基于AHP的实用中长期电力负荷预测最优综合模型

针对中长期电力负荷预测特点，提出一种基于AHP（层次分析法）的中长期电力负荷预测最优综合模型。以历史数据的拟合精度、预测结果与未来经济社会发展的一致性、预测者对模型的信赖度为目标准则；通过自适应进化规划和判断矩陈的方法确定综合模型的最优权重。基于AHP的最优综合模型不令依靠科学计算，而且也考虑了专家经验，将基于AHP的最优综合模型的预测结果与传统综合模型的预测结果相比，误差较小，精度较高，在实用中取得了满意的结果。

Practical optimum combined power system mid-long term load forecast based on the Analytic Hierarchy Process

To improve power system load forecasting accuracy, an optimum combined method based on the Analytic Hierarchy Process was presented in this paper. Compared to current forecast method, the objective function proposed here are not only the highest fitting accuracy, but also concordance with future economic development and believability of the methods. The optimal combined weight vector is computed via self-adaptive evolutionary programming and judgment matrix. In terms of the comparison of the forecasting results between AHP-based optimum combined model and traditional combined model, the former bears smaller error and better accuracy. The practical example shows that this method can combine the advantages from many forecasting methods and experiences from experts. So it is practical, accurate and efficient.