Identification and Correction Method of Bad Data of Renewable Energy Plants with Deep Learning

Abstract

Given the problem of real-time data acquisition errors in renewable energy plants, the data of the renewable energy plants have the characteristics of mass and mutually coupled characteristics, a deep learning-based method for identifying and correcting bad data from renewable energy plants is proposed. Firstly, a deep neural network identification model is constructed to identify the real-time bad data, and the bad data of the real-time identification was obtained. Secondly, the BP neural network correction model was constructed to correct the bad data of the identification, and the reliable data of the operation of the renewable energy station is obtained. Finally, the accuracy and effectiveness of the proposed method are verified through the analysis of the real historical data of a typical wind farm.