基于LMD和模型匹配的家电负荷识别算法
发布时间：2018-07-29 14:15
【摘要】：家电负荷识别是智能用电的重要环节,传统侵入式负荷监测具有成本高、安装维护复杂的缺点,因此以非侵入式负荷监测为基础研究家电负荷识别算法。结合系统辨识的基本原理和方法,以稳态电流、稳态电压为特征,提出一种基于局部平均分解(LMD)和模型匹配的家电负荷识别算法。通过预先获取用电网络中各负荷的稳态数据,构建线性和非线性模型库。利用LMD算法将混合信号分解为单个负荷的用电数据,通过预筛选确定分离数据所属模型库,根据模型匹配原则进行负荷识别。仿真结果表明,所提算法可以准确识别出各负荷的运行状态,运算效率高,并能有效应对用电网络中有新负荷加入的情况。
[Abstract]:Load identification of household appliances is an important part of intelligent power consumption. Traditional invasive load monitoring has the disadvantages of high cost and complex installation and maintenance. Therefore, based on non-invasive load monitoring, the load identification algorithm of household appliances is studied. Combined with the basic principles and methods of system identification, a load identification algorithm based on local average decomposition (LMD) and model matching for household appliances is proposed, which is characterized by steady current and steady voltage. The linear and nonlinear model libraries are constructed by obtaining the steady state data of each load in the power network in advance. The mixed signal is decomposed into the electric data of a single load by LMD algorithm. The model base of the separated data is determined by pre-screening, and the load identification is carried out according to the principle of model matching. The simulation results show that the proposed algorithm can accurately identify the running state of each load, has high computational efficiency, and can effectively deal with the situation where new loads are added to the power network.
【作者单位】： 华北电力大学电气与电子工程学院;国网物资有限公司;
【基金】：中央高校基本科研业务费专项资金资助项目(2016MS13)~~
【分类号】：TM714
本文编号：2152940
[Abstract]:Load identification of household appliances is an important part of intelligent power consumption. Traditional invasive load monitoring has the disadvantages of high cost and complex installation and maintenance. Therefore, based on non-invasive load monitoring, the load identification algorithm of household appliances is studied. Combined with the basic principles and methods of system identification, a load identification algorithm based on local average decomposition (LMD) and model matching for household appliances is proposed, which is characterized by steady current and steady voltage. The linear and nonlinear model libraries are constructed by obtaining the steady state data of each load in the power network in advance. The mixed signal is decomposed into the electric data of a single load by LMD algorithm. The model base of the separated data is determined by pre-screening, and the load identification is carried out according to the principle of model matching. The simulation results show that the proposed algorithm can accurately identify the running state of each load, has high computational efficiency, and can effectively deal with the situation where new loads are added to the power network.
【作者单位】： 华北电力大学电气与电子工程学院;国网物资有限公司;
【基金】：中央高校基本科研业务费专项资金资助项目(2016MS13)~~
【分类号】：TM714
【相似文献】
相关期刊论文 前1条
1 项华珍,代冀阳;H∞控制及其应用[J];南昌航空工业学院学报;1996年01期
，本文编号：2152940
本文链接：https://www.wllwen.com/kejilunwen/dianlidianqilunwen/2152940.html
