文章摘要
韩笑,邓春宇,张玉天,等.基于高频数据V-I特性的延时反馈负荷在线快速辨识算法[J].电力系统自动化. DOI: 10.7500/AEPS20180415003.
HAN Xiao,DENG Chunyu,ZHANG Yutian, et al.A Fast Online Load Identification Algorithm based on V-I Characteristics of High Frequency Data and Delayed Feedback[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20180415003.
基于高频数据V-I特性的延时反馈负荷在线快速辨识算法
A Fast Online Load Identification Algorithm based on V-I Characteristics of High Frequency Data and Delayed Feedback
DOI:10.7500/AEPS20180415003
关键词: 非侵入负荷监测  在线识别  负荷容感性  稳态电流
KeyWords: non-intrusive load monitoring  online identification  load capacitance and electricity sensibility  steady-state current
上网日期:2018-11-30
基金项目:中央高校基本科研业务费专项资金资助项目(2018MS001);中国电力科学研究院创新基金(AI83-18-001)
作者单位E-mail
韩笑 华北电力大学 电气与电子工程学院 北京 昌平区 102206 hxyu33@sina.com 
邓春宇 中国电力科学研究院 dengchunyu@epri.sgcc.com.cn 
张玉天 中国电力科学研究院 zhangyutian@epri.sgcc.com.cn 
张瑜 中国电力科学研究院 bingdianlanxin@126.com 
武昕 华北电力大学电气与电子工程学院 19861002wuxin@163.com 
摘要:
      文中研究了一种基于V-I特性的延时反馈非侵入负荷在线快速辨识算法,该算法根据用电设备负荷容感性不变原理,在相同电压背景下,提取每次暂态发生前电路中的稳态周期电流,利用信号一维加减得到上一次投切的用电设备稳定运行时的周期电流,结合居民用户的用电设备操作习惯对目标函数施加约束,缩小可能进行投切的用电设备的组合范围,优化求解确定用电网络中的负荷状态。此外,引入延时反馈识别投切负荷避免负荷暂态过程对稳态特征提取的影响。利用公开数据集对该方法的有效性进行验证,通过延迟负荷识别可在短时间内即准确高效的判断各用电设备的启停时刻。
Abstract:
      A fast online fast identification algorithm based on V-I characteristics is studied in this paper. According to the principle of constant load capacitance and electricity sensibility of the electrical equipment, the steady periodic current before every transient is extracted under the same voltage background. And the periodic current of the last electrical equipment is obtained by the one dimension addition and subtraction of the signal. Then, combining the operating habits of the residential users to impose constraints on the target function, the combination range of the electrical devices that may be switched can be reduced. Finally, the load status in the power network is determined by optimization. Because this paper realizes the identification based on the load steady-state signal, the effect of the load transient process on the steady state feature extraction can be avoided by the delay identification of the switching load. The public data sets are used to verify the effectiveness of this method, and the start /stop time can be judged accurately and efficiently in a short time by load delay identification.
查看全文   查看/发表评论  下载PDF阅读器