Contact net insulator pollution flashover prediction method

Abstract

The invention discloses a contact net insulator pollution flashover prediction method, which includes:using artificial fish school algorithm to optimize BP neural network, setting up contact net insulator pollution flashover grade prediction model, predicting insulator pollution grade in the present month. As one-dimensional input vector of the pollution flashover probability prediction model, the predicted value of the contact net insulator pollution grade is used to predict the pollution flashover probability of the contact net insulator pollution flashover and the probability of the contact net circuit in the present time. Through the contact net circuit pollution flashover probability prediction model, the pollution flashover probability which is obtained is used to set up a short-period forecast response mechanism of the contact net circuit pollution flashover. The contact net insulator pollution flashover prediction method has the advantage of saving manpower and material resources, having low financial cost, easily operating, being convenient and quick, which realizes provides certain booster actions for realizing electric equipment from scheduled overhauling to status overhauling.

Description

technical field [0001] The invention belongs to the technical field of electrical equipment, and in particular relates to a pollution flashover prediction method for catenary insulators. Background technique [0002] The traction power supply catenary is an open-air power supply equipment. With the increasing air pollution, it has a great impact on the pollution of the insulators. The insulation performance is subjected to real-time tests of severe weather such as rain, snow, fog, and industrial dust and coal smoke from surrounding factories and mines. The discharge flashover phenomenon caused by dirty and wet insulators will cause large-scale power outages of catenary equipment, which seriously threatens the safe production of railway transportation. Although the pollution flashover accident is inferior to the lightning damage accident which accounts for the first power system accident, the loss caused by it is ten times that of the lightning damage. At present, there are ...

AI Technical Summary

Problems solved by technology

[0010] The purpose of the present invention is to provide a pollution flashover prediction method for catenary insulators, which solves the problems that the current pollution flashover prediction is mainly based on the neural network model, and the model is easy to fall into local optimum and the network convergence speed is slow.

Images