Research On Periodic Assessment And Fault Reasoning Method Of Nuclear Emergency Diesel Generator

Nuclear power emergency diesel generator is the last safety defense line of nuclear power plant,whose function is supplying power to emergency equipment in time when nuclear power plant loses both working power supply and standby power supply,ensuring the reactor shutdown safely,preventing damage to main equipment and guaranteeing the safety of personnel and environment.As a long-term standby equipment,the functional status of each system is mainly evaluated and verified by periodic tests.However,due to the changeable working conditions of periodic tests and start frequently and rapidly,which accelerate the degradation rate of components,which requires that the health status of each system of emergency diesel generator be accurately evaluated and warning timely under finite parameters during periodic test.Early warning.Meanwhile,according to nuclear power plants requirements,the failures of emergency diesel generators must be eliminated in limited time and restore the generating set to usable state.Therefore,it requires rapid and accurate location of the fault reason and improve the diagnostic efficiency.In this paper,a series of studies are carried out on how to evaluate the operation status accurately and effectively of emergency diesel generator at regular intervals and how to improve the efficiency of fault diagnosis.(1)Aiming at the current shortcomings of monitoring evaluation methods of emergency diesel generator,an early warning evaluation model for emergency diesel generators is established based on the combination of analytic hierarchy process and fuzzy theory.Considering the variation characteristics of system parameters,the evaluation models for emergency diesel generators are constructed which based on the fusion of subjective and objective weights,the relative deterioration analysis method and the statistical analysis method,Through the principle of maximum membership degree and validity analysis,the health status of each system and the whole machine in the process of periodic test is evaluated.(2)Aiming at the problems of multi-source fault diagnosis information,different semantics,lack of fault cases and so on.the fault modes and fault reasons of various systems of emergency diesel engines are sorted out based on the method of FMEA,and the hierarchical causality among fault events is established based on FTA analysis method.(3)The fault diagnosis model of the emergency diesel generator is establish which based on the knowledge representation method of ontology.Firstly,ontology-based knowledge representation method is used to standardize the concepts of fault phenomena,fault causes and fault structures,and different concepts are given relevant attribute knowledge.Secondly,the relationship between knowledge is established and the relevant diagnostic rules are constructed.Finally,the start-up timeout fault reasons of emergency diesel generators is located successfully based on this method,of which the effectiveness is also verified.