发表机构
University of Stuttgart; Fraunhofer Institute for Manufacturing Engineering and Automation IPA（斯图加特大学; 弗劳恩霍夫制造工程与自动化研究所IPA）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文通过对212项研究的系统综述，提出PIML的四类分类方案，发现其在PHM中性能优于传统方法，但存在领域偏向问题，未来需关注可迁移设计等方向。
AI 中文摘要
在现代工业中，维持复杂系统的可靠性、安全性与效率依赖于预测与健康管理（PHM）。机器学习（ML）在诊断与预测领域的进展中发挥了重要作用，但纯数据驱动模型存在固有局限，如泛化能力差、无法推断因果关系、可解释性不足。物理信息机器学习（PIML）通过将先验物理知识直接融入机器学习流程，帮助缓解这些局限，因此其在PHM中的应用受到越来越多的关注。本研究通过对212项研究进行系统文献综述，探究PIML在PHM中的应用情况。该综述提出了四类分类方案，包括观测偏差、归纳偏差、学习偏差及混合方法，并进一步按PHM任务对研究进行分类。在所有四类方法中，经综述的研究均显示，在广泛的资产上，其预测性能优于传统基准方法；不过，相关文献严重偏向锂离子电池与轴承领域，且以针对特定问题的解决方案为主。总体而言，该综述表明，物理信息方法已带来切实效益，但关于前述部分局限的改进主张缺乏足够支撑证据。未来研究应优先关注可迁移的设计模式、对比集成策略的基准，以及足够轻量且鲁棒、可在真实场景中在线部署的不确定性感知模型。
英文摘要
In modern industry, keeping complex systems reliable, safe, and efficient hinges on Prognostics and Health Management (PHM). Machine Learning (ML) has largely driven advancements in diagnostics and prognostics, yet purely data-driven models face inherent limitations, such as poor generalization, an inability to infer causal relationships, and a lack of interpretability. Physics-Informed Machine Learning (PIML) helps mitigate these limitations by incorporating prior physical knowledge directly into the ML pipeline, thereby fostering growing interest in its application to PHM. This work investigates how PIML is being leveraged in the context of PHM through a systematic literature review of 212 studies. The review introduces a four-class classification scheme, consisting of observational bias, inductive bias, learning bias, and hybrid approaches, and further categorizes studies by PHM task. Across all four classes, the reviewed studies consistently demonstrate improved predictive performance over conventional baselines across a broad range of assets, although the literature is heavily skewed toward lithium-ion batteries and bearings, and dominated by problem-specific solutions. Overall, the review indicates that physics-informed approaches already provide tangible benefits, whereas claims of improvements concerning some of the aforementioned limitations lack sufficient supporting evidence. Future research should prioritize transferable design patterns, benchmarks comparing integration strategies, and uncertainty-aware models that are lightweight and robust enough for online deployment in real-world settings.
CommentsThe version of record of this article, first published in Journal of Intelligent Manufacturing, is available online at: https://link.springer.com/article/10.1007/s10845-026-02930-3 . This arXiv version is content-equivalent but differs in four respects: (i) additional appendix; (ii) disambiguated citations; (iii) numbered section headings; and (iv) figures with selectable and searchable text
DOI：10.1007/s10845-026-02930-3
