基于可用性建模的数控机床主动维修决策方法研究

Functional fault and performance fault are both important factors affecting CNC machine tools’ availability and maintenance strategy. However, the Reliability Centered Maintenance (RCM) widely used nowadays does not take performance fault into account. Meta-action decomposition is a new method to decompose CNC machine tools from the perspective of function and motion, which is proposed on the basis of CNC machine tools fine dynamic control, and can help modeling and fault analysis more precise. But the problems of availability modeling and maintenance strategy based on meta-action unit have not been solved. Therefore, with availability setting as center, meta-action unit setting as basis in the project, the fault formation mechanisms of meta-action unit were analyzed by dynamic simulation, availability was characterized and the parameter models were established firstly. Secondly, the performance characteristics and fault symptoms of meta-action unit were studied by experiments, and the mapping relationships between them and faults were analyzed, then the availability comprehensive evaluation model of CNC machine tools was built based on meta-action unit. Considering availability as constraint condition, the preventive maintenance cycle and predicted maintenance time models of meta-action unit were established, and maintenance strategy optimization model of CNC machine tools was built based on meta-action unit correlation analysis. Finally, the strategy method of availability centered proactive maintenance based on meta-action unit for CNC machine tools was formed to provide theoretical and technical support for improving availability.

数控机床的功能故障和性能故障都是影响可用性和维修决策的重要因素，但是目前广泛应用的以可靠性为中心的维修并未考虑性能故障等问题；元动作分解是基于数控机床精细化动态控制提出来的一种新方法，从功能运动的角度对数控机床进行分解，使建模和故障分析更精准，但尚未解决基于元动作单元的可用性建模与维修决策等问题。因此，本项目以可用性为中心，以元动作单元为基础，通过动力学仿真对元动作单元的故障形成机理进行分析，对可用性进行表征并构建其参数模型；接着采用试验手段对元动作单元的性能特征及故障征兆进行研究，分析它们与故障之间的映射关系，构建基于元动作单元的数控机床可用性综合评估模型；然后以可用性为约束条件，建立元动作单元预防维修周期和预测维修时机模型，并分析元动作单元相关性，构建数控机床维修决策优化模型；最终形成一套以可用性为中心基于元动作单元的数控机床主动维修决策方法，为提高数控机床的可用性提供理论和技术支撑。

本项目系国家自然科学基金青年科学基金项目（批准号：51705048），经3年的研究已按计划完成。项目组针对数控机床可用性差，以可靠性为中心的维修没有考虑性能故障以及元动作理论尚未解决基于元动作单元的可用性建模与维修决策等问题，以可用性为中心，以元动作单元为基础，研究了数控机床元动作单元故障形成机理以及可用性建模技术，在此基础之上进一步研究了基于元动作单元的数控机床可用性综合评估模型。最后，形成了一套以可用性为中心基于元动作单元的数控机床主动维修决策方法，为提高数控机床的可用性提供理论和技术支撑。具体研究内容主要包括以下几个部分：.1）数控机床元动作单元故障形成机理及可用性建模技术研究。对元动作单元的工作过程和耗损过程进行了分析，对元动作故障模式进行了深入的研究，得到了元动作理论上所有的故障模式，避免传统分析中故障模式和故障原因分不清的缺陷。同时，利用QFD和FMECA提取出了元动作单元可用性表征指标，并分别构建了参数模型。最终建立了元动作单元的状态空间模型。.2）基于元动作单元的数控机床综合评估模型研究。对元动作单元的故障信号进行了提取，基于蚁群算法建立了故障与特征之间的映射关系，最后建立了基于元动作单元的考虑可靠性、维修性、精度、精度寿命、性能稳定性等可用性特征在内的数控机床可用性综合评估模型。.3）以可用性为中心的数控机床主动维修决策方法研究。首先对元动作单元的维修特征和维修方法进行了研究，基于Grey-BWM建立了元动作单元维修方式决策模型。然后以元动作单元可用性为约束条件，建立了定期预防性维修周期模型，视情维修优化模型以及预测性维修模型。最后，构建了考虑元动作单元之间经济相关性的数控机床选择性维修决策优化模型，并给出了其遗传算法求解流程。