| 引用本文: | 杨丹旭,宋亚奇,岳建任,李莉.基于投票方差的非侵入式负荷波动检测[J].电力自动化设备,2022,42(7): |
| YANG Danxu,SONG Yaqi,YUE Jianren,LI Li.Non-intrusive load fluctuation detection based on voting variance[J].Electric Power Automation Equipment,2022,42(7): |
|
| 摘要: |
| 针对现有事件检测方法在大功率用电设备开启后漏检小功率用电设备切换事件以及对事件发生时间定位不准确的问题，提出一种基于投票方差的波动检测方法，该方法通过投票机制定位事件，可提高检测精度。实验结果表明，该方法在聚合功率水平发生变化时仍然具有稳健性，可检测出的负荷波动范围为40~2000 W，通过对参数的选择，该方法的检测精度达到95 %左右，能够精准检测到开始时间的事件数占95.30%。 |
| 关键词: 非侵入式负荷监测 波动检测 漏检 时间定位 投票方差 |
| DOI：10.16081/j.epae.202205021 |
| 分类号:TM715 |
| 基金项目:国家自然科学基金资助项目(51677072)；中央高校基本科研业务费专项资金资助项目(2018MS074) |
|
| Non-intrusive load fluctuation detection based on voting variance |
|
YANG Danxu, SONG Yaqi, YUE Jianren, LI Li
|
|
School of Control and Computer Engineering, North China Electric Power University, Baoding 071000, China
|
| Abstract: |
| Aiming at the problems that the existing event detection methods miss the switching events of small power electrical equipment and inaccurately locate the occurrence time of events after turning on high power electrical equipment, a fluctuation detection method based on voting variance is proposed. This method locates the events through voting mechanism, which can improve the detection accuracy. Experimental results show that the proposed method is still robust when the aggregate power level changes, the load fluctuation range that can be detected is 40~2000 W, the detection accuracy of the proposed method reaches about 95 % through the selection of parameters, and the number of events whose start time can be accurately detected accounts for 95.30%. |
| Key words: non-intrusive load monitoring fluctuation detection missed detection time location voting variance |
