Power grid fault diagnosis method based on temporal Bayesian knowledge base (TBKB)

Abstract

The invention discloses a power grid fault diagnosis method based on temporal Bayesian knowledge base (TBKB). The power grid fault diagnosis method comprises steps of establishing a corresponding temporal Bayesian knowledge base (TBKB) diagnosis model; clearly describing work principles and processes of relevant protectors and circuit breakers after elements are broken down; establishing a temporal causal relationship (TCR) among different action events; quantificationally expressing a successive restraint relationship among different action status nodes in the aspect of timing sequence; and identifying conditions of action events, time mark errors and the like of the abnormal protectors and the abnormal circuit breaks. Uncertain states of nodes and information loss nodes which do not pass TCR examination in alarm information are supposed and combined, so that a supposed state combination gather is formed, the fault probability of every supposed state combination is obtained, the fault probability of suspected fault elements is obtained, and fault elements are diagnosed. The protectors and the circuit breakers which operate mistakenly and fail to operate and the time mark error condition of the protectors and the circuit breakers are detected by forward reasoning of the TBKB modal to the fault elements. The total power grid diagnosis method is fast and accurate and is high in fault tolerance.

Description

technical field [0001] The invention relates to the field of power grid scheduling and fault analysis, in particular to a power grid fault diagnosis method. Background technique [0002] Power grid fault diagnosis plays an important role in the identification of faulty components, rapid recovery after faults, and prevention of chain trips. At present, the power system fault diagnosis methods mainly include expert system, Petri net, optimization analysis method, Bayesian network, D-S evidence fusion and so on. However, in some cases, due to the existence of uncertainties such as protection and switch malfunction and refusal, information loss, etc., the existing literature has not made full use of the timing relationship and time stamp information of protection and switch action events. For complex faults, The above-mentioned fault diagnosis methods are difficult to obtain correct diagnosis results. [0003] Bayesian network is good at dealing with the uncertainty of complex...

AI Technical Summary

Problems solved by technology

[0014] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a grid fault diagnosis method based on the time-series Bayesian knowledge base TBKB, so that it can overcome the various states of event variables that cannot be explicitly expressed in the prior art, and do not have strict Quantitative time series expression ability, insufficient time stamping error tolerance during timing inspection, etc.

Images