液压系统运行可靠性的时变耦合非概率建模与融合现场状态信息的实时评估研究

Hydraulic system is the core of power transmission and control in mechanical equipment, its operational reliability affects the reliable and safe operation of the main machinery, and it is the key to enhance the core competitiveness of mechanical equipment, thus it is very urgent and necessary to realize operational reliability real-time assessment for hydraulic system, however the real-time assessment is challenged by the time-varying coupling, uncertainty and the individual particularity of hydraulic system, cause the existing reliability modeling and assessment methods difficult to meet the requirements. This project aims to investigate the operational reliability modeling and operational reliability real-time assessment for hydraulic system. Firstly, T-S dynamic fault tree, T-S dynamic Bayesian network algorithms will be proposed and reliability modeling method based on the proposed algorithms will be put forward, to overcome the dynamic time-varying and coherent coupling characteristics of the hydraulic system. Secondly, in order to solve fault uncertainties and reliability indices, time-varying coupling non-probabilistic reliability modeling method as well as T-S non-probabilistic reliability, importance and sensitivity indices algorithms will be proposed. Thirdly, according to the individual difference and particularity of the hydraulic system, the redefinition of operational reliability will be put forward, and the operational reliability real-time assessment method based on time-varying coupling non-probabilistic model and combining with running condition information will be proposed. Finally, reliability test system platform will be set up and the theoretical research method will be verified. The research of this project will provide the theory basis for the engineering application of the hydraulic system operational reliability assessment, and will provide the technical support for the improvement of the hydraulic system operational reliability.

液压系统作为机械装备动力传输与控制的核心，其运行可靠性关系主机的可靠安全运行，是机械装备核心竞争力提升的关键，因此对液压系统的运行可靠性进行实时评估，就变得十分迫切和必要，但却受到液压系统时变耦合、不确定性、特殊性特点的挑战，现有建模评估方法难以胜任。本项目针对液压系统运行可靠性的建模与实时评估问题开展研究。首先，针对系统的动态时变性和关联耦合性，提出T-S动态故障树、T-S动态贝叶斯网络算法以及基于此的可靠性建模方法；进而，针对故障不确定性和可靠性指标求解问题，提出时变耦合非概率可靠性建模及T-S非概率可靠度、重要度和灵敏度算法；然后，针对系统的个体差异和自身特殊性，提出运行可靠度的重新定义和基于时变耦合非概率模型并融合现场状态信息的实时评估方法；最后，搭建试验系统并对理论研究方法进行验证。本项目研究可为液压系统运行可靠性评估的工程应用提供理论依据，为提高液压系统运行可靠性提供技术保障。

液压系统作为机械装备动力传输与控制的核心，其运行可靠性关系主机的可靠安全运行，是机械装备核心竞争力提升的关键，可靠性评估作为提升产品质量的关键共性技术，对推动产品创新、保证产品质量有重要作用，因此对液压系统的运行可靠性进行评估，就变得十分迫切和必要，但却受到液压系统时变耦合、不确定性、特殊性特点的挑战，现有建模评估方法难以胜任。本项目针对液压系统运行可靠性的建模与评估问题开展研究。首先，针对系统的动态时变性和关联耦合性，提出了T-S动态故障树、T-S动态贝叶斯网络算法以及基于此的可靠性建模方法，解决了具有动态时变特性、复杂的耦合结构与故障因果关系的系统可靠性建模问题；进而，针对故障不确定性和可靠性指标求解问题，提出了时变耦合非概率可靠性建模及T-S非概率可靠度、重要度和灵敏度算法，建立了可靠性评估建模与可靠性指标计算方法，解决了时变耦合系统故障状态不确定、故障统计数据不足及信息不完备情况下的可靠性指标计算求解问题，为运行可靠性评估计算提供可靠的模型与算法；然后，针对系统的个体差异和自身特殊性，提出了基于时变耦合非概率模型并融合现场状态信息的运行可靠度评估方法，建立了基于快速变分模态分解、参数优化多尺度排列熵和特征加权GK模糊聚类等状态信息处理方法，建立了基于快速变分模态分解多尺度排列熵和核模糊C均值聚类、基于最小熵解卷积-快速变分模态分解模糊近似熵和模糊支持向量数据描述等运行可靠性评估方法；最后，搭建试验系统并对理论研究方法进行了验证。本项目针对液压系统运行可靠性评估面临的问题和挑战开展研究，解决液压系统运行可靠性建模与评估问题，为液压系统运行可靠性评估工程应用提供理论支撑，为提高液压系统运行可靠性提供技术保障，为可靠性设计、状态维修、故障诊断提供依据，对求解可靠度等可靠性指标、发现潜在隐患和薄弱环节，对保障系统的运行安全可靠、避免重大事故发生，具有重要现实意义。