Volume 36: Intelligent Energy Solutions for Resilient Urban Systems

Research on Optimization Method of Short-Term Load Forecasting Model Based on CNN-LSTM Xueyuan Zhao, Xiaoyu Ying, Tingting Xu, Yang Tan

Abstract

Accurate power load forecasting can significantly reduce the operating costs of the power grid and is an important guarantee for the stable and efficient operation of the power system. However, the randomness and volatility of short-term power loads are strong, and traditional load forecasting methods are difficult to grasp the patterns of short-term load changes. In order to predict short-term power load more accurately, this paper proposes a short-term power load prediction method based on convolutional neural networks and short-term memory networks (CNN- LSTM), and combines down-sampling processing and time features to extract features from the dataset. The prediction results are compared to improve prediction accuracy. By comparing and analyzing the prediction accuracy of the model based on measured data of public buildings, the reliability of the proposed model was verified, and it was confirmed that its application effect in the field of short-term power load forecasting is good, which can provide theoretical basis and technical support for power planning in the power supply department.

Keywords load forecasting, long-short-term memory, convolutional neural network, optimization method