Font Size: a A A

Research On Reliability Of Overspeed Protection System Of Nuclear Power Turbine

Posted on:2021-04-15Degree:MasterType:Thesis
Country:ChinaCandidate:M H RenFull Text:PDF
GTID:2492306305966139Subject:Master of Engineering
Abstract/Summary:PDF Full Text Request
With global climate change and rising energy demand,nuclear power generation technology has become increasingly prominent in many clean energy due to its advantages of maturity,economy and sustainability,and the safety of nuclear power has also been widely concerned.Turbine is an important part of nuclear power conventional island,its overspeed protection system is the first barrier to prevent serious accidents caused by turbine overspeed,and also an important guarantee for safe and stable operation of the unit.Therefore,the reliable operation of the overspeed protection system is very important when the turbine is overspeed.Taking the overspeed protection system of reheat unit in a nuclear power plant as an example,combining with the reliability analysis method GO,this paper proposes the SVM-NARX reliability judgment network to study the reliability of the system in different states.Firstly,the system is divided according to the operation conditions and functional characteristics,and building system reliability block diagram based on key components.At the same time,by comparing several typical reliability analysis methods and combining with the characteristics of the system,GO is used to quantitatively calculate the reliability of the system under different states.Secondly,according to the characteristics of the system fault,the fault types are divided into instantaneous fault and slow fault,and according to the actual operation experience of the unit and the layout of monitoring instruments,the appropriate monitoring parameters are selected as the system status parameters.For different degrees of slow fault,the system reliability is divided by the system state parameters.Then,according to the different slow fault states of the system,NARX neural network is used to predict the reliability of the system,and using simulation platform to verify the accuracy of experimental results.Finally,the SVM-NARX reliability judgment network of steam turbine overspeed protection system is proposed to realize the accurate judgment of the influence of system state on system performance.According to the reliability judgment results of the turbine overspeed protection system in this paper,the influence degree of different faults on the system performance is clarified.Therefore,the system maintenance plan can be made to achieve predictive maintenance,improve the system availability,and provide theoretical basis for the safety design of the system,which has certain research value and application value.
Keywords/Search Tags:turbine overspeed protection system, GO method, NARX neural network, reliability prediction
PDF Full Text Request
Related items