Research And Application For Fault Early Warning And Diagnosis Of Variable Load Working Condition Of Diesel Engine

Diesel Engine is a typical power machine.It is widely used in the production of power machinery.And it is rapidly developing towards the direction of precision,high speed,automation and systematization.The structure of equipment is increasingly complex and the connection between parts is becoming more and more closely.If it fails,it will affect the entire production process.In severe cases,accidents will occur and cause unnecessary losses.Therefore,along with the rapid development of diesel engines,how to prevent the occurrence of failures and analyze the cause of failures afterwards has received increasing attention from scholars at home and abroad.Based on this,diesel engine condition monitoring and fault early warning diagnosis technology has made great progress.However,the research results at the present stage are mostly limited to the conventional theoretical research.There are few engineering applications and many problems in practice cannot be solved.This article focuses on the above issues,mainly carried out the following four aspects of the study:1?The diesel engine condition identification method is studied.By extracting the characteristic parameters and reducing the dimension and classifying it,a good condition identification result is obtained,which laid the foundation for the next fault diagnosis under variable load conditions.2?Based on the diesel engine precise crankshaft phase characteristic analysis method.Aiming at the inaccuracy of the traditional using the average speed to transform the angle domain,a method of calculating crankshaft phase accurately by using instantaneous speed is put forward.The advantages and disadvantages of the two methods are compared,and the criterions of the two methods are given through examples.3?The fault early warning method based on the envelope correlation analysis of the diesel engine cylinder head vibration signal is studied.Aiming at the problem that the traditional characteristic parameters can not reflect the working state of the diesel engine under variable load conditions,a fault early warning method of envelope correlation analysis using the vibration signal of the cylinder head is proposed.A variety of abnormal conditions and normal conditions of the unit are correlated and verified by real vibration data to demonstrate the feasibility of the method.4?The diesel engine fault diagnosis method based on MEMD and FCM is studied.The problem that the accuracy of fault diagnosis is reduced due to the modal aliasing existing in the EMD.The MEMD method is proposed to suppress the occurrence of modal aliasing.The FCM is used to cluster the vibration signals of the diesel engine under various states.The examples show that this method can obtain satisfactory clustering results for various diesel engine states and improve the accuracy of fault diagnosis.5.Completed a number of diesel engine typical fault simulation experiments.Through the fault simulation experiment of the diesel engine,the real vibration signals of various faults are obtained,and a large amount of fault experimental data is accumulated,which enriched the database of diesel engine faults.Through the fault data,the feasibility of the above methods is verified.A test experiment of an engineering experiment project is completed.Through testing and analysis of the on-site unit,actual unit fault data is used to verify the feasibility of the proposed method and ensure the safe operation of the unit.