基于人工免疫算法的电力负荷预测综合模型  

Comprehensive model of power load forecasting based on artificial immune algorithm

作　　者：王大成[1] 高赐威[1] 程浩忠[1] 奚珣[2] 朱杰[2] 张涌[2] 

机构地区：[1]上海交通大学电气工程系,上海200030 [2]上海市电力公司浦东供电分公司,上海200122

出　　处：《华东电力》2004年第3期15-18,共4页East China Electric Power

基　　金：国家自然科学基金(No.50177017);高等学校优秀青年教师教学科研奖励计划资助

摘　　要：将电力负荷预测综合模型中各单一电力负荷预测模型的最优权重作为抗原,将权重的解作为抗体,通过模拟生物免疫系统的工作原理来搜索最优权重,提出了基于人工免疫算法的电力负荷预测综合模型。应用该模型于某地区负荷预测的实例中,并与基于遗传算法的综合模型和基于直接搜索寻优法的综合模型进行比较、分析。结果表明,该模型具有全局寻优能力好、预测精度高等优点。Taking the optimal weighting of every individual power load forecasting model in the comprehensive model of power load forecasting as antigen and taking the solution of the weighting as antibody, the comprehensive model of power load forecasting based on artificial immune algorithm was put forward through simulating the working principle of biological immune system to search optimal weighting. The model was applied in load forecasting of a certain area and was compared with two other comprehensive models based on genetic algorithm and direct search method respectively. The result showed that this model based on artificial immune algorithm has the advantages of excellent capability of optimal weighting searching and high accuracy of forecasting.

关 键 词：电力负荷预测 综合模型 人工免疫算法 目标函数 电力系统 

分 类 号：TM715[电气工程—电力系统及自动化]