基于加权余弦相似度与极限学习机的电力负荷短期预测

为提高电力系统管理的效率,提出一种基于加权余弦相似度与极限学习机(extreme tearning machine, ELM)的电力负荷短期预测设计。通过熵权法对电力负荷相关物理信息进行权重分配,获得的权重赋予到余弦相似度中,利用加权余弦相似度对历史日与待测日的电负荷数据进行相似度选取,筛选数据作为极限学习机的输入,提高极限学习机回归模型的精度,最终获取电力负荷预测。实验分析与反向传播BP(back propagation)神经网络、支持向量机(spupport vector machine, SVM)预测算法对比,该方法能有效提高预测模型的精度,同时简化计算量。

Short-term Power Load Forecasting Based on Weighted Cosine Similarity and Extreme Learning Machine

To improve the efficiency of power system management, a short-term load forecasting design based on weighted cosine similarity and extreme learning machine (ELM) is proposed in this paper. The weights obtained are assigned to the cosine similarity by the method of entropy weight. The weighted cosine similarity is used to select the similarity between the electric load data of the historical day and the day to be measured. The data are selected as the input of the extreme learning machine to improve the precision of the regression model of the extreme learning machine. Finally, power load forecasting is obtained. Compared with back propagation (BP) neural network and support vector machine (SVM) prediction algorithm, the experimental results show that this method can effectively improve the accuracy of prediction model and simplify the calculation..