| The forage crushing machine is a forage processing equipment developed independently in China.At present,the machines generally have problems such as high failure rate,low reliability and incomplete failure data.The main function design of the forage crushing machine is carried out but the reliability design is not carried out.So the quality of the machine,especially the reliability,is obviously low.Therefore,it is particularly important to design and optimize the system reliability of the machine to improve its reliability.In view of low reliability of the forage crushing machine,firstly,the failure mode and hazard analysis（FMECA）of the machine is carried out based on the fuzzy theory,and the key component and weak link of the machine are identified.Secondly,the Bayesian network is combined with the fault tree to calculate the probability of the top event and the importance of each bottom event.Third,the reliability of the system is predicted by the expert scoring method.Based on the reliability prediction results and targets of the machine,the reliability distribution of the equipment is carried out by the redistribution method.Finally,the reliability check and structure optimization of the weak link of the machine are carried out.The results of research show:（1）The fuzzy FMECA method is used to obtain the comprehensive hazard ranking of the failure modes of the machine.The failure modes such as the fatigue fracture of the throwing blade,the excessive wear of the hammer and the unbalanced rotor rotation are more hazardous.And it is determined that the key component and weak link that affect the reliability of the crusher is the rotor part.（2）According to the fault data of the forage crushing machine,the Bayesian network and the fault tree are combined to calculate the probability of failure of the machine to be 4.87×10-5/h.The root nodes that have a greater impact on the change of the failure probability of the forage crushing machines are insufficient fatigue strength,the hammer’s service life reaching the limit,and the uneven wear of the hammer.（3）The expert scoring method is used to predict the reliability of the machine that is 0.75.The result does not meet the reliability index requirements of agricultural and livestock machinery.The reliability predicted value of rotor part is 0.808,which is the lowest value among the subsystem.In order to improve the reliability,the redistribution method is used to carry out the reliability distribution of the machine.When the reliability of the rotor part reaches 0.85,the reliability of the machine meets the design requirements.（4）According to the load spectrum of the rotor part when machine is working,Matlab is used to fit the maximum stress distribution law of the rotor part and K-S hypothesis test.After the stress distribution law is obtained,the structural reliability of the rotor part is obtained by the stress-strength interference model to be 0.785,which does not meet the reliability distribution requirements,so the structural optimization of the rotor part is required.（5）Based on the Isight,an approximate model of the design variables and optimization goals is established from the sample points of database,and the sequential quadratic programming method is used for global optimization,and the optimized maximum stress value of the rotor part is 87.618Mpa.The reliability of the rotor after optimization is 0.8508,which meets the requirements of the reliability design index,and the optimization result is feasible. |
