[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124699-en":3,"doc-seo-124699-105":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},124699,34359740700684,"Finn","https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487",8,"Research & Report","Application of Machine Learning to Predict Electricity Demand from Electric Vehicles in Workplace Settings - Master Thesis","Sustainability policies worldwide require upgrading electric power systems to meet rising demand driven by continued electrification of transportation. Forecasting electric-vehicle (EV) charging demand becomes essential for planning charging networks and grid infrastructure. This thesis addresses a research gap by focusing on workplace charging and evaluates statistical and state-of-the-art deep learning models for EV charging load prediction using the NREL workplace charging dataset (2017–2020). Tested models include ARIMA, SARIMA, XGBoost, LightGBM, RNN, LSTM, GRU, TFT, and N-BEATS, with TFT and N-BEATS achieving the lowest MAPE (18.9% and 19.5%).","Application of Machine Learning to Predict Electricity Demand from Electric Vehicles in Workplace Settings  \nby  \nJohnny Esteban  \nSubmitted to the Department of Electrical Engineering in partial fulfillment of the requirements for the degree of  \nMaster of Engineering in Energy Engineering  \nat the  \nUniversitat Politècnica de Catalunya  \nJuly 2023  \n© Johnny Esteban, MMXXIII. All rights reserved.  \nThe author hereby grants to the UPC permission to reproduce and to distribute publicly paper and electronic copies of this thesis document in whole or in part in any medium now known or hereafter created.  \nAuthor ................................................................  \nDepartment of Electrical Engineering July 3rd, 2023  \nCertified by ............................................................  \nMonica Aragüés Peñalba Lecturer  \nThesis Supervisor  \nAccepted by    \nEnrico Velo Garcia Degree Coordinator  \n2  \nApplication of Machine Learning to Predict Electricity Demand from Electric Vehicles in Workplace Settings  \nby  \nJohnny Esteban  \nSubmitted to the Department of Electrical Engineering on July 3rd, 2023, in partial fulfillment of the  \nrequirements for the degree of Master of Engineering in Energy Engineering  \nAbstract  \nAs sustainability-oriented policies begin to be implemented across the world, adapting the current electric power system (EPS) to meet the demands required by those policies is key to meeting emissions targets. Part of those policies includes the continued expansion of the electrification of national transportation systems. This electrification of transportation will require the vast expansion of electric vehicle (EV) usage as well as the charging networks that will give them power. The consequent growth in anticipated energy demand must be included in infrastructure planning. As a result, forecasting the charging demand of EVs will be a vital tool to plan for the development of EPS infrastructure. The identification of the best forecasting methods is a key field of research supporting this effort. This thesis analyzed several statistical models and state-of-the-art (SoA) deep learning (DL) machine learning models to determine relevant forecasting tools for predicting EV charging loads in the context of workplace charging. Workplace charging was identified as a gap in research, where fewer attempts to model EV demand at office buildings and places of work had been recorded. The data set chosen was the NREL workplace charging data set, which included daily charging load from 2017-2020 . The time series forecasting models tested include ARIMA, SARIMA, XGBoost, LightGBM, RNN, LSTM, GRU, TFT, and N-BEATS. A machine learning modelling pipeline was developed for each model. Results of modelling determined that the SoA DL models TFT and N-BEATS were the top performing models with a mean average percentage error (MAPE) score of 18.9% and 19 .5%, respectively, followed by XGBoost with an MAPE of 21 . 1% . From a residual error analysis, it was found that TFT poorly estimated peak consumption, but was able to more consistently predict the general trends, as compared to XGBoost and N-BEATS, which performed better with extreme fluctuations, but struggled with non-extreme values.  \nThesis Supervisor: Monica Aragüés Peñalba  \nTitle: Lecturer  \n4  \nAcknowledgments  \nI would like to say thank you to all my professors, classmates, and friends who helped me over the course of the past two years to grow and learn so much about a topic I care for deeply. Without their support I would not have been able to complete such an undertaking and I am truly grateful. It was such a pleasure to learn alongside you in class, projects, and most frighteningly, the exams.  \nEach of the members of my thesis committee has provided me extensive personal and professional guidance and taught me a great deal about both scientific research and life in general. A special recognition to my advisor, Dra. Monica Aragüés Peñalba, who helped me ","cbCaif687KGxPwQK","https://ap.wps.com/l/cbCaif687KGxPwQK","pdf",3365354,1,118,"English","en",105,"# Introduction\n## Context\n## Scope, Goals, and Research Question\n## Literature Review\n## Contribution to Literature\n## Structure of the Thesis\n## Planning\n# Mathematical Basis\n## Data Transformations\n## Error metrics\n## ARIMA Models\n## Deep Learning Algorithms\n## Decision Tree Algorithms","[{\"question\":\"Why is forecasting EV charging demand important in workplace settings?\",\"answer\":\"Workplace charging must be incorporated into infrastructure planning as electrification policies expand EV usage and charging networks, creating increased energy demand.\"},{\"question\":\"Which dataset and forecasting models were used in the thesis?\",\"answer\":\"The study uses the NREL workplace charging dataset covering daily loads from 2017 to 2020 and tests ARIMA, SARIMA, XGBoost, LightGBM, RNN, LSTM, GRU, TFT, and N-BEATS.\"},{\"question\":\"Which models performed best and how did their errors differ?\",\"answer\":\"TFT and N-BEATS were top performers with MAPE scores of about 18.9% and 19.5%. Residual analysis showed TFT estimated general trends more consistently but poorly captured peak consumption, while XGBoost and N-BEATS handled extreme fluctuations better but struggled with non-extreme values.\"}]","Application of Machine Learning to Predict Electricity Demand from Electric Vehicles in Workplace Settings - Master Thesis | PDF",1785893990,297,{"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},"application-of-machine-learning-to-predict-electricity-demand-from-electric-vehicles-in-workplace-settings-master-thesis","",{"@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/application-of-machine-learning-to-predict-electricity-demand-from-electric-vehicles-in-workplace-settings-master-thesis/124699/",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","2026-08-05",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 is forecasting EV charging demand important in workplace settings?","Question",{"text":75,"@type":76},"Workplace charging must be incorporated into infrastructure planning as electrification policies expand EV usage and charging networks, creating increased energy demand.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which dataset and forecasting models were used in the thesis?",{"text":80,"@type":76},"The study uses the NREL workplace charging dataset covering daily loads from 2017 to 2020 and tests ARIMA, SARIMA, XGBoost, LightGBM, RNN, LSTM, GRU, TFT, and N-BEATS.",{"name":82,"@type":73,"acceptedAnswer":83},"Which models performed best and how did their errors differ?",{"text":84,"@type":76},"TFT and N-BEATS were top performers with MAPE scores of about 18.9% and 19.5%. Residual analysis showed TFT estimated general trends more consistently but poorly captured peak consumption, while XGBoost and N-BEATS handled extreme fluctuations better but struggled with non-extreme values.","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"]