基于多级负荷智能协调的母线短期负荷预测

Short-term Bus Load Forecasting Based on Multi-level Load Intelligent Coordination

为提高母线负荷预测精度,提出一种基于多级负荷智能协调的母线短期负荷预测方法。首先对预测母线负荷序列进行历史负荷与当前负荷的相关性分析,再进行系统空间母线与预测母线的相关性分析,根据两次相关性分析结果合理设置算例,得到预测网络的最优输入方式,然后利用长短时记忆网络(LSTM)建立母线短期负荷预测模型,最后运用吉林省某地区的实测数据将提出模型与反向传播(BP)神经网络和支持向量机(SVM)的预测结果进行对比分析,验证本文提出的预测模型具有更高的精确度。

In order to improve the accuracy of bus load forecasting,a bus short-term load forecasting method based on multi-level load intelligent coordination is proposed.Firstly,the correlation between the historical load and the current load is analyzed for the bus load sequence,and then the correlation between the system space bus and the forecasting bus is analyzed.According to the results of the correlation analysis,the cases are set reasonably to obtain the optimal input mode of the prediction network.Then,the bus short-term load forecasting model is established by the long-term and short-term memory network.Finally,the model is compared with the prediction results of BP neural network and support vector machine by using the measured data of a certain area in Jilin Province,which verifies that the prediction model proposed in this paper has higher accuracy.