An efficient particle swarm optimization and rule mining for fault diagnosis of dissolved gas analysis

Dissolved gas Analysis (DGA) is the most vital segment of discovering shortcoming in huge oilfilled transformers. Early recognition of beginning issues in transformers decreases unreasonable impromptu blackouts. The most important and dependable strategy for assessing the center of transformer is the disintegrated gas investigation. In this paper we have used dissolved gas analysis for the analytical and computation study which is used for the evaluation used as a diagnostic tool for evaluating the condition of the transformer. We have proposed an efficient Particle Swarm Optimization using Rule Mining for Fault Diagnosis of Dissolved Gas Analysis. In this approach we first apply associative IEC for finding the faults. Then Item based individual association are applied on different gas ratio. It is on the basis on the values taken as a data set. It shows the individual associated improvement in the concentration quantity. Then by using random particle swarm optimization it is tuned to their maximum threshold value for obtaining the saturation points. The results show the improvement in fault diagnosis and provide and approach for finding associated saturation point.