关键词: 尖峰负荷, 多信息融合, 神经网络模型, 皮尔逊相关系数

Abstract: In order to reduce the complexity of load data and improve prediction accuracy, an neural network model based on multi information fusion for short-term peak load is proposed on this article. The Pearson correlation coefficient is selected to analyze the closeness between weather information such as holidays, temperature, and humidity. In proposed model considering the key weather information fusion, the input load parameters is optimized,a new dataset of the neural network model is reconstructed, and overfitting of the neural networkis avoided,and the accuracy of short-term peak load prediction is improved. A simulation example of peak load forecasting proves that the proposed method is more effective in improving the accuracy of short-term peak load forecasting compared to the single enhanced decision tree model and neural network model that do not consider multiple information fusion.

Key words: peak load, multi-information fusion, neural network model, Pearson correlation coefficient