Research On Diagnosis Technology For Cavitations Faults Of Faults Of Fluid Machinery Based On Acoustic Emission Detection Method

Fluid machinery such as hydraulic turbine and pump is the key energy conversionequipment in hydropower plants and pump stations respectively. Cavitations，which isone of the common faults in fluid machinery, would directly threaten the safe andeconomic operation of units. Due to the deficiencies in the commonly used detectionmethods, cavitations in fluid machinery cannot be detected effectively. Therefore, it’sof great theory and application value to introduce the acoustic emission technique(AET) to the cavitations diagnosis domain. Theoretical and experimental methods arecombined and used in this thesis. The detection technique for cavitations in fluidmachinery based on AET was studied and the corresponding detection system wasdeveloped. The main contents of the thesis follow.The dynamic characteristics of single transient cavitations bubble in the growing andbreaking process was simulated. The simulated results show that the maximum speed canreach up to 130 m/s and the highest pressure can reach to 900 atmospheres when the singlecavitations bubble collapse. Also the statistic characteristics of the cloud cavitations bubblesin growing and breaking process are studied, and the varying laws of cloud cavitationsbubbles radius are simulated. From the simulated results, a local pressure pulse phenomenoncan be discovered obviously. Above-mentioned study conclusion can provide some theoreticalguides for the actual research.The mechanism, characteristics, and processing method of acoustic emission signalsgenerated from fluid machinery cavitations are discussed. Multi-scale and multi-resolutionanalysis on the acoustic emission signals of a centrifugal pump was performed via wavelettransformation. Study results have shown that signals of each wavelet layer are stablewhen the pump is working normally, the signal intensities of each wavelet layerobvious increase when cavitations exist, and abnormal wave crests have appeared inthe second and the third layer of the spectrum with a dispersed distribution.A cavitation detection system for the fault diagnosis of fluid machinery wasresearched and developed using LabVIEW as developing platform. The detection systemincludes hardware and software. The hardware part is simple, easy-assembled and highlyportable. The software part has the capacities of data collection, condition monitoring, datastorage and data recalling. It can provide suitable processing methods for the acousticemission signals under cavitation conditions. And the determination indices can also beextracted. The results from both the lab and the field application have verified theperformance of the developed system, which will further promote the corresponding study ofAET on the cavitation monitoring and diagnosis of fluid machinery.