[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117639-en":3,"doc-seo-117639-105":30,"detail-sidebar-cat-0-en-105":92},{"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},117639,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Machine learning and deep learning algorithms for anomaly detection - Master’s Thesis","Master’s thesis focused on using artificial intelligence for Structural Health Monitoring, where changes in a structure’s nominal configuration can signal defects needing early detection to avoid critical conditions. The work reviews machine learning and deep learning and applies two deep-learning approaches on a scaled four-story building test rig: a physics-informed autoencoder and a data-driven autoencoder. Performances are compared with methods based on experimental modal analysis, using spring stiffness changes to simulate damage, showing superiority of the proposed algorithms, especially physics-informed neural networks for accurate damage localization.","Machine learning and deep learning algorithms for anomaly detection  \nTesi di Laurea Magistrale in  \nMechanical Engineering-Ingegneria Meccatronica  \nGianluca Bombaci, 10482291  \nAdvisor:  \nProf. Simone Cinquemani  \nCo-advisors:  \nIng. Francesco Morgan Bono  \nIng. Luca Radicioni  \nIng. Claudio Somaschini  \nAcademic year:  \n2022-2023  \nAbstract: This Master’s Thesis work is an extension of two research papers that delve into the utilization of artificial intelligence in the field of Structural Health Monitoring. Changes in the nominal configuration of a structure can often indicate the presence of structural defects that require monitoring to prevent them from escalating to critical conditions. Undoubtedly, the capability to automatically detect alterationsin a structure holds significant appeal. When there is a lack of prior knowledge about the system, artificial intelligence, and specifically deep learning models, can effectively identify structural changes and enhance the ability to pinpoint the location of damage. However, it’s important to note that acquiring data related to scenarios involving damaged structures is not always feasible. Within this Master’s Thesis work, a comprehensive overview of Artificial Intelligence is provided, with a specific emphasis on machine learning and deep learning. Furthermore, two deep learning approaches are applied to a test rig featuring a scaleddown four-story building model: a physics-informed autoencoder and a simpler data-driven autoencoder. Subsequently, their performances are compared to conventional methods based on experimental modal analysis. Specifically, modifications to the system are simulated by adjusting the stiffness of the spring. Both of these machine learning algorithms demonstrated their superiority over traditional approaches. Additionally, the physics-informed neural networks exhibited a higher potential for detecting and precisely locating structural damages.  \nKey-words: Artificial Intelligence, Machine Learning, Deep Learning, Structural Health Monitoring, Fault detection, Neural Networks, Convolutional autoencoder, Physicinformed neural network.  \n1. Introduction  \nIn recent years, there has been a growing interest in the field of Structural Health Monitoring (SHM) for civil structures like buildings and bridges, as evidenced by the references [1–3] . Structures are constantly exposed to various environmental factors that can potentially impact their structural integrity. Some examples of these factors include [4]:  \n• Structural cracks that can influence the structure’s stiffness;  \n• Changes in balance or weight distribution affecting its mass;  \n• Wear and loosening in joints altering the boundary conditions for structural dynamics and connections between different sections.  \nTo address these challenges, effective damage detection techniques are crucial. SHM encompasses a range of monitoring strategies that analyze dynamic response measurements, employ feature extraction algorithms, and apply statistical analysis techniques [5] .  \nIn a broad sense, damage in structures can be defined as changes that affect their current or future performance. Detecting such damage requires a comparison between two different states of the system, in which one represents the nominal condition, often corresponding to an undamaged state of the structure. Visual inspections are a common method for locating damage. However, they can be imprecise, unreliable, and time-consuming [6] . In contrast, vibration-based techniques have proven to offer a more dependable approach to assessing a structure’s health [7–10] . Vibration is considered the most robust indicator of a structure’s state compared to other indicators [11] .  \nGiven the wealth of data generated by vibration monitoring, deep learning has emerged as a powerful tool. It can identify meaningful features within large datasets using multiple processing layers [12] . Generally, deep-learning models for damage detection rely on","cbCaiqjFHj7Cc13l","https://ap.wps.com/l/cbCaiqjFHj7Cc13l","pdf",5166810,1,38,"English","en",105,"# Introduction\n## Structural health monitoring motivation and challenges\n## Damage definition and detection approaches\n## Role of deep learning in vibration-based damage detection\n## Autoencoders and convolutional approaches\n## Physics-informed neural networks and hybrid modeling","[{\"question\":\"What problem does the thesis address in structural health monitoring?\",\"answer\":\"It addresses detecting structural defects indicated by changes from a structure’s nominal (often undamaged) configuration, especially when abnormal damaged data is difficult to obtain.\"},{\"question\":\"Which deep learning models are applied to the scaled four-story building test rig?\",\"answer\":\"A physics-informed autoencoder and a simpler data-driven autoencoder are applied to vibration data from the test rig.\"},{\"question\":\"How are structural damages simulated and how are methods evaluated?\",\"answer\":\"Damages are simulated by adjusting the stiffness of the spring, and the proposed algorithms are evaluated by comparing their performance against conventional methods based on experimental modal analysis.\"}]","Machine learning and deep learning algorithms for anomaly detection - Master’s Thesis | PDF",1785677539,96,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-and-deep-learning-algorithms-for-anomaly-detection-masters-thesis","",{"@graph":36,"@context":86},[37,54,69],{"@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/machine-learning-and-deep-learning-algorithms-for-anomaly-detection-masters-thesis/117639/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-02",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the thesis address in structural health monitoring?","Question",{"text":76,"@type":77},"It addresses detecting structural defects indicated by changes from a structure’s nominal (often undamaged) configuration, especially when abnormal damaged data is difficult to obtain.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which deep learning models are applied to the scaled four-story building test rig?",{"text":81,"@type":77},"A physics-informed autoencoder and a simpler data-driven autoencoder are applied to vibration data from the test rig.",{"name":83,"@type":74,"acceptedAnswer":84},"How are structural damages simulated and how are methods evaluated?",{"text":85,"@type":77},"Damages are simulated by adjusting the stiffness of the spring, and the proposed algorithms are evaluated by comparing their performance against conventional methods based on experimental modal analysis.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]