Active Power Distribution Optimization Scheme Of Wind Farm Based On DFIG Turbine Equivalent Model

With the constant rise of China’s wind power installed capacity, grid-connected wind power has tremendous impact on the power system stability because of its volatility and anti-peak regulation. Strengthen the stability and controllability of wind power can reduce the incidence of wind abandoned conditions. To deepen the study of wind turbine power control and wind farm active power allocation scheme is necessary. Wind farms need to be able to follow the grid dispatching signal, reducing the power output volatility. According to this research needs, this paper carried out the following work.Establish the equivalent model of the DFIG by using wind turbines operating data. Analysis of historical data of wind farms, and extract the wind speed and power data. Draw the wind power curve that can accurately reflect the actual condition of the units. Correct the power coefficient by using this curve, then build wind turbine mode. In combination with other parts of the model together to build the equivalent model of wind turbine unit.After receiving the dispatching signal,wind farm need to distribute the active power settings to every unit. By setting a conventional generator group and a dispatching response group, we can reduce the dimension of power distribution problem. Extract the feature vectors of actual operating data, then cluster the units by using fuzzy clustering method. The units are divided into different groups, different groups set different power distribution priority. Combined dispatching response program, distribute the power setting into these clusters with different priorities. This method can reduce simulation time and avoid local optimum in searching process.An active power distribution optimization objective function is proposed. This function can make volatile units running at a relatively low operating condition by setting different power distribution priority values. Negative impact of volatility can offset by the power margin, making the total output smooth and reliable. Simulate this optimized result through the equivalent model of wind turbine. The simulation result show that the wind farm can response to dispatching signals in time.