Research And Application Of Active And Reactive Power Control Based On Measured Data For Wind Farms

Our country is building a new type of power system with the renewable energy as the main body.Wind power is one of the important part of the renewable energy.Its output power characteristics affect the stability of frequency and voltage in power system directly.Power network dispatcher puts forward higher requirements for the rapidity,accuracy and stability of active and reactive power control on the second time scale of wind farms(WFs).With the continuous increasing of the single capacity of wind turbines(WTs)and the single scale of WFs,the dynamic time-varying characteristics of the WTs and WT groups output power on the second time scale(This paper refers to this characteristic as the dynamic characteristic of WTs)will become more and more complex,and it is difficult to describe accurately by using traditional mechanism modeling methods.So,it is more and more difficult to achieve the precise control for the WFs.In the absence of an accurate mathematical model,the following urgent problems need to be solved to achieve the accurate controlling of the WFs,that is,using the measured data of WFs,how to study the dynamic characteristics of WTs and the smoothing effect of WT clusters on the second time scale through data mining,how to tap the power regulation potential of WF itself fully based on the dynamic characteristics of WTs,and develop the active and reactive power control technologies for WFs to improve the speed,accuracy and stability of power control,how to test the developed power control technology in the field and realize the engineering application.Based on the basic characteristics of WTs,this paper had taken the operation data of mountain and plain WFs as the research object,excavated the spatial trajectory of the variation of the operating state points of WTs,and revealed three essential laws of the time-varying dynamic characteristics of WTs under the second time scale namely the general law of output power fluctuation characteristics and cluster smoothing effect,the time-varying law of dynamic regulation performance and dynamic response characteristics.The essential mechanism of dynamic characteristics coordination and complementarity among WTs was analyzed deeply.Based on this,an active power dynamic interval control strategy and a reactive power control strategy based on PID-like control were proposed,and a wind farm dynamic simulation platform was developed which suitable for the simulation and verification of WF active power control.The proposed control strategy had been applied to the actual WF,and had passed the relevant grid-related tests of State Grid,and the control indexes met the relevant technical requirements.The main work of this paper and the research results are as follows:According to the power fluctuation characteristics and cluster smoothing effect of WTs in the second time scale,the mathematical relations among cluster smoothing effect,power correlation and turbines number were established based on mathematical statistics method.And based on the operation data of a mountain WF and a plain WF in China,the correlation of unit power at different spatial scales and wind speeds on the second time scale were quantitatively compared,and the spatial variation characteristics of the cluster smoothing effect were given.Finally,the influence of the smoothing effect on the volatility of the WFs cluster output power was discussed,which provided theoretical basis and data support for the subsequent research of this article.The concept of Dynamic Adjustment Performance(DAP)and Dynamic Response Performance(DRP)for WTs were proposed in view of the difficulty in describing the time-varying dynamic of operation state and wind speed fluctuation with the unified mechanism model.Through the large-scale measured data mining,the In-depth research on DAP and DRP was carried out respectively.By mining the trajectory curve of operating states under limited power operation mode,the dynamic relationship between the output(power)and the input(control parameters)of the WT was quantified,and the evaluation indexes of DAP and DRP were established,which provided theoretical support for the research of active and reactive power control of WFs.How to fully tap the wind farm potential of active power regulation,and to improve the rapidity,accuracy and stability of WF tracking target power,the coordination and complementary mechanism of second level dynamic characteristics of wind farm were analyzed.Based on this,the principle of realizing power regulation synchronization between each WT,between each WT and the whole WF were taken to improve the active power control characteristics of WF,a dynamic interval active power control strategy based on dynamic characteristics of WTs was proposed,and the detailed control framework and control process were given.Finally,the effectiveness and correctness of the control strategy proposed in this paper was verified through the research and analysis of the field application data.Aiming at how to improve the accuracy of reactive power control in WFs,combined with the operation data,the essential principle of reactive power regulation of doubly-fed induction generator(DFIG)and permanent magnet synchronous generators(PMSG)were analyzed from the perspective of the generator operation essential principle,the reactive power regulation capability of DFIG and PMSG were discussed.A reactive power control strategy based on PID-like control was proposed,and a detailed control framework and control process were given.The actual results of field application have shown that the proposed method could improve the regulation accuracy of the reactive power,and could significantly improve the reactive power output characteristics.Aiming at the problems that the existing wind farm simulation platform does not have the function of limited power operation and cannot simulate and verify the active power control,the dynamic simulation model of WT was constructed,and the dynamic WF simulation platform was built.The simulation platform was verified in an actual WF.Through the analysis of actual verification data,it showed that the simulation platform has the conditions of WF active power control simulation and verification.The active power control system based on data mining was installed in a 96 MW mountain WF in China.The technical indicators of the AGC(Automatic Generation Control)had met the relevant requirements of the dispatching department,and the closed loop test of the dispatching AGC master station had been passed.So far,the system has been running safely and stably.