短期尖峰负荷多信息融合的神经网络预测方法

Neural Network Prediction Methods of Short-Term Peak Load Based on Multi-Information Fusion

为了降低负荷数据的复杂度、提高预测精度,提出一种短期尖峰负荷多信息融合的神经网络模型。选取皮尔逊相关系数分析假日、温度、湿度等信息之间的密切程度,将关键气象信息融合进模型中,优化负荷的输入参数,重构神经网络模型的新数据集,并防止神经网络的过拟合,提高短期尖峰负荷预测精度。算例仿真分析表明,所提方法与未考虑多信息融合的单一增强型决策树模型和神经网络模型相比,更能有效地提高短期尖峰负荷预测准确率。

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.