Intelligent Control Applied For Constant Power Control In Wind Power System

The technology of control is the key technology in wind power system. When the wind speed exceeds rated wind speed, because of the limit of the generator mechanical and electrical properties, the energy captured by the wind turbine must be reduced to maintain the output power at rated data. While reducing the load and impact on wind turbine blades, the unit can run safely.PI control is widely used in industrial control process. The issue of setting PI parameter is discussed in this paper. The wind speed has the very strong randomness, the wind electrical machinery's nonlinear response is also very strong. Different PI parameters are needed for different wind speed to make the operation of unit stable and efficient. Using the method of combining of genetic algorithm and neural network, the controller that can give the accurate reference value of PI parameters with the changed of wind speed is designed. The simulation results show the feasibility of the controller.However, PI controller adjusts the system when there is error between the system's output and the rated output. The controller may have some lag. There may be a large overshoot and longer settling time, when the wind speed changes. In this paper, an adaptive neuro-fuzzy feedforward controller is used. A simple feedforward controller is designed by combining neural network and fuzzy control method. The neural network with the characteristics of self-learning can adjust the parameters of the fuzzy controller. The controller can overcome the interference of non-linear. Timely adjustment of pitch angle with the change of wind speed, the response speed of the constant power control system is improved.The simulation results show that the feedforward controller used in conjunction with PI controller can not only guarantees the stability of output power,but also optimize the output power curve,when the wind speed above the rated data.Finally, with the theory above, QT is used to realization of software. Because of the features of cross-platform, the constant power control software designed with QT can be used in different platforms. On this software,the control effect of the above theory can effectively achieve the desired at various experimental wind speed.