parameters and balance the string.
Alternatively, when the environment is not stable but measurements and calculation are done faster than the MPPT is working, there can be no need to stop the MPPT on the inverter. Alternatively, when the environment is not stable, measurements can be taken few times for the same radiation until a stable result is achieved.
Many variations can be applied to the systems and methods, without departing from the spirit presented in the disclosure. For example, additional components can be added, or components can be replaced. For example, rather than using a capacitor as primary energy store, an inductor can be used, or a combination of inductor and capacitor. Also, the balance between hardware and firmware in the micro controllers or processors can be changed. In some cases, only some problematic modules can have a local management unit, for example in a shaded or partially shaded or otherwise different situation. In yet other cases, local management units of strong modules can be virtually shut off. The methods for determining the duty cycles for the solar modules can also be used to determine the duty cycles of groups of cells connected via local management units in a string within a solar panel/module.
In one embodiment, the receiver (301) can manage communications from all the local management units. In one embodiment, each local management unit can have its own receiver. In one embodiment, a receiver can be implemented in hardware (HW) only. In one embodiment, a digital radio can be used as the receiver, in which case an analog to digital converter (ADC) samples the signals and all the processing is done in a microcontroller or a digital signal processor using software (SW), or any combination of SW and HW.
In operation, switch Q1 and switch Q2 are controlled oppositely, and switch Q3 and switch Q4 are controlled oppositely. When switch Q1 is on, switch Q3 is on. When switch Q2 is on, switch Q4 is on. The current can be increased or decreased by adjusting switches Q1, Q2, Q3, Q4. A controller (not shown), suitably connected to a power supply, controls the operation of the switches Q1, Q2, Q3, Q4. In one embodiment, the controller can be off the shelf and possibly modified. In one embodiment, the controller can have analog circuitry. In one embodiment, the controller can be a microcontroller. In one embodiment, the controller could be a combination of these features. As discussed below, a phase shift can be created between the currents controlled by the switches Q1, Q2, Q3, Q4. The inductor LR and the capacitor CR constitute an LC (or tank) circuit. The primary winding Tp of the transformer T is coupled to the secondary winding Ts. Diode D1, diode D2, and capacitor Cout constitute a Delon rectifier circuit. In a positive cycle, diode D1 charges the upper capacitor of capacitor Cout. In a negative cycle, diode D2 charges the lower capacitor of the capacitor Cout. Vout is effectively two times the voltage across the secondary winding Ts of the transformer T.
The local management unit (1200) requires a reliable current limit because it is required to charge a large input capacitance reflected from the inverter (203). The local management unit (1200) needs to allow operation with low input and output capacitance, because reliability does not allow the use of aluminum capacitors due to their limited life expectancy. In many instances aluminum may not be suitable for the local management unit (1200) for reasons of reliability.
Efficiency of the novel topology of the local management unit (1200) should be higher than 96 percent at the range of 20 percent to 100 percent load. The topology of the local management unit (1200) should allow direct control of input impedance for smooth MPPT control, and should minimize the need for damping networks (i.e., snubbers) in order to limit EMI emissions to improve reliability and maximize efficiency. Further, the transformer should be protected from saturation. Isolation voltage must be higher than 2000V, and switching losses reduced (i.e., zero current switching/zero voltage switching). No load condition is to be defined during inverter turn on.
The local management unit (1200) achieves the aforementioned performance goals.
In
For loads higher than 15 percent of the maximum load, switches Q1, Q3 are operated together at 50 percent duty cycle, while switches Q4, Q2 are operated together at 50 percent duty cycle with no phase shift. Input power is controlled by changing operating frequency of the local management unit (1200) above and below the resonant frequency. Turn ratio of the primary winding Tp and secondary winding Ts is set according to MPPT voltage because at this voltage efficiency is at the highest point (i.e., zero voltage, zero current is achieved). For other frequencies, switching is performed at zero voltage because there is current in the primary winding Tp and resonant tank that is maintained, and this current causes voltage shift that allows turn-on to be performed at zero voltage.
Below 15 percent of load, the local management unit (1200) is operated in phase shift mode. In phase shift mode, switches Q1, Q2 are reversed, and switches Q3, Q4 are reversed. However, a phase shift causes switches Q3 and Q4 to conduct together part of the time, and likewise for switches Q1, Q4. A phase shift operation allows no load and light load control. As shown in, for example,
As shown in
In the local management unit (1200), a resonant tank provides a limit to the current through the primary winding Tp. A serial capacitor CR prevents transformer saturation. Output rectifier voltage is clamped to output voltage Vout allowing the use of 600V ultra fast diodes. There are no spikes across the switching transistors. Transformer parameters act as part of resonant tank. Input voltage range and efficiency are optimized for solar module operation by transformer turn ratio and transformer small air gap. Resonant frequency controls input impedance, which is the required control parameter for the application of separate solar modules operating against a fixed voltage inverter load in the system.
Operation of the local management units in
The communication transmission modulator (1910) modulates switching of the pulse width modulation (PWM) operation to transmit data from the local management unit (1900). Various modulation encoding schemes can be used, such as, for example, modified FM (MFM) and Manchester coding. In one embodiment, another modulating and encoding scheme can be used. In one embodiment, the communication transmission modulator (1910) represents the transmission portion of a modem (not shown) that is associated with the local management unit (1900). In one embodiment, the communication transmission modulator (1910) is part of the local management unit (1900). In one embodiment, the communication transmission modulator (1910) is external to the local management unit (1900).
This system allows the use of full duplex (two-way) communications. The receiver at the module side can be implemented within the module circuitry. The limitation of transmit and receive within same circuit does not exist. Transmission from management unit can be used to synchronize modules. Reliability is not affected by transmission. The effect on overall performance is very small because the transmission duty cycle from the module is low.
Weak solar modules in a string bus, and weak string buses in a solar array can bring down the total output power of the solar array. A weak string may produce less current, voltage, and/or power than other strings because one or more solar modules in the string are either partially/wholly in shade or are malfunctioning (or for other reasons). Traditional solar arrays may not be able to overcome this problem since the output of individual solar modules and string buses may not be controllable independently of other solar modules and string buses. The systems and methods herein disclosed monitor and adjust individual solar module outputs such that weak solar modules are balanced with strong solar modules in a string bus, and strong string buses are balanced with weak string buses in a solar array. Once balancing within string buses and between string buses has been accomplished, an inverter using maximum power point tracking (MPPT) can determine the maximum power point for the solar array.
Strong string buses are within a threshold voltage of a string output voltage of a strongest string, and weak string buses are not within the threshold voltage of a string output voltage of a strongest string. The strongest string bus operates at a higher voltage than the other string buses. The weakest string bus operates at a lower voltage than the other string buses.
Three I-V curves for a string bus will now be used to describe balancing solar modules on a string bus.
Three I-V curves for a solar array will now be used to describe balancing string buses in a solar array.
In an embodiment, the voltages of the strong string buses (3202) can be decreased until the string bus voltage balances with the voltage of the one or more weak string buses (3208, 3210). Since the weak string buses (3208, 3210) may not operate at the same maximum voltage (e.g., (3210)>(3208)), strong string buses (3202) can be balanced with an average of the weak string buses (3208), (3210).
It should be understood that
Balancing solar modules in a string bus means that the solar module currents converge. In an embodiment, balancing solar modules in a string bus means that weak solar module currents converge on strong solar module currents. Balancing string buses in a solar array means that string bus voltages converge. In an embodiment, balancing string buses in a solar array means that strong string bus voltages converge on weak string bus voltages. It should be understood that balancing need not mean an exact balance. Two values can only converge to within a threshold value of each other—a relative equivalency. For instance, two currents that are to be balanced by causing them to converge on 1 amp, can be considered balanced if they come within 0.1 amp of the 1 amp goal. Two currents that are to be balanced by causing them to converge can be considered balanced when they are within five percent of each other. However, even when balanced, the process is iterative and will continue indefinitely. This is because the solar module performance changes, the load changes, and local conditions (e.g., clouds, leaves, dirt, to name a few). Furthermore, every balancing of a string bus can require a balancing of solar modules on each string bus, and every balancing of solar modules on each string bus can require a balancing of string buses.
A “solar array” typically comprises two or more solar modules series-connected via a string bus where the output voltage is a sum of the voltages of the series-connected solar modules. In larger solar arrays, string buses can be connected in parallel such that their currents add. A combiner and inverter are not part of the solar array.
Balancing current outputs of solar modules (201a, 201b, . . . , 201n) on a string bus (205a, 205b, 205c) will now be discussed in more depth. The controller (204) can be configured to balance the currents produced by the solar modules (201a, 201b, 201n) on a given string bus (205a, 205b, 205c), and perform this balancing for each string bus (205a, 205b, 205c). As a result, the currents from the solar modules (201a, 201b, . . . , 201n) on a string bus (205a, 205b, 205c) can be balanced.
In order to balance solar modules (201a, 201b, . . . , 201n) on a string bus (205a, 205b, 205c), it can be useful to identify strong solar modules (201a, 201b, 201n) and weak solar modules (201a, 201b, . . . , 201n). This is done by varying the current on a string bus (205a, 205b, 205c), monitoring the resulting change in voltage in each solar module (201a, 201b, . . . , 201n), and comparing the changes in voltage on each solar module (201a, 201b, . . . , 201n) to identify strong solar modules (201a, 201b, . . . , 201n) and weak solar modules (201a, 201b, . . . , 201n).
Varying the current on the string bus (205a, 205b, 205c) can involve the inverter (203) pulling a different current from the string bus (205a, 205b, 205c). It can involve varying an impedance seen by the string bus (205a, 205b, 205c). For instance, the inverter (203) can vary the impedance that the string bus (205a, 205b, 205c) sees, and in doing so the current and voltage produced by the solar modules (201a, 201b, . . . , 201n) on the string bus (205a, 205b, 205c) will vary. In other words, pulling a different current or changing the impedance changes where on the I-V curve each solar module (201a, 201b, . . . , 201n) operates at. Since current for devices connected in series is the same, a change in current on the string bus (205a, 205b, 205c) causes the same change in current for each solar module (201a, 201b, . . . , 201n) on the string bus (205a, 205b, 205c). However, the changes in voltage may not be the same, since the solar modules (201a, 201b, . . . , 201n) can operate at different operating points on the I-V curve.
This can be seen in
By identifying strong and weak solar modules, based on the changes in voltage dV1, dV2, one can determine which solar module(s) (201a, 201b, . . . , 201n) to adjust. Strong solar modules (201a, 201b, . . . , 201n) can be used as a reference. Strong solar modules (201a, 201b, . . . , 201n) may not be adjusted, while weak solar module (201a, 201b, . . . , 201n) voltages can be decreased until their current outputs converge on the strong solar module (201a, 201b, . . . , 201n) outputs (or an average strong solar module (201a, 201b, . . . , 201n) current output). This raises the current output of the string bus (205a, 205b, 205c), while decreasing the string bus (205a, 205b, 205c) voltage output. However, the net effect is greater power output from the string bus since the loss in voltage is more than compensated for by the increased current. The end result can preferably be working points that are proximal for all solar modules (201a, 201b, . . . , 201n), that is balanced or near balanced current outputs. An indication that balancing has been achieved and that the solar modules (201a, 201b, . . . , 201n) are operating near the maximum current output of the strong solar modules (201a, 201b, . . . , 201n), is that a variation in the current along the string bus (205a, 205b, 205c) will cause a nearly equivalent change in voltage for each solar module (201a, 201b, . . . , 201n).
In an embodiment, instead of using strong solar modules (201a, 201b, . . . , 201n) as the reference, an average of all solar modules (201a, 201b, . . . , 201n) on a string bus (205a, 205b, 205c) can be used as a reference. In this embodiment, all solar modules (201a, 201b, . . . , 201n) on the string bus (205a, 205b, 205c) can be adjusted, including strong solar modules (201a, 201b, . . . , 201n), until their current outputs converge on the average. In an embodiment, the controller (204) can identify strong and weak solar modules (201a, 201b, . . . , 201n) of all solar modules (201a, 201b, . . . , 201n) on a string bus (205a, 205b, 205c).
In an embodiment, the change in voltage dVi for each solar module (201a, 201b, . . . , 201n) can be checked for anomalies, and those measurements appearing to be erroneous can be ignored or eliminated and replaced with a new measurement of dVi. In an embodiment, at least one solar module (201a, 201b, . . . , 201n) can be identified as a strong solar module (201a, 201b, . . . , 201n). In an embodiment, the solar modules (201a, 201b, . . . , 201n) identified as strong solar modules (201a, 201b, . . . , 201n) can be left out of the other steps involved in balancing a string bus (205a, 205b, 205c) (e.g., strong solar module (201a, 201b, . . . , 201n) current outputs may not be changed while the current outputs of weak solar modules (201a, 201b, . . . , 201n) are changed).
String buses (205a, 205b, 205c) can be connected in parallel and have an output that is optionally connected to a string combiner (206) (or fuse box or chocks box). In an embodiment, the output of the string buses (205a, 205b, 205c) can be connected to the inverter (203).
In an embodiment, the LMUs (202a, 202b, . . . , 202n) control the voltage and current provided to the string buses (205a, 205b, 205c) from the solar modules (201a, 201b, . . . 201n). In an embodiment, one LMU can control the voltage and current output for more than one solar module (201a, 201b, . . . , 201n). In an embodiment, the number of solar modules (201a, 201b, . . . , 201n) can exceed the number of LMUs (202a, 202b, . . . , 202n). For instance, LMUs (202a, 202b, . . . , 202n) can only be used to control the current and voltage output from solar modules (201a, 201b, . . . , 201n) identified as weak solar modules.
In an embodiment, a controller (204) can control the LMUs (202a, 202b, . . . , 202n). The controller (204) can also monitor the string buses (205a, 205b, 205c) and the solar modules (201a, 201b, . . . 201n) via the LMUs (202a, 202b, . . . , 202n). Data regarding the solar modules (201a, 201b, . . . 201n) and LMUs (202a, 202b, . . . , 202n) can be transmitted via the string buses (205a, 205b, 205c) to the controller (204). In an embodiment, the controller (204) can transmit instructions or commands to the LMUs (202a, 202b, . . . , 202n) via the string buses (205a, 205b, 205c). In another embodiment, the controller (204) can perform the above-noted communications with the LMUs (202a, 202b, . . . , 202n) via wireless communication paths. The controller (204) can also be in communication with the inverter (203). In the illustrated embodiment, the controller (204) is a standalone device. However, in other embodiments, the controller (204) can be a part of other devices (e.g., the inverter (203), or LMUs (202a, 202b, . . . , 202n). In an embodiment, the controller (204) can be a part of one of the LMUs (202a, 202b, . . . , 202n).
In an embodiment, operation of the controller (204) can be based on historical current and voltage data to help in pattern identification. For example, where the controller (204) notices that certain solar modules (201a, 201b, . . . , 201n) become weak solar modules (201a, 201b, . . . , 201n) at a specified time every day, this can be an indication of an object casting a predictable shadow over those solar modules (201a, 201b, . . . , 201n). As a result, instructions can be sent to the affected LMUs (202a, 202b, . . . , 202n) at the time when those LMUs (202a, 202b, . . . , 202n) regularly become weak.
Balancing current outputs of string buses (205a, 205b, 205c) will now be discussed in more depth. Although solar module (201a, 201b, . . . 201n) output current can be balanced on each string bus as described above, each string bus (205a, 205b, 205c) can produce different voltages (i.e., weak string buses can produce less-than-ideal or less-than-maximum voltages).
Since string buses (205a, 205b, 205c) can be connected in parallel, the voltages produced by the string buses (205a, 205b, 205c) can converge. This voltage convergence causes the working points of the solar modules (201a, 201b, . . . 201n) in each string bus (205a, 205b, 205c) to change.
In an embodiment, varying the voltage of the string buses in the solar array involves varying the current drawn from the string buses (205a, 205b, 205c) or varying an impedance seen by the string buses (205a, 205b, 205c). For instance, an inverter (203) connected to the string buses (205a, 205b, 205c) can vary the impedance that the string buses (205a, 205b, 205c) see, and in doing so the current and voltage produced by the solar modules (201a, 201b, . . . 201n) on the string bus (205a, 205b, 205c) will change. In other words, changing the impedance changes where on the I-V curve each string bus (205a, 205b, 205c), and the solar modules (201a, 201b, . . . 201n) on each string bus, (205a, 205b, 205c) operate at. A change in voltage on the string buses (205a, 205b, 205c) causes a change in the current output from each of the string buses (205a, 205b, 205c). However, since the string buses (205a, 205b, 205c) may not operate at the same operating point on the I-V curve, the change in voltage will cause differing changes in current for some or all of the string buses (205a, 205b, 205c).
This can be seen in
By comparing the change in currents dI1, dI2 one can determine which string buses (205a, 205b, 205c) to adjust. Adjusting string bus voltage output involves equally decreasing the voltage output of all solar modules on a string bus (205a, 205b, 205c), resulting in an increase of the current from the string bus (205a, 205b, 205c). In an embodiment, one or more weak string buses (205a, 205b, 205c) can be used as references such that all other string bus voltages are balanced with that of the one or more weak string buses (205a, 205b, 205c). In another embodiment, an average of weak string buses (205a, 205b, 205c) can be used as the reference. In an embodiment, an average of all string buses (205a, 205b, 205c) can be used as the reference.
Not all solar modules can be adjusted. For instance, if two or more string buses (205a, 205b, 205c) are producing about the same current, then those string buses (205a, 205b, 205c) can be used as references (the outputs from solar modules (201a, 201b, . . . 201n) on those string buses (205a, 205b, 205c) will not be changed). Solar module output currents on all other string buses (205a, 205b, 205c) can be such that the string bus output currents converge on the output from the reference string buses (205a, 205b, 205c).
Having identified weak string buses (205a, 205b, 205c), an average change in current dIw for the weak string buses (205a, 205b, 205c) can be determined. The average change in current dIw for the weak string buses (205a, 205b, 205c) can be a reference value. The change in current dIk for each strong string bus (205a, 205b, 205c) can be compared to the average change in current dIw for the weak string buses (205a, 205b, 205c). The difference between dIk for a string bus (205a, 205b, 205c) and dIw indicates by how much the string bus (205a, 205b, 205c) output current should be decreased in order to match the output current of the weak string buses (205a, 205b, 205c) (to push the strong string bus (205a, 205b, 205c) working point (2507) towards the weak string bus (205a, 205b, 205c) working point (2508)).
For instance, and referring to
In an embodiment, before weak string bus (205a, 205b, 205c) outputs are adjusted, one or more weak string buses (205a, 205b, 205c) can be disconnected from the other string buses (205a, 205b, 205c) to determine if disconnecting the one or more weak string buses (205a, 205b, 205c) increases the power output from the solar array (200).
The inverter (203) can convert the direct current (DC) outputs from the string buses (205a, 205b, 205c) to an alternating current (AC) output that can be supplied, for example to a power grid or other load (e.g., a home or business). The inverter (203) can control the current and voltage drawn from the string buses (205a, 205b, 205c), and thus control where on the I-V curve the string buses (205a, 205b, 205c) operate at. For instance, as the inverter (203) can increase impedance seen by the solar array (200), which will cause the current drawn from the string buses (205a, 205b, 205c) to decrease and the voltage to increase, and the working point will shift along the I-V curve to the right towards where the I-V curve meets the x-axis (voltage). Thus, if the string buses (205a, 205b, 205c) are operating at a working point having a higher current and lower voltage than the MPP for the solar array (200), then the inverter (203) can increase the impedance causing the string bus (205a, 205b, 205c) working points to shift towards the MPP. Balancing can be carried out via the methods described with reference to
The order that balancing operations and the applying operation occur in can vary. For instance, currents produced by solar modules in each of one or more string buses can be balanced in a first balance operation (2604). Voltages produced by the one or more string buses can then be balanced in a second balance operation (2614). An MPPT algorithm can then be applied to the solar array in an apply operation (2620). In another embodiment, the method (2600) can begin with the first balance operation (2604) followed by one or more loops of the apply operation (2620). The second balance operation (2614) can then occur followed by one or more further loops of the apply operation (2620). It should be understood that the first and second balance operations (2604, 2614) and the apply operation (2620) can operate in any order or pattern, with each operation repeating one or more times before another operation operates. The first and second balance operation (2604, 2614) can pause between operations to allow the apply operation (2620) to repeat numerous times. In an embodiment, the balance operations (2604, 2614) can run simultaneously with the apply operation (2620). In an embodiment, the applying an MPPT algorithm is not a part of the method (2600). Rather, the method (2600) can merely include balancing the solar modules and the string buses.
As noted above, the method (2600) includes the balance currents produced by solar modules in each of one or more string buses operation (2604). The solar modules can be connected by a string bus of the solar array, for instance in series. In an embodiment, balancing involves (1) varying the current on the string bus, (2) monitoring changes in voltage output for each solar module on the string bus, (3) comparing the monitored changes in voltage output for the solar modules on the string bus, and (4) adjusting the current output of one or more of the solar modules such that the current outputs from all solar modules on the string bus converge (see
In an embodiment, before the first balance operation (2604), strong and weak solar modules can be identified via a first identify operation (2602). The first identify operation (2602) identifies, for each string bus, one or more strong solar modules and one or more weak solar modules. Weak solar modules can be adjusted such that their outputs converge on those of the strong solar modules.
In an embodiment, the first identify operation (2602) can be carried out by: (1) varying the current on the string bus (or impedance seen by the string bus), (2) monitoring changes in voltage output for each solar module, and (3) comparing the changes in voltage output. Strong solar modules can be those having the smallest change in voltage. Strong solar modules can also be characterized as operating at relatively the same current (indicated by similar changes in voltage). Strong solar modules can be those having working points furthest to the right on the I-V curve (e.g., working point (2201) in
Having identified strong solar modules, an average change in voltage dVs for the strong solar modules can be determined. The average change in voltage dVs for the strong solar modules can be a reference value. The change in voltage dVi for each solar module can be compared to the average change in voltage dVs for the strong solar modules. The difference between dVi for a solar module and dVs indicates by how much the solar module output current should be decreased in order to match the output current of the strong solar modules (to push the weak solar module working point (2202) towards the working point (2201) of the strong solar modules).
In an embodiment, once the first identify and balance operations (2602, 2604) have been performed, strong and weak solar modules can again be identified via the first identify operation (2602). This can be done since some strong solar modules can have become weak while the weak solar modules were being adjusted. If a weak solar module is identified that was previously a strong solar module, then its current output can be adjusted in the next loop of the first balance operation (2604).
The method (2600) can also include a second balance operation (2614). In an embodiment, the second balance operation (2614) includes (1) varying the voltage of the string buses in the solar array, (2) monitoring changes in current of each string bus, (3) comparing the monitored changes in current, and (4) adjusting the voltage output of the string buses (by changing the voltage output of all solar modules in a string bus) such that the string bus output voltages converge (see
In an embodiment, before the second balance operation (2614), strong and weak string buses can be identified via a second identify operation (2612). The second identify operation (2612) identifies strong and weak string buses in the solar array. Strong string buses can be adjusted such that their outputs converge on those of the weak string buses. In an embodiment, the second identify operation (2612) can be carried out by: (1) varying impedance seen by the string buses (or the voltage on the string buses), (2) monitoring changes in current output for each string bus, and (3) comparing the changes in current output. Weak string buses can be those having the smallest change in current. Weak string buses can also be characterized as operating at relatively the same current (indicated by similar changes in current). Weak string buses can be those having working points furthest towards the left of the I-V curve (e.g., working point (2507) in
The method (2600) also includes an apply operation (2620) wherein the maximum power point tracking algorithm is applied. Maximum power point tracking (MPPT) is a procedure or algorithm used to determine the maximum power point (MPP) of a system—in this case the maximum power point of a solar array (see, e.g., “Maximum power” in
The apply operation (2620) can operate after either of the balance operations (2604, 2614). The apply operation (2620) can also operate after any number of repetitions or loops of either or both of the balance operations (2604, 2614). In an embodiment, the apply operation (2620) can operate after any number of repetitions or loops of the first identify operation (2602) and the first balance operation (2604); after any number of repetitions or loops of the second identify operation (2612) and the second balance operation (2614); and after any number of repetitions or loops of a combination of the first and second identify and apply operations (2602, 2604, 2612, 2614). For instance, strong and weak solar modules in each of one or more string buses can be identified in the first identify operation (2602). Currents produced by solar modules in each of the one or more string buses can be balanced in the first balance operation (2604). Strong and weak string buses in each of the one or more string buses can be identified in the second identify operation (2612). Voltages produced by the one or more string buses can then be balanced in the second balance operation (2614). An MPPT algorithm can then be applied to the solar array in the apply operation (2620).
In another embodiment, the method (2600) can begin with the first identify operation (2602) and the first balance operation (2604) followed by one or more loops of the apply operation (2620). The second identify operation (2612) and the second balance operation (2614) can then occur followed by one or more further loops of the apply operation (2620). It should be understood that the first and second identify operation (2602, 2612), the first and second balance operations (2604, 2614), and the apply operation (2620) can operate in any order or pattern, with each operation repeating one or more times. In an embodiment, the apply operation (2620) can operate multiple times before the first or second identify operations (2602, 2612) resume. Alternatively, the identify and balance operations (2602, 2604, 2612, 2614) can run simultaneously with the apply operation (2620). In an embodiment, the applying an MPPT algorithm is not a necessary part of the method (2600). Rather, the method (2600) can merely include balancing the solar modules and the string buses.
In this disclosure, systems and methods for balancing string buses have been described where string bus output voltages for strong string buses were decreased until they converged on weak string bus voltages (see
Unlike solar arrays previously discussed in this disclosure (recall
In an embodiment, balancing strings (3410, 3420) using LSMUs (3413, 3423) can entail first identifying the strongest string. In an embodiment, the LSMUs (3413, 3423) can be configured to identify strings (3410, 3420) that are within the range from the output voltage of the strongest string and to identify strings that are not within the range from the output voltage of the strongest string.
In an embodiment, one LSMU (3413, 3423) can be a controlling LSMU. The controlling LSMU can do any one or more of the following: monitor the string output voltages (3414, 3424); transmit data regarding the string output voltages (3414, 3424) to other LSMUs (3413, 3423); monitor the solar array output voltage (3415); transmit data regarding the solar array output voltage (3415) to other LSMUs (3413, 3423); identify the strongest string; identify the weakest string; identify strings that are within the range from the output voltage of the strongest string; identify strings that are not within the range from the output voltage of the strongest string; and select the target voltage.
Optionally, a management unit (3440) can perform the tasks of the controlling LSMU. The management unit (3440) can be in wired or wireless communication with the LSMUs (3413, 3423). The management unit (3440) can monitor the string output voltages (3414, 3424), as well as the solar array output voltage (3415), via optional connection (3442). The management unit (3440) can identify a strongest string and a weakest string along with strings that are within the range from the output voltage of the strongest string, and strings that are not within the range from the output voltage of the strongest string. The management unit (3440) can select the target voltage. The management unit (3440) can communicate with other management units. Those other management units can be in communication with other solar arrays. The other management units can be monitor other solar arrays and thus be external to the solar array 3400.
In an embodiment, balancing strings (3410, 3420) involves converting the string output voltages (3414, 3424) to the target voltage on the parallel bus (3450). In an embodiment, the target voltage is selected to be equal to a string output voltage (3414, 3424) of the strongest string. In this case, string output voltages (3414, 3424) for strings (3410, 3420) that are not within the range from the output voltage of the strongest string can be upconverted and/or downconverted so that the solar array output voltage (3415) of the parallel bus (3450) equals the string output voltage (3414, 3424) of the strongest string. Since the string output voltage (3414, 3424) for the strongest string is equal to the target voltage, the LSMU (e.g., 3413 or 3423) connected to the strongest string (and other LSMUs connected to strings having output voltages within the range of the voltage output of the strongest string, if any) can operate in a bypass mode. In bypass mode, a converter of the LSMU (e.g., 3413, 3423) is turned off to reduce power consumption. Current can pass around or through the LSMU (3413, 3423) while losing no more energy than is lost in normal transit through low-loss or low impedance wires or circuits. In an embodiment, current passes through low-loss portions of the LSMU (3413, 3423). In an embodiment, the bypass mode involves routing current through a transistor, for instance a MOSFET or bipolar transistor.
In an embodiment, the target voltage is equal to an average string output voltage of strings (3410, 3420) that are within the range from the output voltage of the strongest string. In this case an average string output voltage (3414, 3424) can be determined for the strings (3414, 3424) that are within the range from the output voltage of the strongest string where the average includes the voltage of the strongest string. String output voltages (3414, 3424) of strings (3410, 3420) that are not within the range from the output voltage of the strongest string are upconverted so that the solar array output voltage (3415) equals the average string output voltage of the strings (3410, 3420) that are within the range from the output voltage of the strongest string. Only string bus output voltages (3414, 3424) for the strings (3410, 3420) that are not within the range from the output voltage of the strongest string are upconverted. LSMUs (3413, 3423) connected to the strings (3410, 3420) that are within the range from the output voltage of the strongest string, (including the strongest string) operate in bypass mode.
In an embodiment, the target voltage is less than the average output voltage of the strings (3410, 3420) that are within the range from the output voltage of the strongest string. Strings (3410, 3420) having string output voltages (3414, 3424) that are less than the average output voltage of the strings (3410, 3420) that are within the range from the output voltage of the strongest string are upconverted so that the solar array output voltage (3415) equals the target voltage. Strings (3410, 3420) having string output voltages (3414, 3424) that are greater than the average output voltage of the strings (3410, 3420) that are within the range from the output voltage of the strongest string are downconverted so that the solar array output voltage (3415) equals the target voltage. In this embodiment, some LSMUs (3413, 3423) are configured to both upconvert and downconvert voltages.
In an embodiment, the target voltage is greater than the string output voltage (3414, 3424) of the strongest string. In this case string output voltages (3414, 3424) for all strings (3410, 3420), including the strongest string, are upconverted so that the solar array output voltage (3415) equals the target voltage.
In one embodiment of a bypass mode of an LSMU (3413, 3423), the converter of the LSMU (3413, 3423) is not used, and a conductive path around the converter of the LSMU (3413, 3423) is used to bypass the converter of the LSMU (3413, 3423). In other words, when an LSMU (3413, 3423) is in bypass mode, current travels from a string (3410, 3420) to the inverter/combiner/load (3430) with a low voltage loss. For instance, a transistor can route current around the voltage converter of the LSMU (3413, 3423). For instance, a bypass switch (e.g., transistor) could be used to redirect current through a low-resistance wire in the LSMU (3413, 3423) rather than through the circuitry that causes losses. In an embodiment, such a switch can route current around the outside of the LSMU (3413, 3423). In an embodiment, the bypass mode can be the mode that the LSMU (3413, 3423) normally operates in unless instructed otherwise.
In an embodiment, the solar array (3400) can be connected to a combiner, inverter, or load (3430). For the case of a combiner (3430), the LSMUs (3413, 3423) can be connected in parallel via the combiner (3430). This connection can exist within or outside of the combiner (3430). In an embodiment, there can be one to four LSMUs (3413, 3423) in a combiner. However, it is also possible to have more than four LSMUs (3413, 3423) combined in a combiner (3430). The output of the combiner (3430) can be provided to an inverter (not shown). In an embodiment, outputs from multiple combiners (3430) can be combined in parallel and provided to an inverter. In one embodiment, the LSMUs (3413, 3423) and at least a portion of the parallel bus (3450) are located within the combiner box (3430).
In an embodiment, the solar array voltage output (3415) can be provided to a load (3430). Non-limiting examples of loads include computers, cell phones, single or multi-family dwellings, apartment or commercial structures, and rechargeable vehicles, to name a few.
In an embodiment, the management unit (3440) and/or the LSMUs (3413, 3423) can reside on or as part of the solar array (3400). In an embodiment, the management unit (3440) and/or the LSMUs (3413, 3423) can reside on one or more solar modules (3411, 3421). In an embodiment, the management unit (3440) and/or the LSMUs (3413, 3423) can reside outside the solar array (3400).
In an embodiment, the string output voltages of two more strong strings can be used to select the target voltage, when the string output voltages of these strong strings are within a predetermined range. The selected target voltage allows the LSMUs of these strong strings to operate in a bypass mode to reduce power consumption and/or conversion losses by the LSMUs. In one embodiment, there may be multiple groups of strong strings that have voltages within the predetermined range, and the group that can achieve a greatest reduction in power consumption via operating in a bypass mode is used to select the target voltage. The string output voltages of other strings are upconverted and/or downconverted to the target voltage. For example, two second strongest strings may have a lower string output voltage or a lower power output than the strongest string, but have a larger string output voltage or a higher power output than all other strings. When the two second strongest strings are used to select the target voltage they can be operated in the bypass mode. In one embodiment, when the two second strongest strings operate in bypass mode, the solar array output power can be greater than the solar array output power produced when only the strongest string is operated in bypass mode.
In an embodiment, the strings can be dived into a first group of strings and a second group of stings. The first group of strings have a greater string output voltage, or greater string output power, than the second group of strings. The first group of strings have string output voltages or power that are within a range of each other such that when connected to the parallel bus, their string output voltage or power is not degraded (or the degradation is less than operating them in non-bypass mode). In some embodiments, some of the strings in the second group may be stronger than the strings in the first group. The first group of strings is selected to operate in bypass mode to increase the power output of the array. For example, a first amount of power may be lost when a string is connected to the parallel bus via an LSMU in bypass mode, in which the string (and other strings connected to the parallel bus in bypass mode) might move away from the maximum power point because of the influence of the parallel bus; and a second amount of power may be lost when the string is connected to the parallel bus via the LSMU upconverting or downconverting the output voltage of the string. When the LSMU is not in the bypass mode, the LSMU can isolate the string from the parallel bus and thus allow the string to run on or near the maximum power point. However, the operation of upconverting or downconverting consumes power. When the first amount of power is less than the second amount of power, the string is selected to be part of the first group for bypass mode; and when the second amount of power is less than the first amount of power, the string is selected to be part of the second group for non-bypass mode.
As noted, the LSMU (3500) can include the input (3502). The input (3502) can be configured to receive an input voltage from the string. The input (3502) is a portion of the LSMU (3500) capable of forming a conductive path with other electrical devices. For instance, the input (3502) can be connected to the string. In an embodiment, the input (3502) can be a connector or a terminal. The input (3502) can have two poles, one positive and one negative. The negative pole can be grounded.
As noted, the LSMU (3500) can include the voltage conversion portion (3508). The voltage conversion portion (3508) can be configured to convert the input voltage to a target voltage. In other words, the voltage conversion portion (3508) can be a transformer or any voltage-converting device or circuit (e.g., Buck, Buck-Boost, and Cuk converters, to name a few). In one embodiment, the voltage conversion portion (3508) is configured to upconvert the input voltage. In another embodiment, the voltage conversion portion (3508) is configured to downconvert the input voltage. In a further embodiment, the voltage conversion portion (3508) can be configured to either upconvert or downconvert the input voltage, based on a control signal.
In one embodiment, the target voltage for the parallel bus (3450) is determined to increase the total power output from the strings (3410, 3420) to the parallel bus (3450). In the one embodiment, an LSMU is placed in the bypass mode to increase the total energy output of the solar array. For example, if the power consumption of the converter of an LSMU operated in a voltage-converting mode is larger than the drop of power output caused by operating the LSMU in a bypass mode, then the LSMU can be switched to the bypass mode to increase overall power output; otherwise, the LSMU can be switched to the voltage-converting mode
The voltage conversion portion (3508) can include multiple systems, one of which is used to upconvert the input voltage and the other is used to downconvert the input voltage. The LSMU (3500) or the voltage conversion portion (3508) can be configured to operate in a bypass mode. The voltage conversion portion (3508) can cause 1-2% energy loss when it is downconverting or upconverting the input voltage. Thus, when the string that the LSMU (3500) is connected to is identified as a string having a string output voltage within a range from the output voltage of the strongest string voltage of a string output voltage of a strongest string, and the target voltage is within a range from the output voltage of a strongest string, the LSMU (3500) can avoid the 1-2% energy loss by operating in the bypass mode.
As noted previously, the bypass mode is a low-loss mode wherein current is routed through a low-loss conductive path. For instance, the switchable connection (e.g., transistor) (3506) can be used to route current around the voltage conversion portion (3508). This allows current to pass from the string, through the LSMU (3500), and to the combiner, inverter, or load while avoiding the losses of traveling through the voltage conversion portion (3508). The voltage conversion portion (3508) can have a connection to ground.
A local controller (3504) can control the switchable connection (3506). For instance, if the switchable connection (3506) is a transistor, then the local controller (3504) can control the transistor (3506) gate. The local controller (3504) can be implemented in hardware, software, or a combination of the two. In an embodiment, a diode can be connected in parallel to the transistor (3506), or used to replace the transistor (3506) in another embodiment.
As noted, the LSMU (3500) provides the output (3518). The output (3518) can be configured to provide the target voltage to a combiner, inverter, or load. The output (3518) is a portion of the LSMU (3500) capable of forming a conductive path with other electrical devices. For instance, the output (3518) can be connected to an inverter or load. Non-limiting examples of the output (3518) include a connector or terminal, to name two. The output (3518) can be connected to outputs from other LSMUs. In an embodiment, outputs of different LSMUs can be combined within a combiner. Connections between LSMU outputs can be made in parallel. In an embodiment, the output (3518) can provide a fixed voltage inverter voltage to an inverter (or to a combiner and then to an inverter). The output (3518) can have two poles, one positive and one negative. The negative pole can be grounded.
Unlike LSMU (3500), LSMU (3600) can include an input filter capacitor (3614) and can include an output filter capacitor (3616). The LSMU (3600) also can include a diode (3610) or rectifier. The LSMU (3600) can include a local controller (3604), which can be implemented in hardware, software, or a combination of the two. The local controller (3604) can be in communication with the management unit (3440) discussed with reference to
In one embodiment, the LSMU (3600) includes a booster converter which includes an inductor (3608) and a switchable connection (3612). The output voltage of the boost converter is determined by the duty cycle of the switch (3612). In the bypass mode, when the switchable connection (3606) is closed, losses due to current passing through the inductor (3612) and the diode (3610) are reduced (and/or are negligible). Thus the ratio of input voltage to target voltage can be controlled via the local controller (3604). The local controller (3604) can also control the bypass mode by controlling the switchable connection (3606). The LSMU (3600) can include a connection (3620) able to connect to other LSMUs. For instance, the connection (3620) can allow LSMUs to be connected within a combiner box.
The select operation (3702) can include selecting the output voltage of a string having a string output voltage within a range from the output voltage of a strongest string as a target voltage. A string having a string output voltage within a range from the output voltage of a strongest string can be referred to as a “strong string.” Alternatively, the select operation (3702) can include selecting the average string output voltage of strong strings as the target voltage. The select operation (3702) can be carried out by LSMUs or the management unit. In an embodiment, strong strings and weak strings (having string output voltage not within the range from the output voltage of the strongest string) can be identified. Having identified one or more strong strings, the string output voltages from weak strings can be upconverted in order to cause the solar array output voltage (3415) of the parallel bus (3450) to equal (or match or converges on) the string output voltage (3414, 3424) of the strongest string. Identification of strong strings is not a required operation since such identification can have occurred before the method (3700) begins.
The operate operation (3704) can include operating a first LSMU in a bypass mode. The first LSMU converts the string output voltage for a strongest string or a strong string (a string having a string output voltage that is within a range from the output voltage of the strongest string). In an embodiment, the first LSMU is an LSMU that can be bypassed to increase the total power output of the strings. Bypass mode can be used when the solar array target voltage equals or matches the string output voltage of the strongest string or when the solar array target voltage is within the range from the output voltage of a strongest string. The bypass mode can be used to allow current to pass through or around a first LSMU without a voltage drop due to current passing through high resistance or high impedance circuitry in the first LSMU. When an LSMU operates in bypass mode, the string operates at the voltage of the parallel bus, which may be different from the maximum power point voltage of the string. In other words, there can be a small power loss.
The method (3700) can include an instruct operation (3706) where a second LSMU is instructed to upconvert a second string output voltage so that the solar array output voltage equals the target voltage. The second LSMU can be connected to a second string. The second string can be a weak string or a string having a string output voltage that is not within a range from the output voltage of a strongest string. The instruct operation (3706) can include providing instructions (signals or commands) to the second LSMU to upconvert the second string output voltage. In an embodiment, a management unit provides the instructions to the second LSMU. In an embodiment, a controlling LSMU provides the instructions to the second LSMU. In an embodiment, the controlling LSMU is the second LSMU, so the second LSMU provides the instructions to itself.
After the instruct operation (3706), the method (3700) can repeat or loop. Alternatively, the method (3700) can wait for one or more loops of an MPPT algorithm to run. Once the MPPT algorithm has run one or more times, solar modules on the string can be balanced, and the method (3700) can begin again. The inverter carries out the MPPT algorithm. In an embodiment, the MPPT algorithm runs at the same time that the method (3700) operates. In other words, the inverter tries to find the solar array's MPP while the solar array is balancing currents produces by solar modules on each string and balancing voltages between the parallel-connected strings. The MPPT algorithm does not have to stop for balancing of solar modules and strings to take place.
In one embodiment, an inverter is allowed to be powered by two (or more) separately controlled strings, connected to the parallel bus (3450) via respective LSMUs, and to operate with its internal MPPT algorithm (that is typically designed for a single string) to deliver optimal or near optimal overall system performance. This arrangement does not require that the inverter's internal MPPT be turned off or the need to move the inverter to a fixed input voltage, and also not require a separate MPPT per string. In one embodiment, the inverter is to operate at its optimal spot continuously, as power generation conditions change constantly; and hence the system performs continuous rebalancing to achieve optimal operation.
In one embodiment, the inverter's MPPT algorithms pull and tug current to create current changes delta current (dI). The algorithm to control the LSMUs has several operations to calculate the dI for every string and to cause the corresponding current change by adjusting PWM value on LSMUs connected to weaker string(s). The inverter takes further operations to get to the maximum power point.
For example, in one embodiment, the delta current of each string (dIk) is calculated when the string is connected via an LSMU. Weak strings are identified. It may be checked whether disconnecting those strings would produce more power. The average delta current is then calculated for the weak strings (dIW). For each strong string, adjust or keep PWM values of its LSMU to be 100% or near 100% (minimal or no change the string). Other strings are adjusted to match voltage set by strong string by changing their voltage output driving PWM to a different value than 100%. Wait some time to allow the inverter to run several MTTP cycles, before repeating the operations to balancing the strings.
In one embodiment, the decision of whether to up-convert in some cases the strong string may depend on several factors, including weak solar irradiation, temperature, and the parameters of the string vs. the inverter, amongst other consideration, but the decision is not limited to these factors.
In an embodiment, an MPP for the strong string can be determined. The strong string can then operate at its MPP. In an embodiment an inverter can run an MPPT algorithm to determine the MPP for the strong string and the strong string can then operate at its MPP.
In an embodiment, the method (3700) can include identifying one or more strong strings and one or more weak strings. Strong strings and weak strings can be identified as previously described (see discussion of
To further improve the energy production performance of photovoltaic systems and/or reduce the cost of the systems, an enhanced system and method is provided for measuring relevant data of any string to help managers decide whether to add or remove an LSMU from a string.
The modules (3804a-n) implemented in the form of blades can be simple bypass blades which can simply facilitate an electric connection between the strings (3410a-n) and the load bus (3805), bypass blades with measurement and disconnect features, or bypass blades with measurement and disconnect features as well as an LSMU, such as the blade illustrated in
In an embodiment, modules (3804a-n) implemented in the form of blades have five connections/connectors through which the blades are connected to other system components in the cage (3802), which provides a housing to the components disposed therein. The connections include two connections (3803a-n) (one pair each) through which each module (3804a-n) implemented in the form of a blade is connected to the corresponding string (3410a-n); two connections through which each module (3804a-n) implemented in the form of a blade is connected to the load bus (3805); and one connection through which each module (3804a-n) implemented in the form of a blade is connected to local controller (3810) through a controller bus (3808). The load bus (3805) is connected to the inverter or other energy consuming unit (3806).
As illustrated in
In one embodiment, a switch device (4201) enables a user to push a button, for example, to trigger a system switch such as switch (3912) or (3922) illustrated in
In one embodiment, two short pins (4205a-n) are grouped with the high-powered connectors (4204a-n). These short pins (4205a-n) can be, for example, half-length, so as to make contact when module (3804x) implemented in the form of a blade is fully inserted. If module (3804x) in the form of a blade is removed without prior deactivation of the circuit, disconnection of short pins (4205a-n) would occur prior to the disengagement of the high-powered connectors (4204a-n) and thus immediately cause the controller (e.g., 3810, 3911, or 3921) to break the circuit before the main power is disrupted, thus giving the controller (e.g., 3810, 3911, or 3921) time to turn off switches (3922) and (3912).
In one embodiment, the strings (e.g., 4504 and 4514) can be received in and managed via the LSMUs (e.g., 4505 or 4515), which can be any of the LSMUs disclosed herein. For example, a modular unit string management system as illustrated in
In one embodiment, the system management unit (4520) can be utilized to determine one or more operating parameters (e.g., duty cycles, phases, synchronize pulses, etc.) based on measurements of the operating voltages and currents of the solar panels as reported by the individual CLMUs (e.g., 4502a-n). A CLMU (4502a) is configured to determine its own duty cycles for operating the respective solar panels (e.g., 4501a-c) with or without relying upon communicating with other management units. The duty cycles are configured to separately operate the solar modules in the group connected to the CLMU at the maximum power point.
In one embodiment, the CLMU functions as an LMU discussed above for each individual solar panel (e.g., 4603, 4064 or 4605) to separately operate the solar panels/solar modules. However, combining the LMU functions in a CLMU allows the sharing of common components and reducing of the number and cost of management units present in the system.
In the exemplary system (4600) illustrated in
In one embodiment, the CLMU is secured in a location separate from the individual solar panels connected thereto (as illustrated in
In
In one embodiment, the junction box (4614) is secured to the panel physically located in the middle of the panel group layout. Because different configurations are possible, for example, with two, four, five, or any other suitable number of panels, the illustrated positioning may be different in other embodiments of the disclosed system.
In
In
The converters (e.g., 4705, 4706, and 4707) may be any of the types described herein throughout. For example, the converter (e.g., 4705, 4706, or 4707) can be an LMU (101) illustrated in
Also, the methods disclosed above to individually control or operate the solar modules of separate panels can be used accordingly with a CLMU hosted inside the junction box (4614).
Communications unit (4702) may use a wireless connection (4703), or a wired connection (4704) to the string (4601), to communicate with a system management unit (e.g., (204) illustrated in
In one embodiment, galvanic separation may be used on the wires between the controller and the converters for creeping current etc.
While the input voltages of solar modules (4603-4605) connected to the converters (4705-4707) of the CLMU may be different, the output currents of the converters (4705-4707) are the same, due to the series connection of the outputs of the converters (4705-4707). Thus, a single current measurement unit/current sensor can be used to measure the output current of the solar panels connected to a CLMU.
In one embodiment, the CLMU is configured to use a voltage measurement unit for multiple solar panels connected to the CLMU via a set of switches. The CLMU controls the states of the switches to dynamically connect the voltage measurement unit to one of the converters (4705-4707) at a time to measure the input or output voltage measurement of the respective solar panel.
In one embodiment, the system may have a combiner box to combine multiple strings of panels that are connected via CLMUs, in a way as illustrated in
In
It is clear that many modifications and variations of this embodiment can be made by one skilled in the art without departing from the spirit of the novel art of this disclosure. For example, the systems and method herein disclosed can be applied to energy generating systems besides photovoltaic systems (e.g., windmills, water turbines, hydrogen fuel cells, to name a few). Also, although the target voltage has been compared to the average string output voltage of strong strings, other reference points for the target voltage can also be used. Some non-limiting examples include an string output voltage for the strongest string or a median string output voltage of strong strings.
In the foregoing specification, the disclosure has been described with reference to specific exemplary embodiments thereof. It will be evident that various modifications can be made thereto without departing from the broader spirit and scope as set forth in the following claims. The specification and drawings are, accordingly, to be regarded in an illustrative sense rather than a restrictive sense.
The present application is a continuation application of U.S. patent application Ser. No. 13/346,482, filed Jan. 9, 2012 and entitled “Systems and Methods to Reduce the Number and Cost of Management Units of Distributed Power Generators”, which claims the benefit of the filing date of Prov. U.S. Pat. App. Ser. No. 61/512,785, filed Jul. 28, 2011, and titled, “System and Method for Reducing the Number and Cost of Local Management Units in a Distributed Power Generation Site,” the entire disclosures of which applications are incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| 61512785 | Jul 2011 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 13346482 | Jan 2012 | US |
| Child | 15225692 | US |
