Research On State Detection And Fault Diagnosis Of Hydropower Unit Bearing Based On Symptom-driven

With the rapid development of China’s economy,the domestic energy consumption demand is increasing.As a clean energy source,hydropower takes more and more proportion in China’s energy supply,and it is our important choice of dealing with blowout energy demand and renewable energy sources.In recent years,hydropower units have gradually developed towards giant,large-capacity and high water-head,and the problem of security and stability is becoming more and more serious.Bearing as an important component of hydropower unit,its performance will have an important impact on the safety and stability of generating units.The temperature rising and wearing of bearing tile surface often happens in many domestic power plant units,which will have a direct impact on the stable operation of the unit,and even endanger the safety of the power grid.Therefore,it is pretty necessary to carry out state detection and fault diagnosis of bearings.This paper has developed state detection and fault diagnosis system of bearing to monitor the operation state of the bearing in real time,and analyze the possible failure of the bearing,which can provide technical support for the final realization of the unit’s state overhaul.In this paper,the state detection and fault diagnosis system of hydropower unit bearing is studied through adopting a rule-based expert system diagnosis method.And a set of software for the system is developed by using Delphi.The system uses C/S mode to realize data sharing in local area network,and has the functions of real-time monitoring,alarm query,historical trend analysis,bearing state analysis and diagnosis,as well as can fully reflect the operating status information of the unit bearing.In terms of data acquisition and storage,this paper has completed the system hardware design,including measurement parameters,sensor and measurement point layout choice,etc.The monitoring data is stored in the way of database combining with data file,and the real-time database,historical database,alarm database,start-stop database and static parameter database are designed to effectively manage the basic data.In the design of bearing fault diagnosis model,a rule-based expert system is used.The expert knowledge is regularized by using the credibility-based production rule representation,which makes the expert system more open and extensible,as well as let it convenient to supplement and perfect the expert system knowledge base in the future.At last,through the summary of the research content of this paper,the author points out the shortcomings of the system,and provides some reference and direction for the follow-up study.