Study On The Application Research Of Partial Discharge On-line Monitoring For Large Generator In Gezhouba Power Plant

Partial discharges(PD) in large generators can lead to deterioration of the insulation, resulting in insulation breakdown and short circuit. While not presenting the real status of operating one's, traditional off line test can not ensure safe-operation of generators. Insulation faults were still reported now and then. It's necessary to develop a complete, practicable on-line monitoring system of partial discharge in large generator insulation.On the base of many documents, a set of practicable system for monitoring large generator partial discharge was presented. In this paper, all the possible sites for placing sensors were selected, and give the detail illustration on characteristics of the transducer and the parameter which may influent the performance of it's response.New methods based on fuzzy mathematics and charging-quantity were applied to analyze charging signal through neutral line of generator for the first time, determining which phase this signal belonging to. Calculation and test presented these two methods could perform successfully.Aiming to depress the all types of periodical interferences at the field, FIR digital filter technique is proposed. Applications on different filtering functions are studied for different discharge patterns.The PD patterns are also recognized by using Neural Network (NN) which its inputs are statistical operators. Satisfactory results showed that NN can recognize different PD patterns successfully.