Research On Equivalent Modeling And Optimized Control Of Clustered Wind Turbine Group

Wind energy has become an effective way to alleviate energy shortage,cope with climate change,and promote low-carbon economic growth in the worldwide.The issues of "wind curtailment" and frequency stability brought by large-scale wind power grid connection have affected the safe and stable operation of the power system to some extent.When studying the problems of power adjustment and frequency modulation for wind power grid-connected system,the systematic and detailed model for each wind turbine will not only increase the model order,but also enhance the complexity of simulation and calculation.Deviding the wind farm to several clusters and simplying the whole wind farm model from the perspective of "clusters",the analysis and research for the entire wind power grid-connected system can be simplified greatly by studying power adjustment and frequency modulation issues of the clustered wind turbine groups.Taking the direct-drive wind farm as an object,the equivalent modeling method and the optimized control method of power adjustment and frequency modulation for the clustered wind turbine groups are studied in this paper.The main work of this paper includes the following aspects:(1)Based on the operating characteristics of the D-PMSG wind turbine,the wind farm multi-clustering indexes are determined.Based on the traditional FCM algorithm,the contribution of different samples and sample characteristics to the clustering results are considered,the corresponding weights and weight indexes are assigned to the wind turbines with different operating conditions and different operating parameters,an ASW-FCM clustering algorithm is proposed.The actual wind farm system is used as an example to verify and evaluate the proposed algorithm.(2)In order to improve the accuracy of the cluster wind turbine group equivalent model,the influences of wake effect and wind direction change on the input wind speed are considered for the different wind turbines in the same group.According to the constant output characteristics before and after the equivalence for the wind turbine,the equivalent wind turbine model is established.Based on the same voltage difference before and after the equivalence,the equivalent collecting lines model is established.The accuracy of the equivalent model is verified by comparing and analyzing the wind turbine group equivalent model,the wind farm multi-machines equivalent model with the detailed model established on the simulation platform.(3)Aiming at the low efficiency and dimensionality of traditional active power allocation algorithm for wind farm,an active power optimal allocation method based on the wind turbine group power adjustment capability is proposed.The power adjustment capability of the wind turbine group is defined and evaluated based on the common characteristics.A multi-objective optimization model of wind farm power allocation is established with the optimization targets of the power adjustment ability,the start-stop status and the wind farm output power control requirements.The improved NSGA-? algorithm with adaptive adjustment mutation method is adopted to solve the problem.Calculation examples verify that the proposed allocation method can give full play to the power adjustment capability of wind turbine group,optimize the start and stop plan of the wind turbine,and improve the output power stability of the wind farm.(4)In view of the restriction of the traditional additional inertia control and fixed de-loading level control on the primary frequency modulation ability for the wind turbine and group,the traditional additional inertia control is improved based on the fuzzy control principle,the de-loading level under different wind speeds is defined to achieve variable de-loading level control.Combining the variable de-loading level control and fuzzy adaptive additional inertia control for the full wind speed range,a frequency coordinated control method for the primary frequency modulation of the wind turbine and group is proposed.Simulation results under different wind speeds and load disturbances indicate the superiority of the proposed frequency modulation method and the steady-state frequency deviation and frequency change rate under load disturbance can be improved effectively.