A Method of Realizing Fault Diagnosis Control Based on Causal Chain of Event Sets in Power System

Abstract

The invention relates to a fault diagnosis control method in a power system based on an event set causal chain and relates to the technical field of control methods in power systems. The method comprises firstly, a system telesignalisation is described as events which form event sets according to attributes; then the event set causal chain can be formed according to causal relationship of the event sets; and finally a causal chain head is confirmed according to logic relationship of the causal chain. Thus, relation between event sets is established according to the relay protection principle and event generation causal relationship, causes can be found according to results to diagnose faults and repeat events, the difficult problem for acquiring knowledge in the prior art is effectively solved, the telesignalisation serving as judgment basis is convenient to acquire, high in calculation efficiency and capable of recognizing false alarm and failure alarm easily, and diagnosis results have explanatory. The fault diagnosis control method in the power system based on the event set causal chain can meet practicability and accuracy requirements of fault diagnosis of the power system.

Description

technical field [0001] The present invention relates to the technical field of power system control methods, in particular to the technical field of power system fault diagnosis methods, and specifically refers to a method for realizing fault diagnosis control based on event set causal chains in power systems. Background technique [0002] Fault diagnosis of power system can be divided into expert system method, artificial neural network method, genetic algorithm and logic processing method in terms of its application method. The first three belong to artificial intelligence technology. It has been widely used in the field of fault diagnosis by taking into account the characteristics of human experience and certain learning ability. Distinguish from the level of data collection. The fault diagnosis research methods mainly include fault diagnosis based on circuit breaker action information, fault diagnosis based on relay protection action information, fault diagnosis based on...

AI Technical Summary

Problems solved by technology

For the fault diagnosis methods of expert system and neural network, both have their own shortcomings: neural network is based on numerical value and calculation, usually each output neuron corresponds to a fault cause, when a new fault phenomenon (that is, a new input) appears , if some faults are too different from the training and learning samples, there may be no corresponding output unit corresponding to them, and the fault diagnosis system will make mistakes; while the expert system is based on rules and heuristic reasoning, which is characterized by fault symptoms and See the clear corresponding relationship in the resume for the reason. Due to the diversity of knowledge and the change of equipment structure, knowledge needs to be updated continuously. Because of the "bottleneck" problem of knowledge acquisition in the traditional expert system design method, the gap between the fault phenomenon and the fault cause There are a large number of non-one-to-one correspondences, resulting in the incompleteness of the expert knowledge base

Images