Research On800MN Die-forging Hydraulic Press Reliability And Monitoring System Based On Fatigue Analysis

Heavey die-forging hydraulic press is a very important technical equipment forprocessing large forgings, especially the aviation forging products, has the vitalsignificance to the national economy, national defense strength etc., and high attentionshould be payed on ensuring the operation reliability of large die-forging hydraulic pressand on predicting its failure life. Current researches on fatigue reliability in our countryand abroad are mainly focused on the element level, and the research on the system levelfor mechanical system is very seldom. Research results about fatigue reliability ofmachine body structure of die-forging hydraulic press on pre-tightening condition havenot been reported. In this paper the method and means for analysising the strength andpredicting and monitoring the structure fatigue life of the pre-tightened large hydraulicmachine are established.In this paper, based on fatigue reliability analysis theory, the hydraulic forgingmachine body structure system on pre tightening condition were studied, including the useof finite element software and fatigue analysis software, the large die forging hydraulicpress1:1model for finite element analysis of the hydraulic machine and the variouscomponents of the stress and strain results, the fatigue analysis of the fatigue life of eachpart, and reliability analysis reliability life of the whole hydraulic machine; the changeform equivalent stress under the action of external loading large die forging hydraulicpress structure after preloading, theoretically proved that the equivalent stress and there isa possibility of minimum in the process with the change of load in the analytical solution,its position and amplitude are obtained, and the equivalent preload stress fatigue analysisalgorithm; processing of load spectrum, to avoid peak valley node in non circulatingthrough the additional condition when the counting, the method can simplify the countingprocess, and the improved three peak valley count the model not to compress the data; themethod of field investigation, gathering in a period of time corresponding to different loadWorking cycle number cluster, and based on the equivalent load gets press of artificialneural network technology based on spectrum, the fatigue life prediction model based on BP neural network.This paper developed a capable of loading, stress and strain monitoring and dataprocessing of the test monitoring system, this system also has on-line compressorstructural static strength and fatigue life monitoring and prediction function, andconducted laboratory experiments; experiment using the actual use conditions close to thepress at the time of piecewise continuous loading mode, the signal at any time monitoringof each measuring point is abnormal and regularly check the measuring point whether nearcrack method to obtain the fault free loading times, the experimental results show that,using electrometric method detecting the working stress and strain state is feasible,reproduction can be achieved through press technology parameters of the real-time dataacquisition, the system to meet the use requirements, it has practical value.