In this article, a noninvasive remote monitoring method of error of electricity energy meter (EM) is introduced and its achievable precision is investigated. The aim of the method is to determine remotely the smart EM active power relative measurement error using the reference EM connected at the input of a low-voltage distribution grid segment. The meter error estimation is carried out despite other customers’ electrical load changes observable by the reference meter and power fluctuations in the distribution network. The method is based on the real-time analysis of natural customer power consumption profiles without any additional load injection for error-monitoring needs. A power step in the customer power consumption profile is detected by the meter under test and then the step magnitude is delivered for the comparison with the power step magnitude estimated synchronously by the reference meter. The analysis of the method error is conducted by applying simulation of the benchmark distribution grid which includes 26 EMs and utilizing the open data power consumption profiles. It is shown that after two to three days of averaging (including approximately 170 power steps) of the meter error estimates, it becomes possible to achieve an uncertainty of meter error estimation approximately 0.63%. The feasibility of the method implementation scenarios with the state-of-the-art EM data collection protocols in advanced measurement infrastructure (AMI) by means of occupancy index modeling is performed.

中文翻译：

---

一种利用功耗曲线对电能表误差进行无创远程监控的方法

本文介绍了一种无创远程电能表误差监测方法，并对其可实现的精度进行了研究。该方法的目的是使用连接在低压配电网段输入端的参考 EM，远程确定智能 EM 有功功率相对测量误差。尽管参考电表可观察到其他客户的电力负载变化以及配电网络中的功率波动，但仍会执行电表误差估计。该方法基于对自然客户功耗曲线的实时分析，无需任何额外的负载注入来满足错误监控需求。被测仪表检测客户功耗曲线中的功率阶跃，然后传送阶跃幅度以与参考仪表同步估计的功率阶跃幅度进行比较。该方法误差的分析是通过应用包括 26 个 EM 的基准配电网的模拟并利用开放数据功耗曲线来进行的。结果表明，在对电表误差估计进行平均两到三天后（包括大约 170 个功率步长），可以获得大约 0.63% 的电表误差估计的不确定性。通过占用指数建模，在高级测量基础设施 (AMI) 中使用最先进的 EM 数据收集协议执行方法实施场景的可行性。