【作者】 陈保家； 陈雪峰； 张周锁；

【Author】 CHEN Baojia,CHEN Xuefeng,Zhang Zhousuo (Suzhou Academe,Xi’an Jiaotong University,Suzhou 215021) (School of Mechanical Engineering,Xi’an Jiaotong University,Xi’an 710049)

【机构】 西安交通大学苏州研究院； 西安交通大学机械工程学院；

【摘要】 为了实时反映设备的性能与精度,提出了一种基于工件加工质量的刀具可靠性预测方法。研究了性能退化指标选取、瞬时可靠度计算以及预测模型的建立和应用,结合Bayes方法和KM估计器思想,将状态特征指标比例关联关系作为瞬时可靠度,该算法简单高效,且不需要对设备失效概率密度函数进行估计,避免了因样本太少而引起的估计误差。利用刀具加工过程中的粗糙度时变数据,以可靠度为评价标准,正确地预测出了刀具的可靠性和失效时间,该过程表明工件加工质量信息用于可靠性预测的可行性和有效性。更多还原

【Abstract】 In order to reflect the operation performance and precision of the equipment,a reliability forecasting method based on machining quality of work piece is proposed.Extraction of degradation characteristic index,computation of instantaneous reliability,construction and application of the neural network prediction model are studied.Incorporating Bayes theorem and Kaplan-Meier(KM) estimator principle,the proportional association relationships of the degradation features are taken as instantaneous reliability.The algorithm is simple,efficient and not necessary to estimate the probability density function of degradation feature.So it avoids a large estimation error caused by too small sample size.Using the time-varying machining quality data of the work piece,a trained neural network is available of forecasting the reliability and failure time accurately.The results show the feasibility and effectiveness of the proposed method.更多还原