基于多目标蛇优化算法的非侵入式负荷监测研究  

Research on Non-Invasive Load Monitoring Based on Multi-Objective Snake Optimization Algorithm

作　　者：肖月 杨超[1] 农宝翔 刘康 王飘 

机构地区：[1]贵州大学电气工程学院,贵州 贵阳 [2]国网安徽桐城市供电公司,安徽 桐城 [3]多彩贵州航空有限公司,贵州 贵阳

出　　处：《建模与仿真》2024年第1期194-203,共10页Modeling and Simulation

摘　　要：随着经济社会的发展和节能环保的要求,负荷监测已成为一个研究重点,安装简单、经济安全的非侵入式负荷监测(Non-Intrusive Load Monitoring, NILM)更是成为近年研究的热门领域。该文针对NILM研究中存在的负荷分解准确率不高及实际应用所需时间较长的问题,通过将有功功率与稳态电流作为识别特征,引入了由Fatma A. Hashim和Abdelazim G. Hussien于2022年提出的多目标蛇优化算法(Multiple Objective Snake Optimizer, MOSO)并建立数学模型,经过选取家中最常见的电器进行实验测量并分析,得出该方法有效提升了负荷分解的准确率并大大缩减了实验时间的结论。通过与不同算法在同一数据上进行实验分析并对比实验结果,验证了该文算法在准确率及实验效率上有明显提升,证明了该文算法具有优越性。With the development of the economy and society and the requirements of energy conservation and environmental protection, load monitoring has become a research focus, and non-intrusive load monitoring (NILM) that is simple to install, economical and safe has become a hot field in recent re-search. This article addresses the issues of low accuracy in load decomposition and long practical application time in traditional non-invasive load monitoring algorithms in NILM research. By using active power and steady-state current as identification features, the Multi-Objective Snake Opti-mizer (MOSO) algorithm proposed by Fatma A. Hashim and Abdelazim G. Hussien in 2022 is intro-duced and a mathematical model is established. After selecting the most common electrical appli-ances in the home for experimental measurement and analysis, it was concluded that this method effectively improves the accuracy of load decomposition and greatly reduces experimental time. By conducting experimental analysis on the same data with different algorithms and comparing the experimental results, it was verified that the proposed algorithm has significant improvements in accuracy and experimental efficiency, proving its superiority.

关 键 词：非侵入式负荷监测 多目标优化算法 蛇优化算法 遗传算法 负荷监测 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]