Real Time Evaluation Algorithm for Measurement Performance of Substation Voltage Transformer Based on Artificial Neural Network

This paper introduces an online multidimensional evaluation algorithm based on Artificial Neural Network (ANN) for substation voltage transformers and implement it. Voltage transformers are widely used in voltage transformers for voltage measurement of 220kV and above. This type of voltage transformer has low error stability compared with electromagnetic transformers. For a variety of reasons, a large number of measuring transformers in the power grid operate over the period of over-test. In this paper, we design a multi-dimensional online evaluation algorithm for transformer performance based on ANN. The observation space is formed by the transformer output signal, and the mapping space is constructed by the physical correlation of the information in the transformer group, and the true value of the transformer is estimated. Based on the algorithm, the data output evaluation, state analysis and risk prediction system for multi-time-space data monitoring are established to realize online evaluation and risk warning of transformer measurement.