[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118783-en":3,"doc-seo-118783-105":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},118783,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","Practical Applications of Machine Learning to Underground Rock Engineering - Dissertation - Doctor of Philosophy","Rock mechanics engineers increasingly access large volumes of data from underground excavations through improving sensors, cheaper storage, and higher computational power. Machine learning provides a practical way to transform such data into engineering decision support. This dissertation evaluates real-world machine learning applications to underground rock engineering using datasets across multiple rock mass deformation contexts. Results show that maintaining original input data formats during digitalization reduces bias and improves model interpretability. A CNN predicts tunnel liner yield class for Cigar Lake Mine with optimized hyperparameters and interpretable inputs via three IVS methods. An LSTM semiautomates stress-state recalibration in FLAC3D for Garson Mine, improving principal-stress prediction accuracy and verification metrics.","PRACTICAL APPLICATIONS OF MACHINE LEARNING TO UNDERGROUND ROCK ENGINEERING  \nJosephine Sarah Morgenroth  \nA dissertation submitted to the Faculty of Graduate Studies in fulfillment of the requirements of the Degree of  \nDoctor of Philosophy  \nGraduate Program in Civil Engineering  \nYork University  \nToronto, Ontario, Canada  \nJuly 2022  \nCopyright © Josephine Morgenroth, 2022  \nTo Mama and Papa, who each shaped me into the person I am. Thank you for encouraging me to go on every adventure, including this one.  \nABSTRACT  \nRock mechanics engineers have increasing access to large quantities of data from underground excavations as sensor technologies are developed, data storage becomes cheaper, and computational speed and power improve. Machine learning has emerged as a viable approach to process data for engineering decision making. This research investigates practical applications of machine learning algorithms (MLAs) to underground rock engineering problems using real datasets from a variety of rock mass deformation contexts. It was found that preserving the format of the original input data as much as possible reduces the introduction of bias during digitalization and results in more interpretable MLAs.  \nA Convolutional Neural Network (CNN) is developed using a dataset from Cigar Lake Mine, Saskatchewan, Canada, to predict the tunnel liner yield class. Several hyperparameters are optimized: the amount of training data, the convolution filter size, and the error weighting scheme. Two CNN architectures are proposed to characterize the rock mass deformation: (i) a Global Balanced model that has a prediction accuracy >65% for all yield classes, and (ii) a Targeted Class 2/3 model that emphasizes the worst case yield and has a recall of >99% for Class 2. The interpretability of the CNN is investigated through three Input Variable Selection (IVS) methods. The three methods are Channel Activation Strength, Input Omission, and Partial Correlation. The latter two are novel methods proposed for CNNs using a spatial and temporal geomechanical dataset. Collectively, the IVS analyses indicate that all the available digitized inputs are needed to produce good CNN performances.  \nA Long-Short Term Memory (LSTM) network is developed using a dataset for Garson Mine, near Sudbury, Ontario, Canada , to predict the stress state in a FLAC3D model. This is a novel method proposed to semiautomate recalibration of finite-difference models of high-stress environments. A workflow for optimizing the hyperparameters of the LSTM network is proposed. The performance of the LSTM network predicting the three principal stresses is improved as compared to predicting the six-component stress tensor, with corrected Akaike Information Criterion (AICc) values of-59.62 and-45.50, respectively.  \nGeneral recommendations are made with respect to machine learning algorithm development for practical rock engineering problems, in terms of how to format and pre-process inputs, select architectures, tune hyperparameters, and determine engineering verification metrics. Recommendations are made to demonstrate how algorithms can be rendered interpretable with the application of tools that already exist in the field of machine learning.  \nACKNOWLEDGEMENTS  \nThank you first and foremost to my supervisors, Drs. Matthew Perras and Usman Khan. I am extremely grateful to have had the experience of being co-supervised by two experts in different fields, who also have the willingness to collaborate without ego that every PhD student dreams of. Matt and Usman were extremely supportive through both academic and personal struggles that arose in the unusual circumstances of completing this dissertation during a global pandemic. Their patience and (usually!) noncontradictory guidance were instrumental in developing machine learning techniques for rock engineering problems that not only satisfied my academic curiosity but also solved real-world problems. Thanks particularly to Matt for ","cbCaitIMVjNYXduX","https://ap.wps.com/l/cbCaitIMVjNYXduX","pdf",10533791,1,206,"English","en",105,"# Abstract\n## Key findings and general recommendations\n## CNN workflow for tunnel liner yield prediction\n## LSTM workflow for FLAC3D stress-state prediction\n## Interpretability via input variable selection (IVS)","[{\"question\":\"Why does preserving original input data formats matter for the machine learning results?\",\"answer\":\"Preserving the original input format as much as possible reduces bias during digitalization and leads to more interpretable machine learning models.\"},{\"question\":\"What was developed to predict tunnel liner yield class, and how were its settings optimized?\",\"answer\":\"A convolutional neural network (CNN) was developed using Cigar Lake Mine data. Hyperparameters optimized include training data amount, convolution filter size, and error weighting scheme.\"},{\"question\":\"How does the LSTM approach improve stress prediction for high-stress environments?\",\"answer\":\"An LSTM network was developed using Garson Mine data to predict stress state in a FLAC3D model. The approach improves performance for predicting three principal stresses compared with predicting the full six-component stress tensor, using corrected AICc values for verification.\"}]","Practical Applications of Machine Learning to Underground Rock Engineering - Dissertation - Doctor of Philosophy | PDF",1785720240,519,{"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},"practical-applications-of-machine-learning-to-underground-rock-engineering-dissertation-doctor-of-philosophy","",{"@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/practical-applications-of-machine-learning-to-underground-rock-engineering-dissertation-doctor-of-philosophy/118783/",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},"Why does preserving original input data formats matter for the machine learning results?","Question",{"text":75,"@type":76},"Preserving the original input format as much as possible reduces bias during digitalization and leads to more interpretable machine learning models.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What was developed to predict tunnel liner yield class, and how were its settings optimized?",{"text":80,"@type":76},"A convolutional neural network (CNN) was developed using Cigar Lake Mine data. Hyperparameters optimized include training data amount, convolution filter size, and error weighting scheme.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the LSTM approach improve stress prediction for high-stress environments?",{"text":84,"@type":76},"An LSTM network was developed using Garson Mine data to predict stress state in a FLAC3D model. The approach improves performance for predicting three principal stresses compared with predicting the full six-component stress tensor, using corrected AICc values for verification.","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"]