Patent Application
20180313327
The present invention relates to control methods for wind farms connected to a power grid, in the event of disturbances in the power grid, and to associated wind farms.
The use of wind energy as a source of electrical energy is widely known today. Wind energy is obtained from wind turbines which convert kinetic energy of the wind into mechanical energy, and said mechanical energy into electrical energy. Wind turbines generally comprise a tower, a nacelle located in the apex of the tower, and a rotor which is supported on the nacelle by means of a shaft. A plurality of generators of different wind turbines can furthermore be grouped to form a wind farm.
The significant increase in the acceptance of wind energy generation has led many countries and power grid operators to implement strict grid connection requirements for wind farms, which requirements are also known as grid codes. Some grid codes, such as grid codes enforced in countries such as Germany and Spain, to name two examples, require the wind turbine generators of a wind farm to comply with specific requirements continuously and before, during and after disturbance, both in steady-state and during disturbances in the power grid, such that the wind farm remains connected to the grid during disturbance in the grid, and so that it performs reactive control at its point of common coupling (PCC) according to a voltage drop-based injection profile during disturbance in the grid.
In the current prior art, wind farm response during disturbances in the power grid changing the voltage of the point of common coupling of the wind farm such that it is outside the permanent operating range is performed locally in each of the elements forming part of the wind farm (wind turbines and/or reactive compensation units such as STATCOM (“STATic synchronous COMpensator”), if any, for example). This solution uses the quick response capacity of each of the dynamic elements of the farm, but this response is not optimized since each of these elements is not communicated with the rest of the elements nor does it have access to the measurements of the point of common coupling.
It is impossible to assure suitable operation of the whole wind farm with this solution, rather suitable operation of each of the controllable elements of the wind farm is assured individually, and there is a need extrapolate what the operation of the wind farm as a whole shall be by defining an estimated behavior of all the elements of the wind farm, such as: wind in each wind turbine, number of operating wind turbines, and state of the variable sub-station elements (compensation units, capacitors, inductances, switches, etc.). This estimation must be performed for a single scenario, so it does not always represent the actual state of the wind farm and does not assure appropriate compliance with the requirements at the point of common coupling.
This limitation is compensated for by negotiating the possible deviations that may occur at the point of common coupling with the wind farm operator and carrier (TSO, “Transmission System Operator”), and if negotiation is not possible, by installing greater controllable generation capacity in the wind farm which allows covering the worst case scenario (using more wind turbine capacities, and/or installing wind turbines with greater performances, and/or installing compensation units such as STATCOM, for example, and/or increasing compensation unit capacities, for example).
Patent document WO2015078472A1 describes a control with which the injection and absorption of reactive power in a wind farm are controlled. In addition to wind turbines, the wind farm described herein comprises reactive power regulating devices (reactive compensation units), such as MSU (“Mechanically Switched Unit”) and STATCOM devices. The reactive power generated by the regulating devices is controlled by means of the farm controller, such that the combined amount of reactive power produced by the wind turbines and by the regulating devices satisfies a desired amount of reactive power. In case of communication fault between the farm controller and one of the regulating devices, the farm controller is reconfigured to compensate for the capacity of said device and to inject or absorb the required amount of reactive power in/from the grid.
Patent document WO2015086022A1 discloses a method for controlling the injection of reactive current in a wind farm during a grid fault. The amount of reactive current that must be injected by the wind farm to the grid during the fault is measured, a difference between the reactive current that is being injected and the reactive current that must be injected is determined, and the wind turbines of the wind farm are controlled for generating the specific active current difference.
The object of the invention is to provide a control method for a wind farm and an associated wind farm, as defined in the claims.
A first aspect of the invention relates to a control method for a wind farm which is connected to a power grid and comprising a plurality of generating units, such as wind turbines, for example, and a local controller associated to each generating unit.
The presence or absence of a voltage disturbance in the power grid is determined with the method in a dynamic and recurrent manner. When the presence of a disturbance is determined, a control phase is implemented in a dynamic and recurrent manner while said presence lasts, during which the generating units are controlled so that they control the (active and/or reactive) power on the point of common coupling of the wind farm and thereby participate in stabilizing the grid voltage. Once the disappearance of said disturbance is determined, the method stops implementing the control phase.
When the disappearance of the disturbance is determined, and in the absence of another disturbance, in addition to stopping the implementation of the control phase, a stabilization phase is implemented in a dynamic and recurrent manner for a limited time interval. The limited time interval can be predetermined based on previous experiences and/or studies, for example, where the time elapsing between the disappearance of a disturbance and complete grid stabilization is estimated or measured, although it could be also be determined in real time, for example, based on measurements (preferably of the electrical characteristics of the grid). This means that the limited time interval may vary from farm to farm and from case to case, being greater in the case of weak grids and/or large wind farms and/or more sudden disturbances. This limited time interval is usually of the order of several seconds. In the stabilization phase, while the generating units continue to be controlled so that they control the power on the point of common coupling of the wind farm, such that a smooth and controlled transient is provided until the voltage of the grid stabilizes.
In summary, the presence or absence of a voltage disturbance in the power grid is determined in a dynamic and recurrent manner with the proposed method, and:
As long as the presence of a disturbance is not determined (and the stabilization phase is not being executed), the method implements a steady-state phase in which the objective thereof is to comply with the requirements applied to the grid by means of the generating units.
Unlike what occurs in the prior art where, in order to stabilize the power grid, action is only performed independently during the presence of a disturbance in said grid, the proposed method does not only help in stabilizing the grid during the presence of disturbances using the generating units in a coordinated manner but also helps to stabilize the grid during the transient occurring between the disappearance of said disturbances and the steady-state state of the power grid, a correct stabilization of the grid being greatly assured without it furthermore affecting the wind farm capacities once the disturbance disappeared. This furthermore prevents sudden changes in wind farm generation, for example, prevents sudden changes from being able to bring about negative impacts on the grid to which it is connected.
A second aspect of the invention relates to a wind farm which is connected to a power grid and comprising a plurality of generating units. The wind farm is suitable and configured for supporting and implementing a method such as the one of the first aspect of the invention, the same advantages as those mentioned for said method thus being obtained.
These and other advantages and features of the invention will become evident in view of the drawings and detailed disclosure of the invention.
A first aspect of the invention relates to a control method for a wind farm 100 connected to a power grid 1 and comprising a plurality of generating units 2, such as wind turbines, for example, and a local controller 3 associated to each generating unit 2. The wind farm 100 further comprises an associated converter 2a which is associated with each generating unit 2, each local controller 3 acting on its associated converter 2a for controlling the generation of power from the corresponding generating unit 2.
The method is executed in a dynamic and recurrent manner.
The presence or absence of a voltage disturbance in the grid 1 is determined with the method, and a specific control is performed on the generating units 2 of the wind farm 100 when the presence of a disturbance is determined in order to stabilize the grid 1, keeping the units 2 connected to the grid 1. In the context of the invention, voltage disturbance in the grid 1 must be interpreted as what is commonly known as a low-voltage disturbance or LVRT (Low-voltage Ride Through) or as high-voltage disturbance or HVRT (High-voltage Ride Through).
To determine the presence or absence of a disturbance, the method implements the following measurements in a dynamic and recurrent manner:
The presence of a disturbance is generally determined when the measured value of at least one of the measured electrical characteristics exceeds a respective associated predetermined maximum threshold value or is below a respective associated predetermined minimum threshold, the absence of a disturbance being determined otherwise. In particular:
Different controllers 3 can simultaneously determine the presence of a local disturbance, but only controller 5 can determine the presence of a disturbance at the general level of the wind farm.
When the presence of a disturbance is determined, regardless of the controllers 3 and 5 that determined it, a control phase which is executed in a dynamic and recurrent manner, while the presence of the disturbance continues to be determined, is activated. During said control phase, the presence or absence of a disturbance continues to be determined in parallel in the way mentioned above, such that when a disturbance disappears it can be determined with the method, and the control phase is deactivated (or stops to be implemented) when said disappearance is determined.
When said disappearance is determined (and the presence of any other disturbance is not determined), in addition to deactivating the control phase, a stabilization phase which is executed in a continuous and recurrent manner is activated for a limited time interval. When going into the stabilization phase, the farm controller 5 goes from the disturbance state to a stabilization state.
In the control phase, the generating units 2 of the wind farm 100 are controlled so that they comply with the power requirements required for the wind farm 100 in the presence of a disturbance (for example, the generation of current and/or power to be injected into the grid 1). Said control is under the responsibility of the farm controller through the corresponding local controller 3 in each case, or the local controller 3 itself which does not follow the possible instructions that may be received from the farm controller 5, as described in detail below. When the responsibility falls on the local controller 3 itself, in the context of the invention said local controller 3 is said to act independently or in local mode.
In the stabilization phase, while the generating units 2 of the wind farm 100 continue to be controlled in order to comply with the power requirements, for the purpose of providing a smooth and controlled transient after the disappearance of disturbance until the voltage of the grid 1 stabilizes in steady-state.
In general, during the disturbance and at the outlet thereof (during execution of control and stabilization phases) the wind farm 100 must preferably produce reactive current and/or power for stabilizing the voltage of the grid 1. The reactive current and/or power must furthermore be supplied in a dynamic manner according to the measurements taken and to the capacities of the generating units 2.
As long as the presence of a disturbance is not determined (and stabilization phase is not being executed), the method implements a steady-state phase in which the objective thereof is to comply with the requirements applied to the grid 1 (increasing/decreasing reactive power, following an instruction, etc.), by means of the generating units 2. In the steady-state phase, all the controllers 3 and 5 are in steady-state.
During method implementation, the farm controller 5 generates generation instructions at all times, regardless of whether or not a disturbance has been determined, and it transmits them to the local controllers 3, as applicable. In the steady-state phase, these instructions refer to the power to be generated by the generating units 2, and the local controllers 3 act on the corresponding generating units 2 depending on said received instructions.
During the control and stabilization phases of the method, however, the local controllers 3 may or may not follow these instructions, as described in detail below, but in any case this instruction generation allows the farm controller 5 to take over control of the generating units 2 as soon as possible and in the best way possible when the local mode of the corresponding local controllers 3 ends, moment in which the local controllers 3 act on the respective generating units 2 depending on the received instructions (the farm controller 5 acts as master).
In a first embodiment, in the control and stabilization phases the local controllers 3 act in local mode, not following the instructions received from the farm controller 5.
In a second embodiment, at the beginning of the control phase and/or the stabilization phase, and for a limited time interval (a transient) established by the corresponding local controller 3 for stabilizing the value of the local variables, the local controllers 3 act in local mode, and after said time interval has elapsed said local controllers 3 act depending on the instructions they receive from the farm controller 5. In the second embodiment, it is considered that the transient (time interval) has ended or has stabilized when the following conditions are complied with at the same time (in the moment in which all the conditions are present):
Generally, in the control and stabilization phases the actuation of the wind farm 100 as a response to any disturbance is improved when the farm controller 5 controls the local controllers 3, but the transitory response in the event of said disturbance is quicker if the local controllers 3 act in local mode. A combination of both advantages is optimally obtained with the second embodiment. In the second embodiment, the dynamic behavior and controllability of the wind farm 100 is thereby improved, which can turn it into the optimum solution for weak grids 1 and for when strict response times are required, for example.
Each local controller 3 is configured for causing the corresponding generating unit 2 to follow the instructions generated by the farm controller 5, or for said generating unit 2 to act in local mode, as mentioned above. Generally:
When a local controller 3 determines the presence of a disturbance and operates in local mode, said local controller 3 activates a reactive power generation mode for the generating unit 2 on which it acts if one of the following conditions is complied with:
Said local controller 3 activates a reactive current generation mode for said generating unit 2 if said value is outside both ranges, said reactive current generation mode being maintained while said disturbance lasts and said value is maintained outside both ranges. Said generating unit 2 thereby generates power according to a local reactive power reference when the local controller 3 is in reactive power generation mode, and generates power according to a local reactive current reference when the local controller 3 is in reactive current generation mode. The reference value can be an expected value of said electrical characteristic, or the value of said electrical characteristic measured before determining the presence of a disturbance, for example.
When a local controller 3 is in reactive current or power generation mode, it acts on the corresponding generating unit 2 so that said generating unit 2 generates reactive current or power, respectively, depending on said measured value, given that said value gradually changes as the required reactive is being generated (the grid 1 gradually stabilizes), the reactive needs thereby being changed.
The reactive current and power generated by a generating unit 2 are monitored at all times, and in the moment in which the corresponding local controller 3 determines the presence of a disturbance, the value of the generated reactive current and/or power monitored in that moment is frozen, the corresponding frozen value being able to be used optionally in the reactive current generation mode (in the case of frozen reactive current value) for determining the reactive current to be produced, and in the reactive power generation mode (in the case of frozen reactive power value) for determining the reactive power to be produced. In particular, the frozen value can be added to a predetermined offset value, as depicted with the following equations:
I=I
offset
+I
(Vmeasure)
Q=Q
offset
+Q
(Vmeasure)
If no frozen values are used, the terms Ioffset and Qoffset of the above two equations would be equal to zero.
Generally, it is preferable that a local controller 3 always act in the reactive power generation mode, but in some embodiments, despite complying with the conditions for operating in that mode, in a first moment of the stabilization phase and during a time interval which preferably is less than 100 ms, said controller 3 is caused to act in the reactive current generation mode in order to reduce the electric voltage of the grid 1 (if said voltage exceeds the reference value).
The wind farm 100 where the method is implemented can further comprise at least one compensation unit 4 with an associated local controller 6 for providing reactive (current and/or power) to the grid 1 when required, as depicted in
In this case, in the method the local controllers 6 also determine the presence or absence of a disturbance in a manner similar to that of the local controllers 3, and to that end at least one electrical characteristic at a local point of coupling PC associated with each compensation unit 4, preferably the voltage at said point of coupling PC, is measured. The operation of a local controller 6 is analogous to that of a local controller 3, so what is described for said local controllers 3 and the different operating possibilities of said local controllers 3 are also applicable to the local controllers 6.
A second aspect of the invention relates to a wind farm 100 shown by way of example in
The wind farm 100 further comprises an associated converter 2a which is associated with each generating unit 2, the local controller 3 acting on said converter 2a for controlling the generation of energy from said generating unit 2, and in the corresponding embodiments, it may further comprise an associated converter 4a which is associated to each compensation unit 4, as depicted in
The wind farm 100 further comprises sensors SPCC and SLV (and SPC, where appropriate) or detectors required for implementing the method, such as for example, those required in order to be able to measure the electrical characteristics based on which the presence or absence of disturbances in the power grid 1 is determined. These sensors SPCC and SLV (and SPC, where appropriate) are furthermore communicated with the corresponding controllers 3 and 5 (and with the local controllers 6, where appropriate).
| Number | Date | Country | Kind |
|---|---|---|---|
| 201700563 | Apr 2017 | ES | national |
