摘要
为了更准确描述水稻插秧机的失效规律,提高可靠性分析的准确性,对水稻插秧机的故障数据进行分析,采用两参数混合威布尔分布对水稻插秧机进行建模。以残差平方和最小为优化目标,建立参数估计优化模型,利用改进粒子群算法对其进行求解,然后采用K-S检验法对模型进行检验,对比单一威布尔模型、混合威布尔模型与水稻插秧机失效数据之间的拟合程度,得出使用两参数混合威布尔模型评估水稻插秧机可靠性的合理性,在此模型的基础上计算得到水稻插秧机的平均无故障工作时间为161.75 h,中位寿命为147.14 h,特征寿命为191.31 h,且在可靠度为0.6时,预防性维修周期为115.19 h,最后在混合威布尔分布模型的基础上计算出剩余寿命-可靠度的关系,可定量分析插秧机在一定使用时间下的剩余寿命,从而进行预测性维护。
In order to accurately describe the failure rule of rice transplanter and improve the accuracy of reliability analysis,the failure data of a type of rice transplanter were analyzed,and the model of rice transplanter was built by using two-parameter hybrid Weibull distribution.With the minimum sum of the squares of residuals as the optimization objective,a parameter estimation optimization model was established,which was solved by the improved particle swarm optimization algorithm,and then the model was tested by the K-S test method to compare the degree of fitting between the single Weibull model,the mixed Weibull model and the failure data of rice transplanter.The results show that it is reasonable to use the two-parameter hybrid Weibull model to evaluate the reliability of rice transplanter.Based on this model,the average fault free working time of rice transplanter is 161.75 h,the median life is 147.14 h,and the characteristic life is 191.31 h.When the reliability is 0.6,the preventive maintenance cycle is 115.19 h.Finally,based on the hybrid Weibull distribution model,the relationship between residual life and reliability was calculated,which can quantitatively analyze the residual life of the transplanter under a certain service time,so as to carry out predictive maintenance.
作者
文昌俊
陈洋洋
何永豪
陈凡
WEN Chang-jun;CHEN Yang-yang;HE Yong-hao;CHEN Fan(School of Mechanical Engineering,Hubei University of Technology,Wuhan 430068,China;Hubei Key Laboratory of Modern Manufacturing Quality Engineering,Wuhan 430068,China)
出处
《科学技术与工程》
北大核心
2024年第1期163-169,共7页
Science Technology and Engineering
基金
国家自然科学基金(51875180)。
关键词
可靠性
混合威布尔分布
非线性最小二乘法
粒子群算法
剩余寿命预测
reliability
mixed Weibull distribution
nonlinear least square method
particle swarm optimization algorithm
remaining life prediction
