Synchronous Condenser Fault Diagnosis Method Based on RBF Neutral Network

Abstract

The traditional synchronous condenser fault diagnosis algorithm is too dependent on experience, and it cannot accurately reflect the relationship between features and failure type, resulting in the low diagnostic accuracy. Aiming at this problem, the radial basis function (RBF) neural network algorithm is used to diagnose the synchronous condenser fault. The operating parameters of synchronous condenser are used for learning and training, and the corresponding fault type output is obtained. The simulation results show that the diagnosis algorithm has a high accuracy and a small calculation error, proving that the diagnosis method can be widely used for on-line monitoring of the synchronous condenser.