引用本文: | 高浩瀚,张利,梁军,杜涛,包圣.基于改进排列熵算法和Yamamoto算法的非侵入式用电设备状态变化检测[J].电力自动化设备,2020,40(1): |
| GAO Haohan,ZHANG Li,LIANG Jun,DU Tao,BAO Sheng.Non-intrusive electrical equipment state change detection based on improved permutation entropy algorithm and Yamamoto algorithm[J].Electric Power Automation Equipment,2020,40(1): |
|
摘要: |
非侵入式负荷监测已经成为智能电网负荷监测管理的关键技术之一。用电设备状态变化过程所表现出的暂态特征是进行非侵入式用电设备状态辨识的重要依据,但其精准提取取决于用电设备状态变化的准确检测。为此,提出了一种基于改进排列熵算法和Yamamoto算法的非侵入式用电设备状态变化检测算法。首先对排列熵算法进行多尺度改进,利用多尺度排列熵的差值分析确定状态变化发生的区间,然后利用Yamamoto算法进行区间检测,定位状态变化的时刻。仿真分析结果表明,所提算法可准确检测用电设备的状态变化,有效地提高后续利用暂态特征的设备状态辨识准确率。 |
关键词: 非侵入式负荷监测 用电设备 暂态特征 排列熵算法 Yamamoto算法 状态辨识 |
DOI:10.16081/j.epae.201911017 |
分类号:TM73 |
基金项目:国家自然科学基金资助项目(51477091) |
|
Non-intrusive electrical equipment state change detection based on improved permutation entropy algorithm and Yamamoto algorithm |
GAO Haohan1, ZHANG Li1, LIANG Jun1, DU Tao2, BAO Sheng1
|
1.Key Laboratory of Power System Intelligent Dispatch and Control of Ministry of Education, Shandong University, Jinan 250061, China;2.Shandong Shanda Electric Power Technology Co.,Ltd.,Jinan 250101, China
|
Abstract: |
NILM(Non-Intrusive Load Monitoring) has become one of the key technologies of load monitoring and management in smart grid. The transient characteristics in the state change process of electrical equipment are an important basis for state identification of non-intrusive electrical equipment, but its accurate extraction depends on the accurate detection of electrical equipment state change. Therefore, a non-intrusive electrical equipment state change detection algorithm based on improved permutation entropy algorithm and Yamamoto algorithm is proposed. Firstly, the permutation entropy algorithm is improved on multiple scales, whose difference analysis is used to determine the interval of state change. Then, Yamamoto algorithm is used to detect the interval and locate the state change time. Simulative results show that the proposed algorithm can accurately detect the state changes of electrical equipment, and effectively improve the identification accuracy of equipment state based on transient characteristics. |
Key words: non-intrusive load monitoring electrical equipment transient characteristics permutation entropy algorithm Yamamoto algorithm state identification |