摘要
为了提高电力消费预测结果,提出了基于对数均值Divisia指数—改进人工蜂群(LMDI-IABC)的电力消费预测模型。采用对数均值Divisia指数(LMDI)算法将电力消费数据分解为7种影响因素,并求取各个影响因素的贡献率。为了获得准确的电力消费预测结果,提出了一种改进的人工蜂群(IABC)算法,所提算法采用混沌映射初始化种群,采用等距分布式并行搜索,并且引入势场更新蜜蜂位置,从而提高蜂群算法的收敛速度和全局寻优能力。采用6种函数进行仿真对比,对比结果验证所提的IABC具有更快的收敛速度和更高的寻优精度。建立LMDI-IABC的预测模型,将LMDI得到的7种影响因素作为IABC的输入,输出为电力消费值。仿真对比结果验证了所提方法适用于电力消费预测。
In order to improve the power consumption forecasting results,a power consumption forecasting model based on logarithmic mean Divisia index-improved artificial bee colony(LMDI-IABC)is proposed.The power consumption data is decomposed into seven influencing factors by logarithmic mean Divisia index(LMDI)algorithm,and the contribution rate of each influencing factor is calculated.In order to obtain accurate power consumption prediction results,an improved artificial bee colony(IABC)algorithm is proposed,which uses chaotic mapping to initialize the population,uses isometric distribution parallel search,and introduces potential field to update the position of bees,so as to improve the convergence speed and global optimization ability of bee colony algorithm.Six kinds of functions are used for simulation and comparison,and the comparison results verify that the proposed IABC has faster convergence speed and higher optimization accuracy.The forecasting model of LMDI-IABC is established,and the 7 influencing factors obtained by LMDI are used as the input of IABC,and the output is the power consumption value.The simulation results show that the proposed method is suitable for power consumption forecasting.
作者
田云
李铁良
张炜
马赛
金如月
TIAN Yun;LI Tieliang;ZHANG Wei;MA Sai;JIN Ruyue(State Grid Hebei Anping County Electric Power Supply Company,HengShui 053600,China;State Grid Hebei Hengshui Electric Power Supply Company HengShui 053600,China)
出处
《微型电脑应用》
2025年第8期194-198,202,共6页
Microcomputer Applications
