电力中长期负荷模糊优选组合预测方法的研究

Research on Fuzzy Optimized Combination Method for Middle and Long Term Load Forecasting

传统中长期电力负荷组合预测方法在确定权系数时,由于没能充分考虑各单一预测模型拟合值与历史实际值之间的多种误差信息,而使预测结果不够理想。为此,选取多种误差作为评价单一模型优劣的指标,引入多目标系统模糊优选理论,结合熵权法客观分配各种误差指标的权重,在全面考虑多种误差信息的情况下,通过求取各单一模型预测精度对"优"的隶属度,确定组合预测模型的权系数。算例分析表明,该组合预测模型能够有效地提高预测精度。

Because the forecasting results and multiple estimation errors for single forecasting methods have not been considered deeply in traditional combination forecasting methods of middle and long term power load, the forecasting results can not meet the requirements. Therefore, muhiple errors are used to be the measuring index for single forecasting method and the multiple objectives fuzzy optimization theory is introduced to determine the combination method. The entropy weight method is used to distribute the weights of errors. One power system is used to validate the proposed method. The forecasting results and the comparison with the existing methods show that this method is more precise and more effective.