摘要
针对重大装备剩余使用寿命(remaining useful life,RUL)预测中如何构建能够准确反映系统退化并提升RUL预测精度的健康指标(health index,HI)问题,提出一种基于复杂网络的HI构建方法。首先,对多源数据的相关性进行分析,建立基于传感器指标相关性的网络模型,并结合网络动力学对多源数据间的耦合机制进行刻画。然后,利用建立的网络和动力学模型,分析各传感器指标对系统状态的影响程度并得到权重因子,通过线性加权实现多源数据的融合,构建一个复合HI。最后,基于构建的HI,利用Wiener过程进行退化建模和寿命预测,并利用C-MAPSS发动机数据集进行验证。结果表明:构建的HI能较好地反映发动机的退化过程,且在HI的性能评价结果和发动机的寿命预测结果上优于单一传感器指标和现有的一些HIs,验证了所提方法的有效性。
A complex network-based health index(HI)construction method was proposed to address the issue of constructing an HI that accurately reflects system degradation and enhances the prediction accuracy of the remaining useful life(RUL)of major equipment.First,the correla-tion of multi-source data was analyzed to establish a performance network model based on the correlation of sensor indices,and the coupling mechanism among the multi-source data was char-acterized using network dynamics.Subsequently,utilizing the established network and dynamic model,the degree of influence of each sensor index on the system state was analyzed to obtain its weight factor.The fusion of multi-source data was achieved through linear weighting to con-struct a composite HI.Finally,the Wiener process was utilized for degradation modeling and life prediction based on the constructed HI.The proposed method was validated using a C-MAPSS engine dataset.The results indicate that the constructed HI can effectively reflect the degradation process of the engine,and can outperform the single-sensor index and existing HIs in terms of HI performance evaluation and engine life prediction.Thus,the effectiveness of the proposed method is validated.
作者
蔡志强
王兆强
胡昌华
田永政
CAI Zhiqiang;WANG Zhaoqiang;HU Changhua;TIAN Yongzheng(School of Mechanical Engineering,Northwestern Polytechnical University,Xi’an 710072,China;Rocket Force University of Engineering,Xi’an 710025,China;Stata Key Laboratory of Fluid Power&Mechatronic Systems,Zhejiang University,Hangzhou 310058,China)
出处
《火箭军工程大学学报》
2025年第4期1-10,共10页
Journal of Rocket Force University of Engineering
基金
国家重点研发计划(2024YFB3311204)
陕西省自然科学基础研究计划重点项目(2025JC-QYXQ-038)
军事重点实验室开放基金(ICL-2023-0304)
流体动力基础件与机电系统全国重点实验室开放基金(GZKF-202430)。
