上海市全社会月用电量预测模型

Study on Prediction Model for Shanghai Monthly Power Consumption

较为准确的中短期用电量预测是制定科学合理电网运行计划的前提。利用2004—2014年上海市各月全社会用电量、气温、历年GDP、人口和单位GDP能耗等数据,采用年际同比变率分析方法,分析了影响月用电量的主要因子,建立了上海市逐月用电量预测模型。结果表明:采用年际同比变率分析方法能有效剔除气温、用电量等年际变化趋势对相关关系分析结果的影响,所得结论物理意义更为清晰;上海市月用电量呈现冬季和夏季双峰型特征,受经济总量增加等因素的影响,各月用电量都呈现出增加趋势;冬季和夏季的月用电量同比变化率与气温变化相关最为密切,春季和秋季的用电量同比变化率主要与经济总量增长和产业结构调整相关;综合考虑各影响因子建立的预测模型能够实现对各月用电量较精确的预测,也可用来研判经济走势。

Accurate mid- to short-term prediction of power consumption is the precondition for formulating a scientific power grid operation program. By investigating the series data of Shanghai City, including monthly power consumption （MPC）, air temperature （AT）, gross domestic product （GDP）, population and energy consumption per unit of GDP （ECUG）, the main influencing factors of MPC are obtained in this paper, and a MPC prediction model is developed through the Yearly Change Rate （YCR） method. The forecast results show that the YCR method can effectively eliminate the impact of background trend on correlation analysis results and helps to improve understanding the relationship between AT and MPC; the MPC of Shanghai has two summits of winter and summer and shows a tendency of increase as influenced by the increase of economic scale; The YCR of MPC in winter and summer is closely related to temperature variations, while the YCR of MPC in spring and autumn is mainly related to the economic growth and industrial structure adjustment; Taking into account of various influence factors, the MPC prediction model is suitable for power consumption forecast as well as economic situation analysis.