An improved fusion-weighted LSTM network for power equipment condition predict

Aiming at the problems of random lack of data and complex changes of many influencing factors, this paper establishes a monitoring index prediction method based on improved LSTM network, and uses LSTM network combined with differential integrated moving average autoregressive model to analyze the main sample data, introduces auxiliary information, and quantifies the impact of auxiliary information on monitoring indicators through combinatorial empowerment. After case verification, the established prediction model can effectively improve the prediction accuracy of power load and other monitoring parameters, and effectively improve the problem of random lack of data.