Patent Grant
11469596
The present application is a national phase entry under 35 U.S.C. § 371 of International Application No. PCT/EP2019/067779, filed Jul. 2, 2019, published in French, which claim priority from European Patent Application No. 18305899.9 filed Jul. 6, 2018, all of which are incorporated herein by reference.
The invention relates to the general field of power electronics and its applications on electrical power distribution networks or on-board electrical power networks, smart buildings and micro-grids, and more specifically modules for filtering and compensating electrical networks with injection of power from renewable sources.
It is well known that the quality of the electrical signal delivered by the distribution network has a direct consequence on the performance of the systems powered by this network, as well as the lifetime of the items of electrical equipment that constitute the network or are connected to it.
In particular, harmonic distortion, a phase difference between the voltage and the amperage, i.e. the reactive power, and also current and voltage unbalance are factors used to express the quality of the energy travelling through this network.
The term “unbalance” is understood to mean a difference in the physical quantities of the signal between the different phases and/or amplitudes, for example the voltage or amperage levels.
The harmonic distortion rate is a measurement of linearity of signal processing, made by comparing the output signal of an apparatus with a sinusoidal input signal.
The non-linearity of the system distorts this sine wave. The output signal remains a periodic signal that can be analyzed as a sum of sine waves of frequencies that are multiples of the frequency giving the period, known as the fundamental frequency.
Each of these sine waves is a harmonic of order equal to the quotient of its frequency by the fundamental frequency. The harmonic distortion rate is the ratio of the root mean square values of the fundamental frequency to the others.
An apparatus including non-linear loads connected to an electrical network receives a power from the electrical network and reinjects a signal into the network, the signal reinjected into the network being degraded by the operation of the apparatus.
The term “non-linear load” is understood to mean a load based on electronic power components consuming active power over the power supply system (with or without reactive power) thus and most importantly reinjecting the distorting power (related to the harmonics) into the supply network. These harmonics can be of conventional order (5, 7, 11, 13, etc.) for 3-wire power supply systems: three-phase non-linear loads (widespread in industrial areas). In the case of single-phase non-linear loads installed in a 4-wire power supply system (3 phases and neutral) widely used in residential, commercial or administrative areas, third order harmonics and their odd multiples (3, 9, 15, etc.) will circulate in this network in addition to the conventional harmonics.
Moreover, the phase difference between the voltage and the amperage of the signal entails the appearance of a reactive power that drives, among other things, the reduction of the active power transmissible through the network.
The development of technology for producing energy, particularly from renewable sources, has led to the appearance of many individual production units, particularly in so-called positive-energy housing solutions.
When a power production unit, particularly in the case of renewable energy of solar or windpower type, is connected to the network, it is necessary to use one or more inverters or rectifiers-and-inverter pairs in order to manage and convert the power generated by the unit into DC current and DC voltage into an AC signal, before injecting it into the network.
The inverters, usually composed of electronic power switches with controlled opening and closing and double throw, such as an IGBT (Insulated Gate Bipolar Transistor) and GTO (Gate Turn-off Thyristor). By a set of switching operations controlled in an appropriate manner (generally a modulation of pulse width), the source is modulated to obtain an AC signal of the desired frequency.
The switching frequencies of the electronic power components of the inverters generally use high-frequency harmonic components in the signal injected into the network, which degrades the quality of the signal traveling through the network.
In an assembly as shown in
Thus, the high-frequency harmonics of the signal caused by the switching of the inverter 4, are not transmitted via the filter 5 to the network 1; only the desired signal is generated by the inverter 4 and injected into the network 1 or into the load 2.
Various structures of filter 5 have been proposed for this purpose.
A first-order filter, the most conventionally used; composed of a simple inductance with a practically negligible internal resistance, as shown in
This is because, the higher the inductance of such a filter, the higher the ability of the filter to prevent the components due to switching from being injected over the network 1. However, the higher the inductance, the more the speed of variation of the amperage traveling through the filter is reduced and the more the filter will cause a phase difference between the desired actual amperage and the amperage to be injected into the network 1-load 2 assembly via the filter 5.
Conversely, a low value of the inductance allows the majority of the components due to switching to be injected into the network 1 and consequently to affect the electrical installations and equipment.
The proper dimensioning of the first-order output filter will therefore depend on the trade-off to be found between the dynamics and the efficiency of the device based on the inverter, especially when it operates as a shunt active filter with harmonic decontamination.
This trade-off is very difficult to strike without using an auxiliary passive filter 6 installed at the output of the output filter 5 or upstream on the network side, to filter the high-frequency components as shown in
However, this auxiliary filter 6 can cause undesirable second effects such as resonance with other passive elements installed on the electrical network 1.
These electrical resonance phenomena in some cases lead to voltage or amperage peaks greatly above the values permissible by the apparatuses connected to the network 1 and cause the destruction of these devices.
This auxiliary filter also gives rise to an active power consumption by its damping resistance. In addition, the filtering quality of these auxiliary filters degrades over time because of the aging of their passive components.
One aim of the invention is to “decontaminate” the current consumed by a non-linear load, by cancelling, on the network side, its harmonic content, the unbalanced content and also the reactive content of the current in the aim of improving the quality of the voltage on the electrical network side. This invention, having a 4-wire structure, is adapted to the harmonic spectrum containing the conventional orders (5, 7, 11, 13, etc.) of industrial areas as well as those of third order and their odd multiples (3, 9, 15, etc.) of residential, commercial or administrative areas.
Another aim is to maximize the active power producible by a unit for generating renewable energy as in
Another aim is to optimize the energy consumption of a smart building.
Another aim is to optimize the energy production of a unit for producing conventional energy (from fossil fuel: petroleum, gas etc.) within a micro-grid.
Another aim is to optimize the management of the energy traveling between the production units (renewable and conventional) and the consumption units.
The invention can be optionally but advantageously completed by the following features, taken alone or in combination:
Optionally but advantageously, the invention can be completed by the following features, taken alone or in combination:
More precisely, the modeling steps are as follows:
1—installing a measuring device that is a spectrum analyzer/energy quality analyzer, at the intake of each production line of the site being studied, and which supplies all the electrical quantities of the network, namely: the active, reactive and apparent powers, the power factor, the fundamental components of the current and of the voltage, the total harmonic distortion rate (THD) of the current and of the voltage, the individual harmonic distortion rate of the current and of the voltages, etc.,
2—modelling the active and reactive powers by a shunt R-L branch or by a single current source that represents the fundamental current, with an angle corresponding to the power factor; the direction of passing of the current goes from the electrical network 1 toward the load 2 (represented here by the current sources),
3—modeling the individual harmonic distortion rate of the currents and of the voltages by current sources; each corresponding to a harmonic order, the direction of passing of the harmonic currents goes from the load (2) (represented here by the current sources) toward the electrical network (1),
4—connecting all these current sources to one or more resistances with very high values (in the order of the MΩ), to force the passing of the currents in a single direction.
Other features and advantages of the invention will become further apparent from the following description, which is purely illustrative and non-limiting, and must be read with reference to the appended figures wherein:
The invention relates to an electrical system as shown in
The compensating assembly 7 includes:
The control unit 12 is configured to control the inverter 9 in such a way that the inverter 9 generates a signal of amperage Iinj configured in phase opposition to cancel, at the connection point C, the non-active disturbances of the signal generated by the load 2 and injected into the network.
The signal emitted at the output of the filtering unit 10, and therefore at the output of the compensating assembly 7, therefore has the harmonics of the signal caused by the non-linear part of the load 2.
For any harmonic present in the disturbance signal, caused by the load 2, the signal emitted by the filtering unit 10 has a voltage value (in phase opposition) that is the inverse of the voltage value of the signal of the non-linear load 2 at this harmonic n. Thus, by injecting the signal at the output of the compensating assembly 7 at the connection point C, the harmonics of the signal emitted by the non-linear load 2 are cancelled and the current traveling on the network 1 side is devoid of harmonics.
Similarly, the output signal of the compensating assembly 7 has, for the fundamental frequency and the harmonics, a phase difference configured to generate a reactive power (in phase opposition) that is the inverse of the reactive power of the disturbance signal of the load 2.
In this way, the reactive power consumed by the load 2 is completely compensated, within the limit of the dimensioning of the device 7, on the network side.
The control unit 12 is further configured to compensate for the unbalance of the signal traveling through the non-linear or linear load 2.
The term “unbalance” is understood to mean a different current consumption as a function of the different phases of several single-phase charges 2.
It should be noted that any disturbance current with a frequency included in the frequency band of the loop for controlling the injected current, ranging from 0 Hz to 2500 Hz, can be compensated/filtered by the device 7 (cf.
The compensating assembly 7 is configured to be installed in an electrical network already contaminated at the level of its upstream voltage while offering the possibility of operating at two fundamental frequencies (50 and 60 Hz), which can each vary by ±10%, while being capable of adapting to voltage variations of ±15%.
This allows the correct operation of the compensating assembly 7 even in the event of electrical generator units replacing the electrical network.
The control unit 12 is further configured to regulate (charge and keep constant) the voltage Vdc across the terminals of the capacitive storing element 3.
The control unit 12 is further configured to detect the MPPT (Maximum Power Point Tracking) of a renewable-energy-generating power unit 100 and generating, via the inverter 9, the active current corresponding to this maximum power; the voltage Vdc across the terminals of the capacitive storing element 3 is equal, in this case, to the voltage of the maximum power of the renewable energy production unit 100.
The Converting Unit 8
The converting unit 8 includes, as represented in
The inverter 9 is here driven by the control unit 12 in such a way as to occupy, with the units 3 and 11, a shunt active filter function.
The inverter 9, with 4 wires and 4 arms, includes for each of the 4 arms, two switching devices 13 and 14 each connected on the one hand to an output terminal of the capacitive storing element 3 and on the other hand to a common connection point A to which one of the 4 wires is also connected.
This connection point located, for each arm, between the first switching device 13 and the second switching device 14 forms a point of connection of one of the 4 wires of the inverter 9 connected to the other end of the filtering unit 10, the output of which is connected to the connection point C.
Each switching device is composed of an electronic power switch D, which can be controlled to open and close (IGBT, GTO or another) and an anti-parallel diode B (double-throw operation).
The inverter 9 emits, over each wire, a signal having an amperage Iinj.
The inverter 9 is configured to inject powers covering a wide frequency band, usually between the frequency of the fundamental, for example 50 Hz, and the harmonic order 50, having in the example of a 50 Hz fundamental a frequency of 2500 Hz.
The Filtering Unit 10
The filtering unit 10 in this embodiment comprises a filter 11 of LCL type, which has the advantage of being easy to implement with respect to a bulky, heavy and expensive first-order filter.
The filter 11 therefore includes, on each of the 4 wires (three phases with the neutral), a first inductance 15 connected on the one hand to an input wire of the filtering unit 10 (exiting from one of the 4 common connection points A), and on the other hand to a second inductance 16, the second inductance 16 being connected on either side to one of the 4 connection points C via an output wire of the filtering unit 10.
The filter 11 associated with each wire further includes a capacitor 17 connected on the one hand to a common point E and on the other hand to a connection point F between the first inductance 15 and the second inductance 16.
The filter 11 makes it possible to prevent the propagation of the components due to the switching frequency of the inverter 9 to the power supply network 1 without degrading the dynamic response of the compensating system 7. It therefore makes it possible to limit the risk of resonance, and thus dispense with an auxiliary passive filter intended to limit this propagation, if a first-order output filter (a simple inductance) is used in association with this auxiliary filter.
The Control Unit 12
The control unit 12 is configured to regulate the voltage across the terminals of the storing element 3 as well as to identify and control the injected current, via the filtering unit 10, at the connection points C for the 4 wires and for controlling the inverter 9 to generate this current. The identification of the currents to be injected at the connection points C is done by the unit 25 for computing the reference currents.
The current to be injected at the connection points C includes the active current of the renewable-energy-generating power unit 100 as well as non-active disturbance currents, which can have harmonics, unbalance, and reactive power, etc. configured in phase opposition to oppose the harmonics, unbalance and reactive power of the signal traveling through the load 2 in such a way as to reduce them, or even cancel them on the electrical network 1 side.
The controller 23 provides the tracking of the current between the injected current Iinj and the reference current coming from the unit 25. On its side, the controller 62 provides the tracking of the voltage between the voltage Vdc across the terminals of the storing element 3 and the voltage of the maximum power of the unit 100, delivered by the unit 25-B; the unit 3 shall consequently be charged.
Finally, the driving unit 21 provides the control of the switches of the inverter 9 in order to generate Iinj.
The operation and the structure of the control unit 12 shall be subsequently described for one phase. It is understood that each of the 4 wires (the three phases and the neutral) is driven similarly and independently by the control unit 12, which therefore includes a chain of control by wire.
For the control of the switching devices 13, 14 of a wire, the control unit 12 therefore includes two outputs, one connected to the pole for triggering the first switching device 13, the other being connected to the pole for triggering the second switching unit 14.
The two outputs are jointly connected to the output of a logic comparator 18, in such a way as to simultaneously control the switching of the two switching devices 13, 14.
One of the outputs of the control unit 12 includes a logic level inverter 19, in such a way that the activation of one of the switching devices 13, 14 causes the deactivation of the other, advantageously during a single control pulse owing to their common connection, upstream of the logic level inverter 19, to one and the same comparator 18.
The comparator 18 emits a logic output signal corresponding to the comparison between an output signal of a saturation element 20 and an output signal of a switch driving device 21.
The switch driving device 21 controls the switching of the inverter 9 via a non-linear controller 23 by continuous sliding mode, adapted to the filter 11 of LCL type that provides in a closed loop the control of the entire 50 to 2500 Hz frequency band for the injection of the non-active disturbance current and of the active current, as a function of the identification of the reference currents by the computing unit 25.
The switching control of the inverter 9 is done in such a way as to allow to pass through the output filtering unit 10 at the connection point C:
The switch driving device 21 emits a signal (called a carrier wave) at a predetermined frequency, as a function of the dimensioning of the units 9, 10 and 3, in such a way as to cause the switching of the switching devices 13, 14.
In most cases, the control laws, designed to control the voltage inverters which are connected to the electrical network via an LCL filter, have been originally established for renewable energy systems (photovoltaic and windpower.) In this case, provision can be made for linear controls as the inverter has only fundamental components (mainly an active power) to be injected into the network, the phase difference, at the fundamental frequency, between the signal Iref-fondamental identified via the unit 25-B for computing the reference currents and that injected Iin-fondamental being negligible in this case. It should be noted that in this operating mode, the unit 25-A, because of a phase difference problem, is not included in the identifying block and the unit 25 is thus limited to the unit 25-B.
Specifically, in the case of a shunt active filter, the inverter 9 must inject into the electrical network fundamental components (reactive and unbalanced) and harmonic components at the same time, which implies having total control over a very large frequency passband. A considerable phase difference limits the applicability of the linear controllers in the case of harmonic filtering.
The non-linear control of the injected current Iinj makes it possible to overcome this phase difference problem, encountered when using an LCL filter associated with a linear controller.
In this embodiment, the controller 23 carries out SMC (Sliding Mode Control), which makes it possible to ensure the desired dynamic response, a strong robustness/insensitivity to the bounded disturbances and also good control properties within a wide range of operating conditions.
It is well known that conventional sliding mode control with a so-called Sign function, generates a control of the inverter switches at very high frequency (discontinuous control), in order to ensure finite-time convergence on the sliding surface wherein the states of the system are subsequently maintained, even in the presence of bounded disturbances.
The Sign function of a signal is given by the relationship:
In practice, the voltage inverter is controlled by a switching function with a limited, fixed or variable frequency. Very high frequency switching cause overheating that can result in the destruction of the inverter.
In order to avoid operation in discontinuous control mode, this invention makes use of two controllers per continuous-control sliding mode, one being associated with a Sign function approximated to a sigmoid function and the other being associated with a method of artificial increase of relative degree, followed by an integrator. The two controllers are associated with two appropriate sliding surfaces.
The first method consists in replacing the Sign function with a continuous approximation. This is a Sign function approximated to a Sigmoid function given by the formula:
with ε the thickness of the neighborhood of the sliding surface.
The system, in this case, no longer converges on the desired value, but on the neighborhood of this value, which makes it possible to limit very high-frequency switching.
The artificial increase of relative degree method consists in increasing the order of the sliding surface by one degree (leading to an additional derivative), which makes it possible to insert an integrator at the output of the Sign function. The control in this case becomes continuous. Specifically, by once more deriving the tracking error between the signal Iref identified by the unit 25 and the injected current Iinj, the signal at the output of the Sign function represents the derivative of the control (discontinuous until now). An integral action at the output of the Sign function will generate a continuous control.
The two controllers used make it possible to avoid the discontinuity of the control signal, particularly caused by variable switching of the inverter at very high frequencies, caused by conventional sliding mode control.
In addition, three other non-linear controls based on the higher-order sliding mode method are also used. These are the 2-SMC Twisting and 2-SMC super-Twisting second-order sliding mode control algorithms as well as the C-HOSMC continuous higher-order sliding mode control algorithm.
These controllers ensure, within a wide range of operating conditions, as described below, the robustness of performance and stability provided by the conventional discontinuous control sliding mode control method, while offering continuous control that ensures the nominal operation of the inverter (9).
Optionally but advantageously, the switch driving device 21 includes a device for modulating pulse width, wherein the control signal is compared to a carrier wave with a predetermined switching frequency as a function of the dimensioning of the units 9, 10 and 3.
This makes it possible to make the inverter 9 operate at a fixed switching frequency.
A synergistic effect is obtained by hence improving the retention of the high-frequency components, due to the switching, by the filter 11 of LCL type.
A saturation element 20 is conventionally configured to impose upper and lower limits on the control signal. These limits are determined by the amplitude of the carrier wave which in turn linked to the voltage of the capacitive storing element 3.
The input signal of the saturation element 20 comes from the output of a summer 22, the output signal of which is the sum of a control voltage u emitted by a controller 23 and of the voltage VS of the connection point C. The addition of the voltage Vs into the control loop of the injected current Iinj prevents a strong draw of reactive current from the inductors of the LCL filter 11.
The unit 25 for computing the reference currents includes a plurality of measured inputs.
The quantities of the system that are measured include:
The unit 25 for computing the reference currents includes one output for each phase, each output corresponding to the control chain of the associated phase (including the neutral), said control chain comprising a comparator 24, a controller 23, a summer 22, a saturation element 20 and a logic comparator 18 the output of which divides into two branches, one of which includes a logic level inverter 19.
The unit 25 for computing the reference currents therefore emits one setpoint signal per phase, including the neutral, having an amperage Iref.
The amperage Iinj of each phase, including the neutral, of the signal emitted by the inverter 9 is returned to the control unit 12 and compared, via the comparator 24 with the setpoint Iref of the control chain of the corresponding phase, delivered by the unit 25. The difference between Iref and Iinj is corrected via the controller 23 which emits the control signal u.
The voltage Vdc applied across the terminals of the capacitive storing unit 3 is regulated (kept constant) by comparing it, via a comparator 60, with a reference voltage Vdc-ref, which is equal to the voltage of the maximum power point VMPP of the renewable-energy-generating power unit 100, computed via the unit 25-B.
The measured signal of the voltage Vdc has the 300 Hz or other fluctuations filtered out of it, via a second-order low-pass filter 61. The error signal (Vdc-ref-Vdc) is controlled by a PI (Proportional Integral) controller 62 or another, to obtain the maximum power PMPP.
The voltage VS of the network 1 is added in the summer 22 to the control signal u, the output of the summer 22 being limited in the saturation element 20, the output of the saturation element 20 being compared via the logic comparator 18 with the signal delivered by the pulse width modulating device 21 (the carrier).
The output signal of the comparator 18 is at a level 1 if the output signal of the saturation element 20 is greater than the carrier. Otherwise it is at a level 0. Following this logic chain, the switching device 13 or 14, which does not include any logic level inverter 19, is respectively closed or open (the other device operating in a complementary way).
The control unit 12 includes at least one processor and at least one memory, the memory comprising a program executed by the processor in such a way as to implement the method for determining the setpoint signal Iref, containing the non-active current as well as the current IMPP of the maximum power point MPPT, via the unit 25 for computing the reference currents, the controlling of the injected current Iinj via the control unit 23, to regulate the voltage Vdc across the terminals of the storing unit 3 via the controller 62 and generating this injected current by controlling the devices 13, 14 of each wire of the inverter 9 via the driving unit 21.
The method for determining the setpoint signal Iref, delivered by the unit 25 for computing the reference currents, includes the following steps:
Optionally but advantageously, the unit 25 for computing the reference currents is configured to identify, via the unit 25-B, the maximum power point MPPT of operation of the renewable-energy-generating power unit 100, installed within a renewable energy production field, a building or a self-consumption factory 99, in the following way:
Advantageously, only the algorithm for computing the maximum power point (unit 25-B) is to be integrated into the unit 25, via the algorithm for identifying the disturbance currents (unit 25-A). Specifically, the loop for regulating the voltage of the capacitor (unit (3)) of the inverter is already provided for in the unit 25-A to charge the capacitor of the inverter during operation as a pure active filter, in order to compensate for the losses caused by the components of the inverter as well as the LCL filter. In this invention and during operation as an active filter with injection of the maximum power of a renewable-energy source, the reference voltage Vdc-ref becomes the voltage of the maximum power, delivered by the unit 25-B, instead of being predefined according to a pure-filtering-oriented specifications book. This methodology simplifies, minimizes and improves accuracy, in a noteworthy manner, during the implementation of the control part, by comparison with the prior art which makes provision for an additional current loop that also requires the use of a chopper.
The control unit 12 is further configured to carry out the regulation of the voltage Vdc across the terminals of the capacitive storing element 3, and consequently to provide the recharging of the capacitor 3. Specifically, the capacitor 3 has, inter alia, the role of covering the losses of the units 9 and 10 as well as supplying the maximum active current IMPP of the renewable-energy-generating power unit 100 at the connection points C, via the inverter 9 and the filtering unit 10.
The rest of the process within the unit 12:
Next, since the unit 12 has the setpoint current Iref, coming from the unit 25, as well as the injected current Iinj controlled in turn via the controller 23, the inverter 9 is controlled via the driving unit 21 to generate the current Iinj.
This makes it possible to avoid the use of an additional device, a chopper, to generate the maximum power coming from the renewable-energy-generating power unit 100.
The inverter imposes in this case the voltage of the DC side. This entails greater reliability of the shunt active Filter device as well as considerable financial savings.
Optionally, the control unit 12 can include a chopper configured to maintain a predetermined DC voltage across the terminals of the energy-storing element (3) of the inverter. This voltage can correspond to the voltage of a battery bank to be charged. For example, the control unit 12 can include:
In an embodiment, the compensating device 7 is installed within a so-called smart building 27, i.e. the electrical apparatuses that the smart building 27 contains can, inter alia, be controlled and activated selectively by a decentralized management unit 70, for example to operate during so-called off-peak periods of the day, during which the energy demand of the network is low and the cost of the energy, from the consumer point of view, decreases.
The control unit 12 is connected to the decentralized management unit 70 of the smart building 27. The decentralized management unit 70 communicates in real time to the control unit 12 the powers of the loads (electrical apparatuses) turned on or off in the building 27.
The control unit 12, having in real time the maximum power of the renewable-energy-generating power unit 100 delivered by the unit 25-B and the controller 62 as well as the powers of the loads, turned on or off, of the building 27 communicated by the decentralized management unit 70, is configured to regulate the consumption of the different apparatuses according to at least two modes of economical distribution of the loads.
In a first mode of economical load distribution, the smart building 27 includes a renewable-energy-generating power unit 100 and in the event of a high degree of availability of the renewable primary sources, the control unit 12 is configured to drive the decentralized management unit 70 in such a way as to activate the apparatuses of the smart building 27 as a function of the production of the unit 100; this consumption mode is known as the adaptive mode.
In a second mode wherein the renewable primary energy is not very available, the control unit 12 is configured to drive the decentralized management unit 70 in such a way as to selectively activate the apparatuses of the smart building 27, in such a way as to present a flat load curve or another, according to the producible power of the network; this consumption mode is called the modulated mode.
This makes it possible to avoid demand peaks on the network 1 and to benefit from a reduced cost of the energy consumed.
The majority, or even the entirety, of the consumed energy is, in this case, of renewable origin, which makes it possible to minimize the consumption of power coming from the network 1. In the event of there being a great need for power and as a supplement to the second distribution mode and if the electrical network 1 is smart, a third distribution mode adapted to the production of the smart electrical network 1 can be envisioned.
In an embodiment, the compensating device 7 is installed between the network 1 and an industrial site or a residential, administrative or commercial building, comparable from the point of view of the network 1 with a disturbing non-linear/linear load 2.
Before installing such a device on an existing industrial site, it is sometimes necessary to demonstrate the qualities of the device to the owner of the site, who will decide whether or not to equip his site with the device.
It is therefore necessary to model the site in order to simulate its performance with and without the device, in order to appraise the contribution of the compensating device.
Conventionally, the modelling of industrial sites is done by producing a complete model containing all the machines as well as their controllers (based on electronic power converters) present across the area being studied.
However, this type of modelling faces two problems: those related to hardware and to the computer systems, which limit its adoption as a credible model for validating solutions applied to this type of electrical network.
Specifically, it is necessary in this case to have as many measurement apparatuses as controlled machines. In addition, it proves difficult to carry out a simulation of an industrial network with several voltage levels, containing several power transformers, many power lines and cables, and a plurality of machines controlled with several decontamination filters.
In this embodiment, a single measuring device is installed at the intake of each production line.
The modelling of the site is done by replacing the machines controlled by their currents (active, reactive and harmonic) measured by one of the three-phase measuring devices.
The simulation module is thereby greatly simplified because current sources, the number of which is limited by a maximum harmonic order theoretically set to 50, replace the controlled machines model.
It should be noted that a current source is always associated with a very high resistance (in the order of the MΩ) shunt-connected to the current source.
The model is then validated by comparing the results of the simulation with measurements made at the intakes of the production lines.
The model is finally used to simulate, with credibility, the effect of the compensating device 7 on the signal traveling through the loads of the industrial site and on the disturbance current circulating in the network 1.
In an embodiment, the system includes a plurality of renewable-energy-generating power units 100, for example of windpower or photovoltaic type, each one connected, directly or via a device 7, to a local network 28, itself connected to the network 1.
The local network 28 corresponds to the producing, consuming and consuming-producing stakeholders (of the self-consumption or positive-energy buildings).
Most of the time, each production unit 100 is connected to the local network 28 means of a compensating device 7 configured to maximize the power production of the production unit 100 as well as to prevent the propagation of the electrical disturbances, if any are present, on the upstream side of the compensating device 7 toward the local network 28 and of the electrical network 1.
Each compensating device 7 is connected, by computer systems, to a semi-decentralized management system 29, to which it communicates, in real time, information about the actual and predicted energy consumption and production of the production units 100.
Furthermore, the semi-decentralized management system 29 receives information in real time on the producible energy of the conventional (fossil fuel) energy production units 80, installed within the local network 28.
The semi-decentralized management system 29 manages the economic distribution of the conventional (fossil) energy generators within the local network 28, as a function of the total production from renewable sources as well as the total consumption within the same local network 28.
The semi-decentralized management system 29 receives in real time the information about the renewable- and conventional-source powers at any time over the local network 28.
The system also includes a plurality of consumption sources, for example a smart building or a smart industrial site, each consumption source being comparable from the point of view of the network to a non-linear/linear load 2; each of the consumption sources being connected to the local network 28, is equipped with a compensating device 7 configured to decontaminate the current circulating on the local network side 28 of the disturbances caused by the load 2 and to drive the activation of the different apparatuses of the consumption source via the decentralized management unit 70.
Each compensating device 7 is connected to the semi-decentralized management system 29, to which it communicates information concerning the instantaneous consumption and the future consumption as a function of the scheduled operation of the consumption sources, in such a way as to estimate the energy requirement.
Since the semi-decentralized management system 29 receives at all times all the data (actual and predicted) of the producible energy as well as those to be consumed by the different stakeholders of the local network (producing, consuming, consuming-producing), it can intervene, only as needed when the total energy requirement within the local network 28 is greater than the total production, on the drivers of the decentralized control units of the smart buildings 27 to tip the consumption mode into modulated mode (flat load curve) for the benefit of the overall system.
The term “consuming-producing source/stakeholder” is understood to mean a self-consumption or positive-energy building. The semi-decentralized management system 29 is therefore configured to estimate, over a given period, the total energy requirement of the local network 28 that it supervises.
According to the power production and energy requirement estimates, the semi-decentralized management system 29 is configured to
The management system 29 is therefore authorized to prioritize the operation of certain consumption sources over others, in order to distribute the energy requirement over the given period.
In addition, the semi-decentralized management system 29 can distribute the power demand over time in such a way that, when a power coming from the network 1 is necessary, it is consumed during low-demand periods in such a way as to minimize the loading of the network 1 during demand peaks.
It should be noted that the maximization of the power of the renewable energy production units 100 as well as the optimization of the consumption, within a smart building 27, are provided locally by the devices 7 via the control units 12. The semi-decentralized management system 29 is thus called as it is used only to ensure the economical distribution of the production of the conventional energy production units 80 and to rectify the consumption as needed.
Following a study concerning the invention conducted by the inventors, the inventors desire to make the following comments.
1. General Structure of the Shunt Active Filter of the Prior Art
General Topology
The control circuit, meanwhile, is composed of:
Output Filter
The output filter is a passive filter used to connect the voltage inverter to the electrical network. The output filter is dimensioned to satisfy the two following criteria:
with Ih_L the harmonic current contained in the current of the load IL and Iinj the current of the active filter injected into the network,
Two types of output filter can be used: a first-order output filter and a third-order output filter.
First-Order Filter
This type of filter is the most used in literature; it is composed of a simple inductance Lf of almost negligible internal resistance.
A filter of this type does not make it possible to simultaneously satisfy the two dimensioning criteria of the output filter. Specifically, only a relatively low value of Lf can produce a good dynamic response in the active filter by satisfying the equality relationship above.
Unfortunately, a low value of Lf allows most of the components due to switching to propagate on the network side and consequently to affect the installations and electrical equipment.
Conversely, a relatively high value of Lf will prevent these components from propagating over the electrical network but will affect the dynamic response of the active filter and will then degrade the compensation quality.
The proper dimensioning of the first-order output filter will therefore depend on the trade-off to be found between the dynamic response and the efficiency of the shunt active filter.
This trade-off is very difficult to strike without using an auxiliary passive filter installed at the output of the inverter or upstream on the network side.
However, this auxiliary filter can cause undesirable side-effects such as resonance with other passive elements installed on the electrical network.
It also gives rise to an active power consumption by its damping resistance.
In addition, the filtering quality of these auxiliary filters degrades over time because of the aging of their passive elements.
Third-Order Filter (LCL): Modelling in the s-Plane
The third-order output filter is an alternative to a heavy, bulky and expensive first-order filter, making it possible to dispense with the problems described in the case of the first-order output filter.
This output filter is composed of two inductances (Lf1, Lf2) of respective internal resistances (Rf1, Rf2) and a capacitance Cf with a small damping resistance Rf (voir la
It should be noted that (Ls, Rs and es) respectively represent the inductance and the resistance as well as the electromotive force of the electrical network upstream.
This type of filter, owing to the additional degree of freedom, provided by the capacitance Cf, can fulfil the two dimensioning criteria of the output filter that we have previously described.
The equations that model the output filter are:
with Vf the output voltage of the inverter, B1(s)/A(s) the transfer function of the output filter with the network corresponding to the original system (to be controlled) and B2(s)/A(s) the transfer function corresponding to the disturbance model.
These disturbances are caused by the voltage of the electrical network e(s) will no be considered equal to the connection voltage Vs for electrical networks with a high short-circuit power.
with
From the preceding relationships, if we neglect all the resistances (except for Rf), the following relationship, valid at frequencies greater than 50 Hz, can be established:
The resonant frequency fcp of the filter LCL, if in this case the resistance Rf is also neglected, is given by the relationship:
The LCL output filter is dimensioned to reject the components due to the switching frequency of the inverter, which has been set to 16 kHz, to correspond to an industrial application situation of the kind proposed here.
Thus, a rejection of over −50 dB is obtained for a cut-off frequency of 1900 Hz.
This choice has made it possible to properly attenuate the high-frequency components, as shown in the diagram of the gain of the output filter as a function of the frequency from
Third-Order Filter (LCL): State Space Modelling
Based on the Laplace representation (Maths. 1), the representation of the LCL filter in the state space can be given by the following equations:
The vector-matrix representation is given by:
{dot over (x)}=Ax+Bvf+Pvs (Maths. 3)
with
2. Linear Control by RST of the Voltage Inverter of the Prior Art Connected to the Network Via the LCL Filter
RST Controller
The control strategy is based on the estimation of the current disturbances by means of an identifying algorithm. Next, the voltage inverter, controlled by the PWM (Pulse Width Modulation) controller, generates the currents injected into the network Iinj, which must track the identified reference currents Iref (Iinf→Iref). The control loop thus closed is designed to provide high-accuracy tracking.
The general diagram of the current-controlling system is shown in
The voltage of the network Vs here represents an external disturbance, the effects of which are compensated by adding the same network voltage to the control signal (u). This will prevent the reactive fundamental current from passing from the network to the active filter via the inductances of the LCL filter.
From the equation (1) and by using an RST controller, we obtain:
With R(s), T(s) and S(s) the polynomials of the controller.
The order of R(s) and S(s) is the same as the order of the LCL system
therefore the polynomials R(s) and S(s) are of third order. The polynomial T(s) is chosen such that
for the whole frequency band (50-2500 Hz) included in the reference signal Iref; T(s) can be a simple gain in this case.
It should be noted that
represents the transfer function of the RST controller; the common denominator D(s)=(S A+R B1)(s), named arbitrary stability polynomial, contains the poles of the closed loop. These poles are placed in a sector of the (2×45°), to provide a damping of 0.7.
Finally, the poles of the control loop are placed in order to provide a quick and accurate response, with a good rejection of disturbances. It should be noted that the values of the poles are limited by the cut-off frequency in a closed loop.
Effect of the Phase Difference
RST controllers, and all linear controllers, can be used when the references to be tracked are composed of constant signals or signals at a single and relatively low frequency (the case of compensation of the reactive power or the unbalance at the fundamental frequency of 50 Hz). At this frequency, the phase difference between the identified references (Iref) and the output of the closed loop injected (Iinj) is acceptable. On the other hand, if the reference to be tracked is composed of signals with several frequencies, the phase difference is no longer negligible. This is because the phase difference increases with the frequency. The effect of the phase difference of the structure shown in
The transfer function of the control loop with the RST Iinj(s)/Iref(s) is given in
The gain and the phase of the closed-loop transfer function Iinj(s)/Iref(s) are given via Table I, for multiples of the fundamental frequency ranging from 50 Hz to (23×50) Hz. It should be noted that the frequency was limited to 1150 Hz here since the electrical networks self-filter high-order harmonic currents via their inductance.
Table. I. Gain and phase of the closed-loop transfer function with RST
It should be noted that the RST linear controller provides a unit gain (0 dB) for practically the whole frequency band (50-2500 Hz: see
Specifically, at this frequency, a phase difference of (−1°) is negligible. After this frequency, the phase difference is no longer negligible, and the shunt active filter cannot compensate the harmonic currents.
3. Non-Linear Control (by Sliding Mode-of First Order) of the Voltage Inverter According to the Invention Connected to the Network Via the LCL Filter
In most cases, the control laws, designed to control the voltage inverters which are connected to the electrical network via an LCL filter, were originally established for renewable energy systems (photovoltaic and windpower). In this case, provision can be made for linear controllers since the inverter has only fundamental components (an active power) to inject into the network: the phase difference being negligible in this case.
However, in the case of a shunt active filter, the inverter must inject into the electrical network fundamental components (reactive and unbalanced) and harmonics at the same time, which means having total control over a very wide passband of frequencies. It should be noted that a considerable phase difference limits the applicability of linear controllers in the case of harmonic filtering.
The alternative for overcoming this phase difference problem, encountered when using an LCL filter associated with a linear controller, is non-linear control. In this invention, we are using the SMC (Sliding Mode Control) method to provide the desired dynamic response, a strong robustness/insensitivity to bounded disturbances as well as good control properties under a wide range of operating conditions.
However, it is well known that a conventional SMC controller generates a control signal, from the switches of the inverter, at very high frequency, in order to provide finite-time convergence on the sliding surface wherein the states of the system are maintained for the remaining time, even in the presence of bounded disturbances. In practice, the voltage inverter is controlled by a limited/fixed frequency switching function. Thus, variable switching at very high frequency, if it occurs, could cause overheating that can result in the destruction of the inverter.
Consequently, a conventional continuous-control SMC (with Sign function) is used. Specifically, in order to avoid discontinuous control (variable switching of the inverter at very high frequencies), caused by conventional SMC, the choice was made to use two controllers per sliding mode. One is associated with a sign function approximated to a sigmoid function and the other is with a method that will be called AIRD (an Artificial Increase Relative Degree, followed by an integrator). Finally, the control of the active filter generated by the continuous SMCs will be modulated by a PWM, in order to allow the active filter to operate at a fixed switching frequency, adapted, on one side, to the nominal operation of the electronic power components of the inverter and easy, on the other, to filter out by the LCL, which facilitates, inter alia, the blocking of the high-frequency components by the LCL filter.
Design of a Conventional Sliding Mode Controller
Let us consider the system (3) with: y:=iinj representing the input, u:=vf of the control signal and w:=vs disturbances, and for Rf=0, the equation (3) becomes:
{dot over (x)}=Ax+Bu+Pw, y=Cx,C=[0 1 0] (Maths. 4)
with
It is clear that the relative degree of the system (4) is r=3, since CB=CAB=0 and
so a sliding variable is chosen of the form:=K0e+K1ė+K2ë (Maths. 5)
with e=iinj−iref. It should be noted that ė, ë are obtained using a higher-order sliding mode differentiator.
The positive coefficients K0, K1 and K2 of equation 5 are selected to make the system (Maths. 4) exponentially stable with the desired convergence rate, in the sliding mode defined by =0.
To design an SMC controller that derives →0 in finite time and which keeps the states of the system (Maths. 4) inside a sliding surface
=0 for the entire remaining time, the dynamics of the sliding variable are derived:
with
It is supposed that:
are bounded,
Then, there exists η>0 such that
The condition of existence of a sliding mode .
<0 can be easily fulfilled by the SMC
u:=vf=−λ sign() (Maths. 8)
with u the control signal applied, via the PWM, to the inverter, λ is set and defined by the saturation limiting block at 420V (see is proportional to
which explains the choice of K0, K1 and K2 in (7):
K0=306000000, K1=35000 and K2=1
In this context, it is important to indicate that the Sign function in (Maths. 8) has been approximated to a Sigmoid function sign
with ε=2×108
SMC with Artificial Increase of Relative Degree (AIRD)
The system (Maths. 4) is extended as follows:
x4=u, {dot over (x)}4=v
with v the new control, the system, in this case, becomes
Given that the relative degree r=4 of the system (9), the sliding variable is chosen of the form:e=K0e+K1ė+K2ë+K3
(Maths. 10)
with e=iinj−iref.
The derivative with respect to time of the sliding surface (Maths. 10) gives:
with
It should be noted that ė, ë and (in Maths 11) are obtained using a higher-order sliding mode differentiator.
The positive coefficients K0, K1, K2 and K3 are selected, in this case also, to provide an exponential dynamic stability for the system (Maths. 9) with the desired convergence rate in the sliding mode defined by e=0.
It is supposed, in this case, that all the disturbances and their derivatives, all the derivatives of the tracking error and all the state variables of the system are bounded in a reasonable state domain, which includes the operating point.
Then, there exists η1>0 such that |e|v=0≤η1 and the condition of existence of a sliding mode
.
<0 for
e≠0 can be easily fulfilled by the SMC
v:=λ sign(e) (Maths. 13)
where the value of λ=107 is chosen to limit the effect of the disturbances caused by the additional derivative coming from the artificial increase of relative degree.
Finally, a continuous control signal u=∫ v dt is generated and limited by the saturation block to 420V (see e is proportional to
which explains the choice of K0, K1, K2 and K3 in (Maths 11):
K0=1.7576 e+13, K1=2.0280 e+09 and K2=78000 and K3=1
4. Non-Linear Control (by Higher-Order Sliding Mode) of the Voltage Inverter According to the Invention Connected to the Network Via the LCL Filter
In this invention, the use of modulated higher-order sliding mode control algorithms: followed by Pulse Width Modulation (PWM), guaranteeing a predefined switching frequency for the controlling of the components of the voltage inverter associated with the LCL filter, is proposed for the first time.
It should be noted that the document [D2: ALALI MHD A E ET AL: “A Lyapunov approach based higher order sliding mode controller for grid connected shunt active compensators with a LCL filter”, 2017 19TH EUROPEAN CONFERENCE ON POWER ELECTRONICS AND APPLICATIONS (EPE'17 ECCE EUROPE)] proposes a higher-order Lyapunov control in which continuity of the controller is not provided: the control is discontinuous (see
In addition, this control is impossible to implement, because of its discontinuity and also its dependence on disturbances caused, inter alia, by the habitually disturbed network voltage, especially in industrial areas.
It should be noted that the Lyapunov control approach is adapted to systems that process constant signals (with a frequency equal to zero) in a healthy environment (free of disturbances). Specifically, addressing the problem of harmonics (with a frequency band including 50-2500 Hz) in a disturbed environment (electrical networks in general and especially industrial areas) while providing stability of the active filter, renders the control delivered by this control approach discontinuous.
In this invention and for the first time, three continuous sliding mode control algorithms of second order and higher order are used. This involves the 2-SMC Twisting and 2-SMC super-Twisting second-order sliding mode control algorithm and also the C-HOSM continuous higher-order sliding mode control algorithm.
These controllers supply, within a wide range of operating conditions, a desired dynamic response of the system in sliding mode, a strong insensitivity to the corresponding bounded disturbances and an improved stabilization accuracy. It should be noted that these algorithms, C-HOSM and 2-SMC Twisting and Super-Twisting, supply finite-time convergence on the sliding surface.
Therefore, the control unit 12 can thus include at least one of the three following higher-order sliding mode controllers:
Design of the C-HOSM Controller
The C-HOSM control algorithm can control systems with an arbitrary relative degree. Here, the sliding variable is selected as follows:=ė+ce (Maths. 14)
The relative degree is r=2 in accordance with the equation. (Maths. 4).
The origin of the equation (Maths. 14) is as follows. In a noisy measurement environment, the sliding variable will converge on a domain, the size of which is proportional to the amplitude of the noise w(t) of the measurement
. Consequently, the real sliding mode
=ė+c e=w(t) and the effect of the noise on the tracking error e is attenuated owing to a low-pass filter of a cut-off frequency equal to c (in our case, c=104).
The dynamic response of the input-output sliding variable S is computed in accordance with the equations. (Maths. 4), (Maths. 14)=v+f(x,t) (Maths. 15)
The control law of the C-HOSM algorithm is designed as follows:
v=v1−v2 (Maths. 16)
where:
v1=−γ1┌┘α
┘α
where ┌┘α
|α
) (Maths. 17)
With:
γ1,γ2>0 selected so that the polynomial p2+γ2p+γ1 is Hurwitz, with a desired pole placement,
(α1,α2) computed such that
if α=1/2 therefore,
v2 is chosen for a disturbance rejection such that:
v2=−ω (Maths. 18)
where:
{dot over (ω)}=−β Sign(σ)−d/dt(λ┌σ┘1/2)⇒ω=−β∫0t Sign(σ)dt−λ┌σ┘1/2
and σ=−∫0tv1dt (Maths. 19)
It should be noted that the control law in (Maths. 16)-(Maths. 19):
Computation of γ1,γ2: They are computed as coefficients of a second-order polynomial; the eigenvalues of this polynomial are chosen to provide a given transient response while being limited by the switching frequency of the voltage inverter, imposed by the carrier of the PWM.
Computation of λ,β: They are computed as follows:
λ=1.5√{square root over (L)},β=1.1L (Maths. 20)
In this case, L is equal to 3×1018
Finally, the control signal u applied to the components of the inverter is given:
u=((Ls+Lf2)Lf1Cf)v
with (Ls+Lf2) Lf1 Cf=a1 of the transfer function B1(s)/A(s) of (Maths. 1).
Design of the 2-SMC Twisting Controller
The sliding variable is chosen such that:=γ1e+γ2ė+ë, γ1,γ2>0 (Maths. 21)
Then, =v+φ where φ(x, t) represents the cumulative disturbances of the system shown by (Maths. 4). Deriving the derivative of the sliding variable
gives:
={dot over (v)}+{dot over (φ)} (Maths. 22)
with {dot over (φ)} supposed to be bounded in a reasonable state field of the system given by (Maths. 4).
The 2-SMC Twisting control, which makes ,
tend to zero in finite time, is designed as a function of the derivative of the control signal v:
{dot over (v)}=−α(β sign()+sign(
)) (Maths. 23)
Computation of γ1,γ2: They are selected as coefficients of a Hurwitz polynomial; the eigenvalues of this polynomial are chosen as a function of this criterion, while being limited by the switching frequency of the voltage inverter, imposed by the carrier of the PWM.
Computation of α,β: with α,β>0; They are respectively equal to 108, 0.6. They are chosen to provide a rapid response as well as an acceptable reduction of the very high switching frequency.
Finally, the control signal applied to the components of the voltage inverter is
u=∫0t{dot over (v)}dt (Maths. 24)
It should be noted that the control signal u in (24) is continuous, whereas {dot over (v)} has a very high-frequency (discontinuous) control.
Design of the 2-SMC Super-Twisting Controller
The sliding variable is chosen such that:=γ1e+γ2ė+ë, γ1,γ2>0 (Maths. 25)
Then:=u+f(x,t) (Maths. 26)
with f(x, t) representing the cumulative disturbances of the system, shown in (Maths. 4), the derivative of which is supposed to be bounded.
The 2-SMC Super-Twisting control, which makes ,
tend to zero in finite time is designed such that:
u=−λ┌┘1/2−v (Maths. 27)
with
{dot over (v)}=−β Sign() and v=−β∫0t sign(
)dt (Maths. 28)
Computation of γ1,γ2: As in the two preceding cases, these parameters are computed as coefficients of a second-order Hurwitz polynomial, consequently the eigenvalues of this polynomial are chosen to provide a rapid response, while being limited by the switching frequency of the voltage inverter, imposed by the carrier of the PWM.
Computation of λ,β: They are chosen as follows:
λ=1.5√{square root over (L)},β=1.1L (Maths. 29)
With a value of L high enough that {dot over (f)}(x, t)≤L, L being in our case equal to 1×107. It should be noted that the Super-Twisting control law (Maths. 27)-(Maths. 29) generates continuous control.
Simulation Results of the Inverter Controlled According to the Invention
Results of the Simulation with Matlab-Simulink
First-Order Continuous SMC Controllers
The simulations were carried out, firstly, via a simple Simulink scheme. In this scheme, the identified reference harmonic currents Iref are modelled by current sources of order 5, 7, 11, 13, 17 and 19, which represent the same harmonic spectrum of the current that will be analyzed via a case study below. The simulations were carried out first using the conventional SMC then the SMC with sigmoid function and finally with the SMC combined with the AIRD algorithm.
Higher-Order SMC Controllers
The simulations were carried out, for the same characteristics of the preceding case, this time using first the conventional SMC then, and in order C-HOSM, 2-SMC Twisting, 2-SMC Super-Twisting and finally the Lyapunov approach of the document D2.
Results of the Simulation with Matlab®-Simulink®-Simscape Power System® (Case Study)
This is at the Melaiece textile factory located in the industrial area of Sheikh Najjar, in Aleppo, in Syria; the electrical network of the factory is given in
The production line of this factory contains 56 asynchronous motors including 40 controlled by power inverters and with a nominal power varying between 0.25 and 75 kW. The power factor of the factory is of 0.76 without compensation of the reactive power, then goes back up to 0.98 with compensation via banks of controlled capacitor batteries.
Modelling of Low-Voltage Networks
Given the large number of machines controlled in an industrial site (40 in this case), it becomes difficult, or even impossible, to model the electrical networks of these installations.
A first restriction is hardware-related. Specifically, there need to be as many measuring apparatuses as there are controlled machines (40 in this case: these are energy quality analysis devices, shown in
In addition, it proves difficult to carry out a simulation of an industrial network at several voltage levels, containing several power transformers with long power cables and lines, as well as dozens of machines controlled and combined with several active decontamination filters.
The idea here is to replace the machines controlled by their currents (active, reactive and harmonics) delivered by the analysis devices.
In this case, an analyzing device will be installed at the intake of each production line, so only one measuring apparatus per production line instead of one measuring apparatus per machine.
From the point of view of the computer systems, the simulation scheme is no longer burdensome since current sources, the number of which is limited by a maximum harmonic order, theoretically set to 50, will replace the model of dozens of controlled machines.
It should be noted that a current source is always associated with a very high shunt-connected resistance (of a few MΩ).
The direction of passing of the currents is chosen according to the convention of the direction of travel of the powers through an electrical network: the harmonic currents are disturbance currents, so they are injected by the so-called contaminating load, whereas the fundamental current is supplied by the network.
It should be noted that the fundamental current can be replaced by a three-phase R-L load; the sources of the fundamental currents I1 will, in this case, be removed from the preceding model.
Validation of the Proposed Model
In order to model the production line of the textile factory being studied, electrical measurements were taken at the intake of the production line, as shown in
It should be noted that these measurements were limited by the maximum harmonic order 50 (50×50 Hz).
Finally, the parameters of the electrical network, shown in
Based on the measurements shown in
In Table III, the quantities PLoad, QLoad, SLoad, respectively are the active, reactive and apparent powers of the overall load of the production line.
Based on the actual measurements as well as the technical data sheets shown via
By comparing the simulation results with the measurements given in
These results validate the proposed models and especially the non-linear loads model (of machines controlled by power converters; such as inverters, rectifiers etc.) Based on these results, it may be considered that the results that will now follow are also valid.
Result of the Simulations of the Active Filter Installed in the Network of the Factory being Studied
First-Order SMC Controllers
This part will validate the use of the active filter associated with the LCL filter with the control algorithms, conventional SMC, SMC with a sigmoid function and SMC with the AIRD method.
The technical data sheet of the components forming the active filter is given in Table IV.
The active filter is triggered after 5 periods of the mains power supply (up to 0.1 s); it will operate, first during 3 periods (up to 0.16 s) with the conventional SMC (with discontinuous control).
Afterwards, during 3 periods (up to 0.22 s), it will continue with the continuous control SMC (associated with a sigmoid function).
Finally after 0.22 s and until the end of the simulation, (0.28 s), the active filter will be controlled by a continuous control coming from the SMC with the AIRD algorithm.
It can be deduced from
This can be observed by negligible THDs of the three-phase currents after filtering. Specifically, the (THD-Is) after filtering is of: 0.4%, 0.8%, 0.5%, these values being the highest of the control methods discontinuous control (conventional) SMC, continuous-control SMC (with a sigmoid function) and continuous-control SMC (with the AIRD method) respectively. It should be noted that the THDs of the currents before the activation of the active filter were respectively for the three phases (11.6%, 10.7% and 12%).
The significant reduction of the THD of the current leads, as shown in
It should be noted that the international standards (e.g. IEEE STD 519-2014) impose a maximum voltage THD between 5%-8% in industrial areas, whereas the EDF recommendations require a total rate of voltage harmonic distortion of less than 1.6% for each receiver.
Higher-Order SMC Controllers
This part will validate the use of the active filter associated with the LCL filter with the C-HOSM then 2-SMC Twisting and finally 2-SMC Super-Twisting control algorithms.
The technical data sheets of the components forming the active filter is given in the previous Table IV.
It can be deduced from these figures that, despite a deteriorated and unbalanced voltage of the network and also the presence of the LCL filter, the three control methods provide quasi-perfect tracking, after filtering, by a sinusoidal shape of the current on the electrical network side.
Specifically, the (THD-Is) after filtering for the three phases are: (0.92%, 0.65%, 0.75%), (0.55%, 0.35%,0.42%) and (0.31%, 0.15%, 0.23%) for the control algorithms C-HOSM, 2-SMC Twisting and 2-SMC Super-Twisting respectively. It should be noted that the THDs of the currents before the activation of the active filter are for the three phases (11.6%, 10.7% and 12%) respectively.
These very good current filtering results are expressed as a significant drop in the voltage three-phase THD (THD-Vs) of (6.7%,6.2% and 6.8%) for the three phases before and after filtering of (0.59%,0.55%,0.66%), (0.23%,0.16%,0.25%), (0.2%,0.15%,0.22%) respectively, using the control algorithms C-HOSM, 2-SMC Twisting and 2-SMC Super-Twisting respectively.
It should be noted that the three control algorithms proposed preserve the robustness and very good filtering quality provided by discontinuous conventional SMC, while guaranteeing, via continuous control, a voltage THD well below the 1.6% required by the most stringent recommendation (EDF standard). This very important result is achieved in spite of the unbalanced currents and voltages of the network, the significant harmonic distortion in the voltage of the site being studied, and also the presence of the LCL filter.
It should be noted that the international standards (for example, IEEE STD 519-2014) impose a maximum voltage THD between 5%-8% in industrial areas, whereas the EDF recommendations require a total harmonic distortion rate of less than 1.6% for each receiver.
Computation and Integration of the Maximum Power Point in the Control Scheme of the Active Filter:
The aim here is to track the maximum power point of a photovoltaic generator, in order to increase the efficiency of this generation system.
In this invention, the algorithm used for the extraction of the maximum power point is the P&O (Perturb and Observe) algorithm, which is based on the perturbation and observation of the voltage of the PV generator, until the obtainment of the maximum voltage which will correspond to the MPPT point.
DC Voltage Regulation Loop (Maximum Power Tracking):
The loop for regulating the voltage of the capacitor, on the DC side of the inverter, will now be made use of to provide the tracking of the maximum power of the PV generator. Specifically, in the case of a shunt active filter which is not connected to a PV generator, the energy storing capacitor regulates itself (charges while maintaining a constant voltage) via the electrical network, through the inverter, to compensate for the Joule-effect losses of the electronic power components of the inverter and of the LCL output filter. The voltage of the capacitor Vdc must track a reference voltage Vdc-ref the amplitude of which is chosen to reinforce the dynamic response of the system, while complying with the dimensioning of the electrical components of the shunt active filter.
In this invention, the same loop for regulating the voltage of the capacitor is used to provide the tracking of the maximum power point, based on the tracking of the power PMPP and not of the current IMPP. Specifically, unlike the methods for regulating the capacitor of the inverter, which use the tracking of the current IMPP if a renewable energy generation system is present, this method will directly extract the maximum power of the PV generator.
Modelling of the Capacitor Voltage Regulating Loop:
The relationship between the active power produced by the PV generator and the voltage across the terminals of the capacitor can be written in the form:
The relationship (Maths 30) being non-linear, and for small variations in the voltage Vdc around its reference Vdc-ref, it can be linearized via the following relationship:
It should be noted that for: Vdc-ref=VMPP (delivered by the P&O algorithm of the unit 25-B), this gives Ppv=PMPP. Consequently, the voltage of the capacitor computed in the Laplace domain becomes:
Based on the relationships (Maths 31) and (Maths 32) and using a linear regulator (Proportional-Integral) or another, the loop for regulating the DC voltage Vdc and consequently the power PMPP can be shown via
Integration of the Loop for Tracking the Maximum Power within an Active Filter:
The advantage of tracking PMPP (compared to IMPP) is being able to incorporate (with as few computation operations as possible) the PMPP tracking loop into the algorithm for identifying the disturbance currents of the control part of the shunt active filter. In this case, the reference voltage Vdc-ref of the algorithm for identifying/regulating the voltage of the capacitor of the active filter, is replaced by the voltage of the maximum power VMPP (Vdc-ref=VMPP) coming from the algorithm (P & O) as shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 18305899 | Jul 2018 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2019/067779 | 7/2/2019 | WO |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2020/007884 | 1/9/2020 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 20130030590 | Prosser | Jan 2013 | A1 |
| 20160329714 | Li et al. | Nov 2016 | A1 |
| Number | Date | Country |
|---|---|---|
| 201813171 | Apr 2011 | CN |
| 102299556 | Dec 2011 | CN |
| Entry |
|---|
| European Search Report for Application No. 18305899.9 dated Feb. 26, 2019, 3 pages. |
| International Search Report for Application No. PCT/EP2019/067779, dated Oct. 22, 2019, 3 pages. |
| Mohamad A. E. Alali, et al., “A Lyapunov Approach Based Higher Order Sliding Mode Controller for Grid Connected Shunt Active Compensators with a LCL Filter”, 2017 19th European Conference on Power Electronics and Applications (EPE'17 ECCE Europe), published Sep. 2017, pp. 1-10, IEEE. |
| Number | Date | Country | |
|---|---|---|---|
| 20220200282 A1 | Jun 2022 | US |
