Wind power is renewable and spotless energy with progress projections. By reason of the effect of randomness, unpredictability of wind quickness, the wind turbine features of movable pitch control system with nonlinear robust coupling, time-varying strictures, it is problematic to acquire the precise mathematical model; and with the intention of comprehend the continuous power output, operating district of the fast, high-precision control dynamic act, this paper planned the AC servo fan adaptable pitch system based on BP network adaptive control PID. Over the simulation, to confirm this system attained the control result that with high steady-state correctness and was rapid to follow control result. Background/Objectives: Wind power generation is to convert wind energy into electrical energy, renewable energy, and clean energy with advance prospects. Due to the impact of randomness, instability and the aerodynamic effect of wind speed, the wind turbine characteristics of variable pitch control system with nonlinear, strong coupling, time-varying parameters, it is difficult to obtain the accurate mathematical model. Results: Self-tuning PID control scheme based on BP neural network made the output state of BP neural network correspond to proportional kp, integral ki, and differential kd of PID. The PID control factors were optimized correction online with the help of neural network's self-learning ability and the ability to approximate any nonlinear function, then weight coefficients of neural network was adjusted, the PID control coefficient under some ideal control rules was achieved when the system in stable states (see the simulation curves of b).
Read full abstract