遗传算法加权指数灰色模型的应用  

Application of Genetic Algorithm Weighted Exponential Grey Model

作　　者：陈坚强 詹棠森 CHEN Jianqiang;ZHAN Tangsen(School of Information Engineering,Jingdezhen Ceramic University,333403,Jingdezhen,Jiangxi,PRC)

机构地区：[1]景德镇陶瓷大学信息工程学院,江西景德镇333403

出　　处：《江西科学》2024年第5期924-928,1033,共6页Jiangxi Science

基　　金：国家自然科学基金项目(71763013);江西省教育厅学位办项目(JXYJG-2023-160)。

摘　　要：由于传统的灰色模型和指数平滑模型在用电量预测方面精度欠佳,为了对江西省全社会用电量进行短期预测,从统计年鉴官网获得2000—2022年江西省全社会用电量数据,然后将灰色模型与指数平滑模型采用遗传算法决定加权组合,得到加权指数平滑灰色模型。利用加权指数平滑灰色模型对该数据进行模型构建,最后在验证模型有效性的基础上进行未来用电量预测。结果显示,所提出的方法相对于其他单项模型和传统的组合模型而言,预测得到的结果更加准确,平均相对误差更小,具有较高的预测精度,为用电量预测提供更可靠的参考,对用电量管理和规划具有重要意义。Due to the poor accuracy of the traditional gray model and exponential smoothing model in predicting electricity consumption,a short-term forecast of the total electricity consumption in Jiangxi province was conducted.Data on the electricity consumption of the whole society in Jiangxi province from 2000 to 2022 were obtained from the official statistical yearbook website,and then the gray model and the exponential smoothing model were weighted by genetic algorithm to determine the weighted combination,resulting in the weighted exponential smoothing gray model.The results show that compared with other single models and traditional combination models,the proposed method has more accurate prediction results with a smaller average relative error and higher prediction accuracy.This method provides a more reliable reference for electricity consumption prediction and is of great significance for electricity consumption management and planning.

关 键 词：灰色模型 指数平滑模型 加权组合 遗传算法 用电量 

分 类 号：F206[经济管理—国民经济]