复杂工程系统预测与健康管理和测试性设计的集成技术关键问题研究

Prognostics and Health Management (PHM) is an advanced technology in the developed world such as the United States of America and the European Union, to improve the “RMSST” (Reliability, Maintainability, Safety, Supportability and Testability) and reduce the cost in the whole life cycle of engineering systems. Researches on the theories and methods of PHM have being carried out widely and deeply in China with tremendous world-class achievements; however, there is still a huge gap between China and the western developed countries on the application fundamental research of PHM. . This project focuses on the common fundamental problems associated with the application of PHM, and studies the key issues in integration technology of PHM and Design of Testability (DoT) for complex engineering systems such as the high-powered marine diesel engines. The research contents include the following aspects: (1) PHM-oriented life-cycle DoT theory, which will combine the idea of feedback and self- adaptation in control systems to solve the problems of testability modeling, generation of diagnostic test strategy, test sequencing optimization under unreliable tests, life-cycle test evaluation, and so on; (2) New data management and PHM techniques based on life-cycle comprehensive test information, which aims to solve the structured or semi-structured data modeling of multi-source and heterogeneous test information, fault diagnosis, prognosis and health evaluation methods under multi-layer uncertainties; and (3) Application of the new developed theories or methods for the electronically controlled common-rail system of marine diesel engines. . The research task is to solve the new scientific problems in the cross realm between PHM and DoT. The expected achievements can offer new perspectives, methodologies and technologies in the field of testability design, health management, and intelligent maintenance for complex engineering systems, in hopes that, actual applications of PHM technology can be achieved widely in China.

预测与健康管理（PHM）是欧美国家提高系统可靠性、维修性、安全性、保障性、测试性和降低系统全寿命周期费用的重要前沿技术，我国在理论研究方面国际先进，但PHM应用基础研究远滞后于国外发达国家。本项目针对PHM技术应用共性基础问题，以大功率船用柴油机为对象，研究：（1）面向PHM的全寿命周期测试性设计新方法，融合反馈和自适应思想，解决动态故障测试性建模和诊断策略生成、不确定测试优化以及全寿命周期测试性评价等问题；（2）基于全寿命周期综合测试信息的数据管理与PHM新技术，解决多源、异构测试信息结构建模、多层不确定性下的诊断、预测与健康评估实用新技术等问题；及（3）上述新方法和新技术在大功率船用柴油机共轨系统上的工程验证。项目研究内容是健康管理和测试性设计交叉领域的新问题，研究成果将为推进PHM理论全面转向工程应用、提升我国复杂工程系统测试性设计、健康管理和系统维护的应用基础研究水平提供实践经验。

复杂工程系统对预测与健康管理（PHM）先进技术需求急迫，但因普遍存在的故障测试代价高、微小渐变故障检测性能差、系统健康退化预测难等问题，以及系统可测状态信息具有多源异构、多时空耦合、非均衡、小样本、无标签等特点， PHM技术的工程实践不及预期。.本项目针对复杂工程系统PHM技术研究与应用中的共性基础问题，开展了面向PHM的全寿命周期测试性设计、全寿命周期健康状态数据建模与管理、以及PHM新技术研究与应用验证等研究工作，提出了基于定量因果图的最优传感器配置方法、基于深度copula函数的测试序列优化方法、基于D-S证据理论的多源信息融合方法、基于智能模型的故障诊断和寿命预测方法等若干具有良好原始创新性的理论方法和应用技术，初步构建了面向PHM的动态测试性设计框架和基于信息融合的诊断预测方法体系，显著提升了工程系统中微小、复合故障的检测、诊断和预测能力。.项目组针对大功率船用柴油机电控共轨系统开展应用验证，实现高压油泵、喷油器、限流阀等关键部件卡滞、磨损等12类微小故障诊断准确率90%以上。另外，相关技术推广应用于城轨车辆车门系统，实现60余种故障和退化模式的准确诊断和可靠预测，项目成果推广应用成效显著。.项目执行期内，已发表学术论文44篇（SCI检索23篇，EI 检索19篇）；申请发明专利13件（已授权5件）；获江苏省科学技术奖一等奖1项、广东省科技进步二等奖1项以及中国自动化学会、中国电子学会、中国机械工业学会科研奖励4项；参加国内外会议20余人次（做大会报告3人次），邀请国内外相关领域专家讲学10余场次；培养相关领域专业人才10人。.项目研究成果为测试性与PHM融合设计理论提供一些新思想和新方法，部分新技术获得成功应用，为PHM应用实践提供了成功经验。