[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121723-en":3,"doc-seo-121723-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121723,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","Physics-Informed Machine Learning for Predictive Maintenance - Applied Use-Cases - Guidelines and Use-Cases","Physics-Informed Machine Learning for Predictive Maintenance combines physics and engineering knowledge with data-driven machine learning and deep learning to support condition-based and predictive maintenance decisions. The work investigates physics-informed data augmentation for ML algorithms through three industrial use cases spanning anomaly detection, fault diagnostics, and prognostics of remaining useful life. It details benefits and prerequisites for each technique and provides guidelines for practical deployment in other technical systems, especially where labeled data is scarce.","Physics-Informed Machine Learning for Predictive Maintenance: Applied Use-Cases  \nLilach Goren Huber  \nSchool of Engineering Zurich University of Applied Sciences Winterthur, Switzerland lilach.gorenhuber@zhaw.ch  \nThomas Palm  \nSchool of Engineering Zurich University of Applied Sciences Winterthur, Switzerland [janthomas.palme@zhaw.ch](janthomas.palme@zhaw.ch)  \nManuel Arias Chao  \nSchool of Engineering Zurich University of Applied Sciences Winterthur, Switzerland manuel.ariaschao@zhaw.ch  \nAbstract—The combination of physics and engineering information with data-driven methods like machine learning (ML) and deep learning is gaining attention in various research fields. Oneof the promising practical applications of such hybrid methods is for supporting maintenance decision making in the form of condition-based and predictive maintenance. In this paper we focus on the potential of physics-informed data augmentation for ML algorithms. We demonstrate possible implementations of the concept using three use cases, differing in their technical systems, their algorithms and their tasks ranging from anomaly detection, through fault diagnostics up to prognostics of the remaining useful life. We elaborate on the benefits and prerequisites of each technique and provide guidelines for future practical implementations in other systems.  \nIndex Terms—physics-informed Machine Learning, ConditionBased Maintenance, Predictive Maintenance, Anomaly Detection, Fault Diagnostics, Fault Prognostics, Deep Learning.  \nI. INTRODUCTION  \nDecision making for optimal health management of industrial assets has been traditionally performed based on domain knowledge and physical models, whenever available. However, in recent years, with the abundance of machine data and the industry 4.0 revolution, there is an increasing trend towards data-driven solutions, focusing on condition-based and predictive maintenance algorithms [1]–[4] . A natural step forward is being made with the recent movement into hybrid methods, that combine the best of both worlds: exploiting the vast basis of domain knowledge and years of experience on one hand, and making use of data-driven innovations and resources on the other hand. This combination of physics with data-driven models is known as ”physics-informed machine learning”(PIML) and can be applied in many fields and various ways, as summarized in a recent seminal review paper [5] . One of the approaches mentioned there is to use physics information in order to augment and supplement the training data for machine learning (ML) algorithms. As in other fields, some examples of PIML applications for equipment prognostics and health management (PHM) have been recently demonstrated [6]–[8] . In this paper we demonstrate different approaches to physics-informed (PI) data augmentation for ML algorithms applied to PHM problems. The approaches are demonstrated  \nL.G.H was supported by Innosuisse-Swiss Innovation Agency Grant No. 55018.1  \non 3 different industrial use cases. The use cases differ not only in their application fields but also in the task that the ML is aimed at, ranging from anomaly detection, through degradation trending and diagnostics and up to prognosis of the remaining useful life (RUL) of the machines.  \nThe use cases concretely demonstrate the various benefits of PIML for practical applications. A central advantage over pure data-driven approaches is an enhanced prediction accuracy even when labeled data is scarce, which is a common challenge in PHM problems. The second obvious advantage is that physics allows for a high degree of interpretability of the model outputs compared to models based on data alone. This, in turn, has the added value of increasing the trust and acceptance of the local domain experts in the outcomes of the models. In the other direction, the possibility to supplement traditional knowledge-based approaches with modern datadriven ones offers higher fidelity of the models on individual units d","cbCaim2lZvb6g7bD","https://ap.wps.com/l/cbCaim2lZvb6g7bD","pdf",1133616,1,7,"English","en",105,"# Introduction\n## Hybrid physics-informed machine learning approach\n# Use Case I: PIML for fault detection in solar power plants\n## Tracker faults and detection/localization goals\n# Use cases II and III\n## Degradation trending, diagnostics, and RUL prognostics\n# Comparison and guidelines","[{\"question\":\"What problem does physics-informed machine learning address in predictive maintenance?\",\"answer\":\"It supports maintenance decision making by combining domain physics knowledge with data-driven ML, improving performance for condition-based and predictive maintenance tasks.\"},{\"question\":\"How many industrial use cases are presented, and what tasks do they cover?\",\"answer\":\"Three use cases are presented, ranging from anomaly detection to fault diagnostics and up to prognostics of the remaining useful life.\"},{\"question\":\"What are the key advantages of physics-informed approaches over purely data-driven methods?\",\"answer\":\"They improve prediction accuracy when labeled data is scarce and increase interpretability, which helps gain trust from local domain experts.\"}]","Physics-Informed Machine Learning for Predictive Maintenance - Applied Use-Cases - Guidelines and Use-Cases | PDF",1785806494,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"physics-informed-machine-learning-for-predictive-maintenance-applied-use-cases-guidelines-and-use-cases","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/physics-informed-machine-learning-for-predictive-maintenance-applied-use-cases-guidelines-and-use-cases/121723/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does physics-informed machine learning address in predictive maintenance?","Question",{"text":75,"@type":76},"It supports maintenance decision making by combining domain physics knowledge with data-driven ML, improving performance for condition-based and predictive maintenance tasks.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How many industrial use cases are presented, and what tasks do they cover?",{"text":80,"@type":76},"Three use cases are presented, ranging from anomaly detection to fault diagnostics and up to prognostics of the remaining useful life.",{"name":82,"@type":73,"acceptedAnswer":83},"What are the key advantages of physics-informed approaches over purely data-driven methods?",{"text":84,"@type":76},"They improve prediction accuracy when labeled data is scarce and increase interpretability, which helps gain trust from local domain experts.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]