基于容量利用特征的行业售电量预测方法研究

为了使电力企业能够从用电根源把握行业整体的用电趋势，引入基于容量利用特征并且考虑了外界经济因素影响的电量预测方法。首先，通过对行业的运行容量进行分离，采用行业不同业扩报装类型生长曲线还原真正的存量容量。再提取行业利用小时数，应用相关系数法和K-L信息量法对影响行业利用特征的外在因素进行辨识，确定主导因素指标。最后，创建基于容量利用特征的电量预测方法，基于存量产生电量和业扩增量产生电量形成对行业的预测模型。实证分析表明，这种方法能有效地提高预测精度。

Research of industry electricity sales forecasting method based on capacity utilization characteristics

In order to enable power companies to grasp the trend of electricity consumption of the whole industry from the power source, this paper proposes a forecasting method of industry electricity sales based on capacity utilization characteristics and taking external economic factors. Firstly, it separates operation capacity of each industry and restores real memory capacity by using the growth curves of different types of business expansion. Then, it extracts industry capacity utilization hours, identifies the external factors which affect the industry utilization characters by correlation coefficient method and K-L information method to determine the dominant factor indexes. At last, it creates power prediction method based on capacity utilization characteristics. The prediction model of industry electricity sales is based on the generation of electricity generated from the stock and the industrial expansion. The results of the case study show that the prediction method can effectively improve the prediction accuracy.