Wind turbine maximum power point tracking control based on unsupervised neural networks

Abstract The main control goal of a wind turbine (WT) is to produce the maximum energy in any operating region. When speed under its rated value, must aim at tracking power point best curve for specific WT. This challenging due non-linear characteristics system and environmental disturbances it subjected to. Direct (DSC) one techniques applied address this problem. In strategy, necessary design controller adjust generator torque so follow optimum speed. work, we improve DSC by implementing with radial basis function neural network (NN). An unsupervised learning algorithm designed tune weights NN learns law that minimizes error. With proposed methodology, electromagnetic allows optimal extraction obtained, thus coefficient (${C}_\mathrm{p}$) values. proposal tested on OpenFAST model National Renewable Energy Laboratory 1.5 MW Simulation results prove good performance neuro-control approach as maintains WT variables into appropriate range tracks operation It has been compared included giving up 7.87% more power.