Research On Fatigue Life Prediction Of Wire Rope Based On Grey Particle Filter Method Under Small Sample Condiction

Wire rope is a key component used in engineering practice.At present,the research on its reliability is mostly based on a large number of fatigue tests to predict the fatigue life.However,the wire rope used in engineering has high cost and long fatigue failure time.Considering from the economic and practical point of view,it does not have the conditions of large sample test,and it is impossible to obtain reliable prediction results from small sample fatigue data relying on traditional statistical theory.Therefore,the research on the fatigue life prediction of wire rope based on small sample condition has very important theoretical value and engineering practical significance.In this paper,the 6×31WS+FC type wire rope is taken as the research object.Under the condition of fatigue test data of small sample size,the fatigue life of wire rope under various working conditions is predicted.This paper mainly carries out the following works:Firstly,the dangerous parts and mechanical model of wire rope are determined.According to the structure of 6×31WS+FC wire rope and fatigue test conditions,the mechanical model and three-dimensional model of wire rope are established.The dangerous parts with the largest equivalent stress are determined by numerical calculation,and the correctness of the dangerous parts and the mechanical model is verified by finite element simulation.Secondly,the stress-reliability life sequence of the dangerous parts of wire rope is established.According to the test conditions,the dynamic simulation of wire rope is carried out,and the obtained load-time curve is substituted into the mechanical model to obtain the equivalent stress-time history of the dangerous parts of wire rope,and then the symmetrical alternating stress spectrum under each tension is compiled;Use the bootstrap method to regenerate the small test sample,and the reliability life corresponding to each symmetrical stress spectrum is calculated combined with Weibull distribution,which provides a basis for the subsequent fatigue life prediction of wire rope.Then,based on the grey particle filter method,the fatigue life prediction of wire rope under small sample condition is realized.With the stress-reliability life sequence of small sample in the high-cycle fatigue stage for conditions,the non-equidistant GOM(1,1)model is established to obtain the P-S-N curve for predicting the law of high-cycle fatigue.Based on the fatigue behavior in the high and low cycle stages,the error correction formula of grey prediction curve is derived.and the reliable life prediction of wire rope in full fatigue stage is realized by combining with the parameter estimation function of particle filtering method,and the relative error between prediction result and reliability life obtained by the test data is below 5%.It is shown that the theoretical prediction is in good agreement with the experimental results.Finally,the application scope of the grey particle filter method is determined.The grey particle filter method is used to predict the whole stage fatigue life of wire rope under combined loading conditions,different production and technological process conditions,and different pulley diameter conditions,and the results are compared with the test results.The results show that the relative error between the results of grey particle filter method under small sample condition and reliability life obtained by the test data is below 8%,in predicting the whole stage fatigue life of wire rope under single and combined loading condition and test parameter changing condition.And the prediction accuracy is greatly improved compared with nominal stress method and local stress-strain method,the grey particle filter method is proved to be feasibility and reliable in predicting the fatigue life of wire rope.