Experimental Design And Analysis Of Fault State Of Nuclear Power Gearbox

The safe development of nuclear power is a concentrated expression of a country’s comprehensive strength.The seawater circulating pump gearbox in nuclear power plants is one of the core components of nuclear power units,and it operates under high loads in complex environments for a long time.If it breaks down,it requires shut down the entire unit for5-10 days,which will cause huge economic losses.Therefore,timely detection of faults have become a "stuck neck" problem that needs to be solved urgently for nuclear power equipment.One of the causes of faults is the loss and deterioration of lubricating oil.Therefore,it is very important to carry out real-time monitoring and fault analysis of the status of the gear and lubricating oil in the nuclear power gearbox.However,there are three major technical challenges in the research.The first is that the long-term operation of the nuclear power unit cannot be interrupted for testing.The second is that the lubrication system is difficult to monitor in real time online;the third is that the friction state of the gear surface is difficult to inspect in real time.In view of the above problems,with the support of the National Major R&D Program "High-power Nuclear Power Gearbox Fault Diagnosis and Health Monitoring Technology" project,the main research content of this paper is as follows:First of all,in view of the problem that the nuclear power unit cannot be shut down for the test,by studying the structure of the planetary reduction transmission device,studying the lubrication method and optimizing the structural composition,the load mechanism is designed,the gearbox test platform is designed,and the performance test of the test bench is completed.Second,aiming at the difficulty of real-time online monitoring of the lubrication system,an online oil monitoring system was designed to achieve real-time collection of oil data;an early warning method for oil pollution was proposed based on the threshold value method;a lubricating oil medium electrical constant following the gearbox operation prediction model was constructed,which realizes the dynamic prediction analysis of the oil state.Third,aiming at the problem of not being able to monitor the friction state of the gear surface in real time,an acoustic emission online detection device and an acoustic emission signal energy impact model were designed and constructed to monitor.It was also found that there is noise interference in the acoustic emission signal.Fourth,in view of the problem of noise interference in the signal,this paper proposes a singular value decomposition noise reduction method based on the difference spectrum of component energy entropy.After the signal is subjected to singular value decomposition,every two singular value groups are reconstructed into components to form a component matrix,and then calculate the component energy entropy difference spectrum,reconstruct the signal based on the maximum point of the energy entropy difference spectrum,and complete the signal noise reduction process.Fifth,test data processing and analysis.The characteristics of acoustic emission signals from normal gears and faulty gears under different working conditions were analyzed,and the results show that under different working conditions,the characteristic parameters of acoustic emission amplitude from normal gears were small,the time domain waveform has no obvious impact characteristics,the steepness coefficient is small,and the absolute maximum value of voltage amplitude was also small.The amplitude characteristic parameters of the faulty gear acoustic emission signal and the absolute value of the time domain signal voltage were greater than the normal gear signal,and there were obvious impact characteristics,and the steepness coefficient is large.The final results show that by monitoring the state of the oil online and detecting the characteristic changes of the acoustic emission signal from the gearbox in real time,the fault of the gearbox can be monitored.