Operation State Assessment and Prediction of Distribution Transformer Based on Data Driven

Abstract

With the rapid development of Power Internet of Things, power grid monitoring data and analysis methods are increasing, so real-time dynamic monitoring of power equipment becomes possible. This paper presents a data driven method for evaluation and trend prediction of distribution transformer operation state. The key features reflecting dynamic change of operation state are extracted from voltage and current data of distribution transformer, and characteristic data flow is input into dynamic evaluation model to make real-time portrait description of distribution transformer operation state. According to time order and change trend of characteristic data flow, Long Short-Term Memory network (LSTM) is used to analysis regulation of characteristic data, and Support Vector Regression model (SVR) for its prediction. The future characteristic data flow is obtained, which is input into the dynamic evaluation model to realize the future operation trend prediction of the distribution transformer. Finally, examples are given to illustrate the feasibility, advanced nature and applicability of the method.