Open AccessPitch Control of Wind Turbines Based on BP Neural Network PI
Author(s) -
Jing Du,
Bo Wang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1678/1/012060
Subject(s) - gain scheduling , control theory (sociology) , turbine , artificial neural network , wind power , pitch control , pid controller , matlab , computer science , controller (irrigation) , scheduling (production processes) , wind speed , control engineering , control (management) , engineering , temperature control , physics , electrical engineering , aerospace engineering , artificial intelligence , agronomy , operations management , meteorology , biology , operating system
For wind turbine operating above the rated wind speed, the output power of the generator is maintained near the rated power through pitch control. Due to the non-linear relationship between the pitch angle and the wind speed, the traditional PI controller is not ideal for control above the rated wind speed. Therefore, the currently used PI controller generally incorporates gain scheduling technology. A wind turbine pitch controller based on BP neural network PI is proposed in this paper, which optimizes PI parameters and has better control effects. First, the PI control strategy with gain scheduling is briefly introduced, and then the principle and implementation steps of BP neural network PI are given. Finally, compared the control effects of the two control strategies of PI with gain scheduling and BP neural network PI, a new co-simulation between Simpack and MATLAB/Simulink is built, and it is proved that the BP neural network PI can improve the effect of pitch control.