Abnormal wind speed date recognition based on convolution neural network

Abstract:
Wind speed data of wind power generation is of great value in the era of big data. Accurate identification of wind speed is very important to evaluate the operation status and power prediction of the wind farm and the formulation of a dispatching plan by the power dispatching department. In this paper, a method for identifying abnormal wind speed data of wind power generation is proposed. Taking the customized data collected every day as samples, the convolutional neural network is trained. The experimental results show that the convolutional neural network can preliminarily distinguish normal samples from different types of abnormal samples. This paper analyzes the influencing factors of the experimental process and results, which provides a reference for the identification of abnormal data in the power system.