Individual pitch control based on radial basis function neural network

Abstract

As the increasing of structure in wind turbine (WT), the flapping vibration force of blade is more and more serious, and the output power will be unstable in operation of the merged power networks. In this paper, to improve the WT dynamic performance in running processes, by analyzing the WT aerodynamics, wind shear, and tower shadow effect, we have designed based on radial basis function neural network (RBFNN) control strategy for individual pitch control (IPC), using RBFNN approach pitch control system unknown nonlinear functions, and introduced into the adaptive law online adjustment the system error, to improve the dynamic performance of pitch control system and alleviate structure of fatigue loads. Finally, the results show that based on RBFNN for IPC produces adaptability dynamic performance. It can effectively improve power quality to reduce fatigue load in key components of WT.