短期负荷预测中气象因素处理的费歇信息方法

在能提供大量实时负荷数据和气象数据的智能电网大数据环境下,挖掘合适的气象因素处理方法对提高短期负荷预测精度尤为重要。针对一个或多个气象变量,解决一维或多维费歇信息计算问题。在此基础上,提出基于费歇信息的气象因素建模方法及新预测模型。实际测试结果表明:采用所提模型可以获得更精确的预测结果,解决了短期负荷预测中对气象因素处理的主观随意性问题。

Fisher information method for processing weather factors in short-term load forecasting

In the big data environment of smart grid that can afford massive real-time load data and weather data, it is especially important to find appropriate weather factor processing method for improving the accuracy of short-term load forecasting. Aiming at one or multiple weather variables, the problems of one or multi-dimension Fisher information calculation are solved. On this basis, the modeling method of weather factors based on Fisher information and the new forecasting model are proposed. The practical test results show that the proposed model can obtain more accurate forecasting results, which solves the subjective randomness problem in dealing with weather factors of short-term load forecasting.