Short-Term Load Forecasting of Smart Grid Based on Load Spatial-Temporal Distribution

Various incentives in the smart grid, which force power consumer to take respond demand and change the characteristics of load, will affect power load forecasting. In this paper, it is concluded that the load spatial-temporal distribution and the load are relevant in the smart grid, and the impact of load spatial-temporal distribution is considered in forecasting model. It sifted multivariate load series highly relevant to prediction point with correlation analysis, and modeled the load series from the viewpoint of time with long short-term memory (LSTM). By comparing with the artificial neural network and support vector machine, it is verified that the model proposed has higher accuracy.