供电计量电表故障自动优化检测仿真

Automatic Optimization Detection Simulation of Power Meter

对供电计量电表故障的自动检测,能够有效提升供电计量电表工作稳定性。对电表故障的检测,需要获得各个独立元的监控量均值,计算电表使用过程各个指标数据的变化特征,完成对供电计量电表故障的自动检测。传统方法先提取电表故障特征数据,给出故障判别阈值,但忽略了计算电表指标数据的变化特征,导致检测精度偏低。提出基于多变量轨迹分析的供电计量电表故障自动检测方法。上述方法先采用KICA从正常电表测量数据中提取独立元,用GMM拟合高精度电表数据各独立元的概率密度函数,建立基于GMM的监控量和其控制限,计算各个独立元的监控量均值,计算电表使用过程各个指标数据的变化特征,定义指标数据关键变量的轨迹趋势图,并完成对供电计量电表故障自动检测。仿真证明,所提方法故障检测精度较高,提升了对供电计量电表的设计质量。

Traditional detection method of ammeter fault always neglects the calculation of change feature of am- meter index data, and the detection precision is low. In order to solve the problem, based on multi-variable trajectory analysis, an automatic detection method for ammeter fault of power supply measure is proposed in this article. KICA was used to extract independent element from measurement data of normal ammeter in the research and GMM was used to fit the probability density function of each independent element of ammeter data with high precision. Monito- ring amount and control limit were built. The mean monitoring amount of each independent element and the change feature of data of each index during process of using ammeter were calculated respectively. The track tendency chart of key variable of index data was defined. Simulation results prove that the method has higher detection precision and can improve the design quality of ammeter.