Wind Turbine Unit Power Prediction Based on Wavelet Neural Network Optimized by Brain Storm Optimization Algorithm

Abstract

The construction of the wind power curve is of great significance to the wind turbines. Based on the accurate model of wind power curve developed, it can be employed for the wind power prediction and fault diagnosis. Normally, the wind turbine manufacturer provides the standard wind power curve, which is measured at standard conditions. However, the actual situation of the wind turbine is different from the standard state and is constantly changing. The wind power curve needs to be modified. The wind power curve essentially establishes a functional relationship between wind speed and active power. The neural networks have the ability to approximate function. In this paper, based on the actual data from a wind farm in Shanxi Province, the wavelet neural network is used to model the wind power curve, and the initial parameters are determined by using the brain storm optimization algorithm. The probability of the non-convergence in the learning process of the wavelet neural network is greatly reduced. Extensive experimental results are presented to validate the effectiveness of the proposed approach.