Combining LIDAR and LADRC for intelligent pitch control of wind turbines

At present, most of the pitch control methods are based on PI controller, the pitch control system has poor disturbance resistance, and the research of variable parameter feedforward based on Light detection and ranging (LIDAR) and the Linear Active Disturbance Rejection controller (LADRC) composite control is rarely studied to reduce the blade root load, so this paper conceives a hybrid intelligent and adaptive pitch control approach to reduce a wind turbine generator speed fluctuation and its blade root load. Specifically, we combine the Radial Basis Neural Network and Finite Impulse Response filter (RBFNNFIR) based on LIDAR wind measurement. We then use a variable bandwidth of LADRC controller. Overall the approach enables and facilitates self-adaption and self-adjustment. We use Matlab s-function to call the multi-freedom mathematical wind turbine model based on FAST code, the composite intelligent control algorithm is established in Simulink. Initial results from the statistical analysis of the experiments under different turbulent wind conditions shows that the hybrid intelligent pitch control approach can reduce the generator speed fluctuation by about 40.8%, and the blade root max value of load moment by about 13.1%, compared with the baseline values of the traditional variable gain PI control algorithm. (C) 2021 The Authors. Published by Elsevier Ltd.

Automated summary

This study proposes a hybrid intelligent and adaptive pitch control approach to reduce wind turbine generator speed fluctuation and blade root load. By combining RBFNNFIR based on LIDAR wind measurement with a variable bandwidth of LADRC controller, the approach enables self-adaptation and self-adjustment. Initial results show that this hybrid approach can significantly reduce generator speed fluctuation and blade root load compared to traditional PI control algorithm.