Transmission Line Fault Diagnosis Based On Electric Field

Recently, with the rapid growth of power need and development of our country's power industry, the 500kV power transmission system has been our main power system network. HV transmission line fault diagnoses is becoming more and more important. While Artificial Neural Network technology gain advantages over the traditional theories in such aspects. It is just based on the point to assume that Artificial Neural Network technology can be employed to solve soft fault diagnoses of transmission line. Firstly, in this paper the HV system's power-frequency electric field problem is been studied seriously. Charge Simulation Method is an effective method that is used to calculate Static Field. The Charge Simulation Method was discussed and its basic ideas and application steps were concluded. To more easily simulate assumption about transmission lines are made. This paper makes a research on the mathematic model of power-frequency electric field of high voltage transmission lines by means of the Charge Simulation Method based on image method. With MATLAB this paper simulates the resultant electric field of three- phase transmission line which has horizontal arrangement. It derives that the size and orientation of resultant electric vector related to not only time, but also spatial site. The dynamic rotary process of electric vector can be observed distinctly via the simulation. According to the track figure of rotary electric vector at different site, and a summarization is made on the distribution pattern of power frequency electric field.The neural network is a large scale of parallel nonlinear system, it has strong associative learning,self-organizing,self-adaptive and high nonlinear operation ability, so it has strong judge ability of identifying the causality of these complicated variables. The diesel engine fault is diagnosed and analysised by using the Back Propagation (BP) neural network. BP neural network is systematically studied, the structure of three layers RBF neural network,the setting of network parameters and the choice of training pattern are detailed discussed, In the transmission line fault diagnosis, the methods of extracting the finite electric field value can not reflect the characteristics of the various failure modes. There- fore, in this paper, we extract the ratio of the electric field value in different fault conditions to the value of normal condition as the characteristic parameters, this method improves the BP neural network's ability of identifying fault types. Then fault characteristic parameters are trained and identified by LM algorithm, taking the characteristic parameters as the input vector, it's diagnosis result as the output vector. The simulation experiment shows that the fault diagnosis result based on neural network is well consistent with measured values. As long as we choose enough typical initial fault sample to train neural network, the network fault-tolerant and the stability are better. The method of fault pattern recognition based on neural network can fully use information feature, realize the mapping relation between input and output, get the accurate result.