Development Of Fault Diagnosis System And Algorithms And Its Application Research On Rotating Machinery

Rotating machinery is widely used in many important areas, such aselectricity, machinery manufacturing, of which the health condition may notonly affect the operation of the equipment itself, but also get the productionaffected, also may cause severe losses of the national economy, and even leadto catastrophic crash. Condition monitoring and fault diagnosis of mechanicalequipment is very necessary. Application of signal processing for featureextraction from vibration signal, and then diagnosis of their state of health, isan important research direction in the field of fault diagnosis of rotatingmachinery and equipment.Firstly, algorithms of holospectrum based on VS.NET software platformare achieved and improved in this article. And then the analog signals, andother signals from rotor test rig are analyzed to study the applications ofholospectrum and vibration modes in vibration condition monitoring andfault diagnosis of mechanical equipment. Considering displacement signalsmore effectively sometimes, while subject to the actual conditions,acceleration signals can only be measured in many cases, numericalintegration of the vibration signal was studied in the article.Most of the algorithms of fault diagnosis achieved are usually based onMATLAB and Labview software platform, in order to meet the needs of the engineering, algorithms of fault diagnosis based on VS.NET softwareplatform are proposed. And a condition monitoring and fault diagnosissystem is developed on VS.NET in this article, for the practical application inengineering. Article has studied the real-time acquisition of vibration signals,and a multi-channel real-time signal acquisition system is established, andthen a database system is established based on Access, finally, vibrationcondition monitoring and diagnosis system of mechanical equipment is built.On the basis of this system, for practical cases is studied to explore itsapplication in the field of engineering.The main research contents are as follows:(1) Applications of holospectrum and vibration modes in vibrationcondition monitoring and fault diagnosis of mechanical equipmentConsidering the algorithms of holospectrum require high accuracy ofprocessing the amplitude, phase, and frequency components of the signals,and general methods of band-pass filtering can not meet the requirements,using frequency-domain method instead of band-pass filtering method tosolve that is studied.Considering the needs of other signal processing methods in the study ofapplications of holospectrum in vibration condition monitoring and faultdiagnosis of rotating machinery and equipment,“Analysis System ofHolospectrum Based on VS.NET” is designed. In this system, waveform,amplitude spectrum, shaft orbits,2D-holospectrum, waterfall diagram,holographic waterfall diagram is all integrated. Based on that, the accuracy ofthe algorithms is verified and applications of holospectrum on fault diagnosisis explored by comparative analysis of the signals from rotor. (2) Research of numerical integration of the acceleration signalsConsidering the advantages of acceleration sensors with a light weight,small size, wide frequency range, high sensitivity, while subject to the actualconditions, acceleration signals can only be measured in many cases,numerical integration of the acceleration signals was explored in the article.The advantages and disadvantages of integration in time domain andfrequency domain are compared.(3) the establishment and application of condition monitoring anddiagnosis system of mechanical equipmentIn order to carry out experiments and research better, establish conditionmonitoring and diagnosis system of mechanical equipment, which consists ofthree parts:"Multi-channel Real-time Signal Acquisition System","DatabaseSystem Based on Access ","Signal Analysis System". When the device isworking, vibration state of rotor system is monitored online through real-timesignal acquisition subsystem, alarm data and timing data will be obtained andstored in the database subsystem. Analysis of data retrieved from the databaseis completed offline by signal analysis subsystem. The system is establishedto provide a condition monitoring and diagnosis system and method ofvibration of equipment based on VS.NET software platform, closer topractical engineering.