三相电能表计量误差自动检定方法设计

Design of Three-Phase Energy MeterMeasurement Error Atuomatic Calibration Method

为了提升三相电能表计量误差检定方法的有效性、避免冗余数据给检定过程带来的干扰,提出基于聚类优化的三相电能表计量误差自动检定方法。利用模糊C-均值(FCM)聚类算法提取三相电能表检定数据的核心特征,并构建检定数据核心特征数据集。利用核函数划分原始的特征空间,以映射到更高维的希尔博特空间。建立支持向量机(SVM)模型,并将核心特征数据集输入该模型,以较少数据实现对三相电能表计量误差的自动检定,从而提高检定的精准性。试验结果表明:该方法的检定时间较少,均在2.5 s以下;三相电能表计量误差检出率均达到90%以上,能够有效提升检定效率。该方法在计量误差检定中具有重要作用。

To improve the effectiveness of the three-phase energy meter measurement error calibration method and avoid the interference of redundant data to the calibration process, an three-phase energy meter measurement error automatic calibration method based on clustering optimization is proposed.Fuzzy C-mean (FCM) clustering algorithm is utilized to extract the core features of the calibration data of three-phase energy meters and construct the core feature data set of the calibration data.The kernel function is utilized to divide the original feature space, which is mapped to the higher dimensional Hilbert space.A support vector machine (SVM) model is established, and the core feature data set is input into the model to realize the automatic calibration of three-phase energy meter measurement error with less data, to improve the accuracy of calibration.The test results show that the calibration time of this method is lower, which is all below 2.5 s;the detection rate of three-phase energy meter measurement error all reaches more than 90%, which can effectively improve the calibration efficiency.The method has an important role in the measurement error verification.