Improved Artificial Fish Swarm Algorithm for Fault Diagnosis of Dry-Type Transformer

Real-time fault diagnosis of dry-type transformer plays a crucial role in the safe operation of power grids. In order to improve the accuracy of the fault diagnosis of dry type transformer, Based on artificial fish swarm optimization algorithm, this paper proposes an improved artificial fish-swarm rear-end clustering algorithm, defines the similarity factor and cluster discriminated factor, and introduces the improved algorithm to the dry-type transformer fault diagnosis, eventually established the simulation of artificial fish-swarm rear-end behavior of dry-type transformer fault diagnosis model. Experimental results show that this method has fast convergence, high accuracy in the dry-type transformer fault diagnosis, and confirmed the effectiveness of the method.