Integrating SVM Classifier and Distribution State Estimation for Detection and Identification of AMI Customer´s Meter Data

In this paper, inaccurate meter data estimation results obtained from support vector machine (SVM) classifier and distribution state estimation(DSE) procedures are presented. The proposed technique makes use of the actual measurements (AM) data available from distribution automation (DA) and accurate AMI meter data to adjust priori missing accuracy of meter data considered as pseudo measurements (PM). Through observations from fully deployed AMI smart metering devices, preliminary results demonstrate that the proposed technique could lead to the DSE solution that matches AM and provides accurate estimates for missing accuracy of AMI meter data.