The Construction And Application Of Vibration Fault Diagnosis Model For Turbo Generator Set

Steam turbine generator set is the core equipment of power production.Whether it can operate safely,stably and reliably is very important for the safety,economy and other assessment indexes of the unit.Turbo generator is a high-speed rotating machine,vibration will inevitably appear in operation.When the vibration exceeds the limit value,it will affect the stable operation of the unit.Excessive vibration may sometimes cause catastrophic accidents of the unit.Therefore,vibration is an important safety performance index to measure the reliability of the unit.At present,large capacity and high parameter units have become the main units in China.With the older and more complex unit structure,the increasing length of shafting,and the increasing operating steam pressure and temperature,there are likely to be many new and difficult vibration problems in the process of start-up,shutdown and operation.Therefore,on-line monitoring and fault diagnosis of turbo generator vibration is a very important topic in power plant.In this paper,the semantic ontology technology and the vibration diagnosis technology of Turbine Electric Generator are integrated,and the main contributions and innovative achievements of this paper are as follows:(1)A knowledge model suitable for vibration diagnosis of steam turbine generator set is established.In view of the complexity,heterogeneity,difficulty in representation and sharing of terms in the field of vibration diagnosis for steam turbine generator units,the advantages of ontology in the knowledge representation of vibration diagnosis are considered,and the traditional seven step method as the construction method of vibration diagnosis ontology for steam turbine generator set is improved,and the defects in ontology evaluation and tracking update are made up.According to the construction principle of ontology,with the help of Python web crawler technology to quickly collect and sort out the network knowledge,and based on this,the protégé turbine generator set vibration diagnosis domain ontology is successfully constructed,which provides a clear formal representation method for vibration diagnosis knowledge.(2)The feasibility and effectiveness of vibration diagnosis ontology for turbo generator set is verified.Aiming at the possible inconsistency in ontology,a consistency checking algorithm based on tableau algorithm is designed to check ontology.In this paper,the SQI mechanical vibration comprehensive simulation test-bed is used to simulate the different vibration of steam turbine generator set.The vibration information obtained is collected and analyzed,and the reasoning test of ontology knowledge is carried out through an example.(3)A vibration diagnosis method based on ontology and case-based reasoning is proposed.This paper expounds the advantages of ontology in case representation,analyzes the main components of case representation,quantifies the mathematical model of semantic distance,semantic depth and semantic density,simplifies and improves the semantic similarity algorithm,establishes a hierarchical retrieval model of ontology and case reasoning based on semantic and case attribute similarity algorithm,and conducts vibration diagnosis through case study Verified.(4)Based on the existing software,combined with the main faults of steam turbine generator set,creatively completed the system development framework and reasonable processing of operation process,including the creation of ontology knowledge base and ontology and case-based reasoning retrieval module,as well as the design of the main function interface of the system.The development of vibration diagnosis system is realized by integrating protégé,visual studio C# and SQL server,which improves the human-computer interaction of the whole vibration diagnosis process and makes the operation process simple and efficient.Finally,an example is given to verify that the system can provide decision support for vibration diagnosis of steam turbine generator set and improve the efficiency of vibration diagnosis.At the same time,the accuracy of vibration diagnosis can reach 80%.