[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121027-en":3,"doc-seo-121027-105":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":4,"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},121027,5909877438554,"Maeve","https://ap-avatar.wpscdn.com/avatar/5600025385ad2bf12a7?_k=1778553567797529272",8,"Research & Report","On Automatic Machine Learning for Industrial Condition Monitoring - Dissertation","This dissertation investigates how automated machine learning can be applied to industrial condition monitoring. It tackles recurring practical challenges by developing a modular machine-learning approach that can be deployed with minimal machine-learning expertise through full automation. The framework targets diverse monitoring scenarios using mutually complementary algorithms and toolbox-like structures. Model outputs are physically interpretable, improving trust and enabling deeper process understanding, while maintaining prediction robustness comparable to neural networks. It also supports deployment on low-cost edge hardware near sensors to cut energy consumption and communication bandwidth versus cloud-based solutions, and includes novelty detection for outlier detection, supervised-learning monitoring, and previously unknown fault identification. The approach is extensively validated against alternatives in multiple application scenarios, with continued development at ZeMA’s Data Engineering and Smart Sensors group and the Lab for Measurement Technology.","On Automatic Machine Learning for Industrial Condition  \nMonitoring  \nDissertation  \nzur Erlangung des Grades des Doktors der Ingenieurwissenschaften der Naturwissenschaftlich-Technischen Fakultät der Universität des Saarlandes  \nvon  \nTizian Schneider  \nSaarbrücken  \n2024  \nTag des Kolloquiums: 19.07.2024  \nDekan: Prof. Dr. Michael Vielhaber  \nBerichterstatter: Prof. Dr. Andreas Schütze  \nProf. Dr. Robert Schmitt  \nVorsitz: Prof. Dr. Dirk Bähre  \nAkad. Mitarbeiter: Dr. Amine Othame  \n-2-  \nAbstract  \nThis thesis studies the question of how to utilize automated machine learning in industrial condition monitoring. The typical issues encountered are addressed, and a modular machine-learning approach is developed to solve them. Namely, it is easily applied with little machine learning knowledge due to full automation. It applies to various condition monitoring scenarios due to mutually complementing algorithmsand toolbox-like structures. Its results are physically interpretable, creating an extra layer of trust and allowing deeper process understanding, while the prediction quality is on par with neural networks in robustness tests. Furthermore, the approach facilitates deployment on low-cost, high-efficiency edge hardware close to the sensors. That, in turn, reduces energy costs and required communication bandwidth compared to cloud computing. Additionally, the approach includes novelty detection and concepts that utilize it for outlier detection, monitoring of supervised learning, and detection of previously unknown faults. All named capabilities have been extensively and successfully compared to other approaches in different exemplary application scenarios. This success started the Data Engineering and Smart Sensors group at ZeMA and the Lab for Measurement Technology that further researched and extended the approach, e.g., by traceable uncertainty estimation following the Guide to the Expression of Uncertainty in Measurement.  \n4  \nKurzfassung  \nDiese Thesis beschäftigt sich mit der Frage, wie automatisiertes maschinelles Lernen für industrielle Zustandsüberwachung eingesetzt werden kann. Ausgehend von dabei typischerweise auftretenden Problemen wird ein automatisiertes Konzept zu deren Lösung entwickelt. Es ist durch die Automatisierung ohne tiefes Verständnis maschinellen Lernens einsetzbar. Weiterhin deckt es mit sich gegenseitig ergänzenden Algorithmen und einer offenen Baukastenstruktur ein breites Anwendungsspektrumab. Die gelernten Modelle sind physikalisch interpretierbar, was zu ihrer Vertrauenswürdigkeit beiträgt und den Aufbau zusätzlichen Prozessverständnisses ermöglicht. Gleichzeitig ist ihre Robustheit der von neuronalen Netzen gewachsen. Das Konzept kann auf kostengünstiger Rechenhardware direkt am Sensor umgesetzt werden, was im Vergleich zu Cloud-Computing notwendige Bandbreite und Energiebedarf reduziert. Darüber hinausgehend werden Konzepte zur Anomalieerkennung entwickelt, die Ausreißerdetektion, Überprüfung des überwachten Lernens oder Erkennung bisher unbekannter Schäden ermöglichen. Alle genannten Fähigkeiten wurden in mehreren Anwendungen mit denen anderer Konzepte verglichen. Die erzielten Erfolge führten zur Entstehung der Gruppe für Data Engineering and Smart Sensors am ZeMA und am Lehrstuhl für Messtechnik, in der diese Forschung weitergeführt und ausgebaut wurde. Zum Beispiel wurde die Berechnung der Messunsicherheit nach Guide to the Expression of Uncertainty in Measurement erweitert.  \n6  \nTable of Contents  \n1 INTRODUCTION ......................................................................................................................9  \n2 THESIS STRUCTURE AND PUBLICATIONS ................................................................... 10  \n2.1 APPENDED PAPERS AND AUTHOR’S CONTRIBUTION AS PART OF THE THESIS......................... 11  \n2.2 OTHER APPENDED PAPERS BASED ON THE AUTHOR’S WORK................................................ 12  \n3 VISION SENSOR 4.0..........................","cbCair9r8lzyMBEM","https://ap.wps.com/l/cbCair9r8lzyMBEM","pdf",22893398,1,217,"English","en",105,"# 1 INTRODUCTION\n# 2 THESIS STRUCTURE AND PUBLICATIONS\n## 2.1 APPENDED PAPERS AND AUTHOR’S CONTRIBUTION AS PART OF THE THESIS\n## 2.2 OTHER APPENDED PAPERS BASED ON THE AUTHOR’S WORK\n# 3 VISION SENSOR 4.0\n# 4 BACKGROUND, CHALLENGES, AND CURRENT RESEARCH\n## 4.1 APPLICATIONS AND DATA PROPERTIES\n## 4.2 MACHINE LEARNING CHALLENGES\n## 4.3 DATA SCIENCE APPROACHES","[{\"question\":\"What problem does the dissertation address in industrial condition monitoring?\",\"answer\":\"It studies how to use automated machine learning for industrial condition monitoring and addresses typical issues that arise in practice.\"},{\"question\":\"How does the proposed approach improve usability and deployment?\",\"answer\":\"It is designed for full automation, so it can be applied with little machine-learning knowledge, and it can run on low-cost edge hardware near sensors.\"},{\"question\":\"What additional capabilities does the approach include beyond prediction?\",\"answer\":\"It incorporates novelty detection concepts used for outlier detection, monitoring of supervised learning, and detection of previously unknown faults.\"}]","On Automatic Machine Learning for Industrial Condition Monitoring - Dissertation | PDF",1785733380,547,{"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},"on-automatic-machine-learning-for-industrial-condition-monitoring-dissertation","",{"@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/on-automatic-machine-learning-for-industrial-condition-monitoring-dissertation/121027/",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","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the dissertation address in industrial condition monitoring?","Question",{"text":75,"@type":76},"It studies how to use automated machine learning for industrial condition monitoring and addresses typical issues that arise in practice.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed approach improve usability and deployment?",{"text":80,"@type":76},"It is designed for full automation, so it can be applied with little machine-learning knowledge, and it can run on low-cost edge hardware near sensors.",{"name":82,"@type":73,"acceptedAnswer":83},"What additional capabilities does the approach include beyond prediction?",{"text":84,"@type":76},"It incorporates novelty detection concepts used for outlier detection, monitoring of supervised learning, and detection of previously unknown faults.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]