京津唐电网电力日峰荷与气象指数的关联性分析

采用4个气象指数来量化温度、湿度和风速等气象因子对负荷的综合影响。首先通过小波去噪消除负荷和气象数据中的伪数据和噪声，然后采用灰色关联理论分析了负荷与各气象指数之间的关联关系，选择其中关联性最好的气象指数和负荷进行敏感度分析，并采用支持向量机算法对电力日峰荷进行预测。最后介绍了在此基础上开发的京津唐电网电力日峰荷与气象指数关联性分析系统所采用的关键技术。

Relevance Analysis of Meteorological Index and Peak Load in Tianjin-Beijing-Tangshan Power Grid

Four meteorological indices are adopted to quantities the comprehensive impacts of meteorological factors such as temperature, humidity and wind speed on power loads. Firstly, by means of wavelet de-noising the pseudo data and noises in both load and meteorological data are eliminated; then by use of gray correlation theory the incidence relation among power load and meteorological indices is analyzed; and then the meteorological indices and load which possess best relevance one another are chosen to perform sensitivity analysis, and using support vector machine algorithm the daily peak load is forecasted. Finally, the key techniques, used in the relevancy analysis system of daily peak load and meteorological indices for Beijing-Tianjin-Tangshan Power Grid, which is developed on the basis of above-mentioned contents, are presented.