【摘要】 针对数控机床可靠性研究过程中可靠性数据收集周期长、费用较高的问题,提出了基于人工神经网络理论和算法的数控机床可靠性数据扩充方法。在对收集的少量可靠性数据进行初步分析的基础上,对人工神经网络网络进行训练,并应用训练后的神经网络对原始可靠性数据进行扩充。扩充后的可靠性数据与原始可靠性数据具有相同的失效分布规律。对扩充后的可靠性数据通过应用最小二乘法和K-S检验法进行分析,可以确定数控机床可靠性数据分布模型。以9台某型号数控车床3个月的可靠性数据为例证明该方法能够在可靠性数据较少的情况下实现对数控机床可靠性数据分布模型的确定,同时分布模型的拟合更加准确。更多还原

【Abstract】 Aiming at the fact that the collection of reliability data in reliability research of numerical control(NC) machine tool is time-consuming and costly,a reliability data expansion method was proposed based on artificial neural network.An artificial neural network was trained with a few reliability data that collected in-situ and analyzed primarily,and the trained network was used to expand the collected data.The failure distribution of the expanded reliability data was the same with the original ones.The reliability data distribution model of the NC machine tool was determined by the analyses of the expanded data by the least square method and the K-S test method.Taking the reliability data of 9 NC lathes of the certain type in 3 months as an example,it was demonstrated that the proposed method can be used to determine the reliability data distribution model of the NC machine tool based on a few reliability data,and the fitting precision is good.更多还原