This dissertation studies a framework in support electric vehicle (EV) charging station expansion and management decisions. In the first part of the dissertation, we present mathematical model for designing and managing electric vehicle charging stations, considering both long-term planning decisions and short-term hourly operational decisions (e.g., number of batteries charged, discharged through Battery-to-Grid (B2G), stored, Vehicle-to-Grid (V2G), renewable, grid power usage) over a pre-specified planning horizon and under stochastic power demand. The model captures the non-linear load congestion effect that increases exponentially as the electricity consumed by plugged-in EVs approaches the capacity of the charging station and linearizes it. The study proposes a hybrid decomposition algorithm that utilizes a Sample Average Approximation and an enhanced Progressive Hedging algorithm (PHA) inside a Constraint Generation algorithmic framework to efficiently solve the proposed optimization model. A case study based on a road network of Washington, D.C. is presented to visualize and validate the modeling results. Computational experiments demonstrate the effectiveness of the proposed algorithm in solving the problem in a practical amount of time. Finding of the study include that incorporating the load congestion factor encourages the opening of large-sized charging stations, increases the number of stored batteries, and that higher congestion costs call for a decrease in the opening of new charging stations. The second part of the dissertation is dedicated to investigate the performance of a collaborative decision model to optimize electricity flow among commercial buildings, electric vehicle charging stations, and power grid under power demand uncertainty. A two-stage stochastic programming model is proposed to incorporate energy sharing and collaborative decisions among network entities with the aim of overall energy network cost minimization. We use San Francisco, California as a testing ground to visualize and validate the modeling results. Computational experiments draw managerial insights into how different key input parameters (e.g., grid power unavailability, power collaboration restriction) affect the overall energy network design and cost. Finally, a novel disruption prevention model is proposed for designing and managing EV charging stations with respect to both long-term planning and short-term operational decisions, over a pre-determined planning horizon and under a stochastic power demand. Long-term planning decisions determine the type, location, and time of established charging stations, while short-term operational decisions manage power resource utilization. A non-linear term is introduced into the model to prevent the evolution of excessive temperature on a power line under stochastic exogenous factors such as outside temperature and air velocity. Since the re- search problem is NP-hard, a Sample Average Approximation method enhanced with a Scenario Decomposition algorithm on the basis of Lagrangian Decomposition scheme is proposed to obtain a good-quality solution within a reasonable computational time. As a testing ground, the road network of Washington, D.C. is considered to visualize and validate the modeling results. The results of the analysis provide a number of managerial insights to help decision makers achieving a more reliable and cost-effective electricity supply network.

The transport industry is facing major changes in the transition from traditional diesel-powered vehicles to electrified vehicles. The transition to electric vehicles in the transport industry is necessary to reach the environmental goals of the Paris Agreement. Through research, data sharing between actors was identified as a potential factor that could be used in the development of the electromobility sector, but sufficient information on this subject was lacking. This led to the basis for the thesis project. The thesis project was carried out in collaboration with Volvo Group, hereby interchangeably called Volvo, to investigate how data sharing can be used to facilitate the transformation to electromobility in the transport industry. The purpose of the thesis was to: Investigate how Volvos Value Offering can be improved by mapping out potential actors in the electromobility eco-system and how they could benefit from shared data. The thesis was based on the three research questions: What values and offers can be created in the charging infrastructure industry through shared data and what challenges, risks and opportunities do this create for the stakeholders involved? What information gaps hinder the development of the electromobility market, in general and, more specifically, in relation to data sharing? And lastly, how can Volvo take advantage of business opportunities in the electromobility market, in general and, more specifically, in relation to data sharing? The methods used to answer these questions were media analysis, 19 in-depth interviews, and a workshop with Volvo. The media analysis resulted in a mapping of the involved stakeholders in the electromobility development industry, how data sharing is used today and the actors' stance on data sharing. The interviews were held with respondents from the energy industry, tech companies, researchers, haulage companies and the truck manufacturer Volvo Group. The interviews were organized using the Gioia method and resulted in six different global themes on electromobility and data sharing. Results from the media analysis and the interviews were compiled into three scenarios. These were then presented to Volvo in a workshop, where they described how they would act as a major truck manufacturer in each scenario respectively. After compiling the results from the three methods, the research questions could be answered. The first research question was answered by the fact that the transport industry has a low degree of data maturity. The reason being that there are several perceived risks among the actors regarding data sharing in the form of losing competitive advantages, increased risks of cyber-attacks and GDPR violations. Despite the low degree of data maturity, there were also new opportunities that could be identified with data sharing. The biggest identified opportunity in this thesis was that data sharing can accelerate the development and expansion of the charging infrastructure, if vehicle data and energy data can be shared between actors. The second research question was answered simply by the fact that due to the low data degree of maturity, very little data is shared at present. The biggest identified information gap was the “chicken and egg” situation in the industry, where energy actors are waiting for initiatives from the automotive industry before making any decisions, and vice versa. The third research question was answered by identifying that Volvo's greatest opportunities as truck manufacturers exist through collaborations with other companies to establish standards for data sharing and data selling, offering charging solutions for their electric trucks and, finally, logistics optimization services based on real-time data. As the three research questions were answered, the purpose of the study was therefore fulfilled. The initial scope of the thesis was expanded from focusing solely on Volvo's opportunities as a truck manufacturer, to include opportunities for actors in the entire electromobility industry such as energy companies, charging post companies, haulage companies and tech companies. The study concluded by showing that there are great potential business and optimization opportunities and societal benefits with data sharing in the EMOB industry if the actors are willing to collaborate to set standards and drive development together. / Transportindustrin står inför stora förändringar i omställningen från traditionella dieseldrivna fordon till elektrifierade fordon. Omställningen inom transportindustrin är nödvändig för att nå miljömålen inom Parisavtalet. Forskning visar att datadelning mellan aktörer är en potentiell faktor som skulle kunna användas inom utvecklingen av elektromobilitetssektorn, men tillräckligt med information om detta område saknas. Detta blev grunden för examensarbetet. Examensarbetet genomfördes i samarbete med Volvo Group, hädan efter kallat Volvo, för att undersöka hur datadelning kan användas för att underlätta elektromobilitetsomvandlingen inom transportbranschen. Syftet med arbetet var att: Undersöka hur Volvos värdeerbjudanden kan förbättras genom att kartlägga potentiella aktörer i ekosystemet för elektromobilitet och hur de kan dra nytta av delade data. Arbetet utgick ifrån de tre forskningsfrågorna: Vilka värden och erbjudanden kan skapas inom laddinfrastruktur branschen genom delade data, vilka utmaningar, risker och möjligheter skapar detta för de inblandade intressenterna? Vilka informationsluckor hindrar utvecklingen av elektromobilitetsmarknaden, generellt och, mer specifikt, i relation till datadelning? Och slutligen, hur kan Volvo ta vara på affärsmöjligheter inom elektromobilitetsmarknaden, generellt och, mer specifikt, i relation till datadelning? Metoderna som användes för att besvara dessa frågor var mediaanalys, 19 djupintervjuer, samt en workshop med Volvo Group. Mediaanalysen resulterade i en kartläggning av drivande aktörer inom elektromobilitetsbranschen, hur datadelning används i dagsläget och aktörernas inställning till datadelning. Intervjuerna hölls med respondenter från energibranschen, techbolag, forskare, åkerier och Volvo Group. Intervjuerna organiserades med Gioia metoden och resulterade i sex olika globala teman om elektromobilitet och datadelning. Resultat från mediaanalysen och intervjuerna sammanställdes i tre scenarion. Dessa presenterades för Volvo i en workshop där de fick resonera hur de skulle agera som en stor lastbilstillverkare i respektive scenario. Efter sammanställning av resultaten från de tre metoderna kunde forskningsfrågorna besvaras. Den första forskningsfrågan besvarades med att transportbranschen i sig har låg datamognadsgrad. Det eftersom det fanns flera uppfattade risker hos aktörerna kring datadelning i form av förlorade konkurrensfördelar, ökade risker för cyberattacker och GDPR överträdelser. Trots den låga datamognadsgraden så finns det nya möjligheter med datadelning, där den största identifierade möjligheten i detta arbete är att datadelning kan påskynda utvecklingen och utbyggnaden av laddinfrastrukturen om fordonsdata och energidata kan delas mellan aktörer. Den andra forskningsfrågan besvarades med att på grund av den låg data mognadsgraden så delas väldigt lite data i dagsläget. Det största identifierade informationsluckan var “hönan eller ägget” situationen i branschen, där energiaktörer väntar på initiativ från fordonsbranschen innan de tar några beslut, och vice versa. Den tredje forskningsfrågan besvarades med att Volvos största möjligheter som lastbilstillverkare finns genom samarbeten med andra företag för att etablera standarder för datadelning och dataförsäljning, erbjuda laddlösningar till sina elektriska fordon och slutligen logistikoptimeringstjänster baserade på realtidsdata. I och med att de tre forskningsfrågorna besvarades, uppfylldes därmed syftet med studien. Omfattningen av arbetet expanderades dock från att enbart fokusera på Volvos möjligheter som lastbilstillverkare, till att omfatta aktörer inom hela elektromobilitets branschen så som energibolag, laddstolpsbolag, åkerier och techbolag. Studien visar att det finns stora potentiella affärs och optimeringsmöjligheter och samhällsnytta med datadelning inom elektromobilitetsbranschen om aktörer är villiga att samarbeta för att sätta standarder och driva utvecklingen tillsammans.

Is battery energy storage a feasible solution for lowering the operational costs of electric vehicle fast charging and reducing its impact on local grids? The thesis project aims at answering this question for the Swedish scenario. The proposed solution (fast charging station coupled with storage) is modelled in MATLAB, and its performance is tested in the framework provided by Swedish regulation and electricity tariff structure. The analysis is centred on the economic performance of the system. Its cost-effectiveness is assessed by means of an optimisation algorithm, designed for delivering the optimal control strategy and the required equipment sizing. A mixed-integer linear programming (MILP) formulation is utilised. The configuration and operative costs of conventional fast charging stations are used as a benchmark for the output of the optimisation. Sensitivity analysis is conducted on most relevant parameters: charging load, battery price and tariff structure. The modelling of the charging demand is based on statistics from currently implemented 50 kW DC chargers in Sweden. Overall, results show that with current figures the system may be an economically viable solution for both reducing costs and lowering the impact on the local distribution grid, at least during peak-period pricing. However, sensitivity analysis illustrates how system design and performance are highly dependent on input parameters. Among these, electricity tariff was identified as the most important. Consequently, detailed discussion on the influence of this parameter is conducted. Finally, the study shows how the system is in line with most recent directives proposed by the European Commission.

The increasing number of electric vehicles (EVs) in the transport sector has rendered the conventional fuel-based vehicles obsolete along with the fuel filling stations. With the growth in EVs, there has been an increase in the public charging infrastructure with fast charging equipment being used to charge the EVs in least possible time and also address the issue of ‘range anxiety’ among the EV owners. Many countries like South Korea and Germany has seen policies being implemented to install fast chargers for EVs in existing fuel filling stations. This study aims conduct a techno-economic feasibility to analyse the potential of implementing Electric Vehicle Supply Equipment (EVSE) with fast charging capacity into existing fuel filling stations. The potential of using solar photovoltaic system (PV) and battery storage systems (BESS) to reduce the load from the grid is also explored. Scenarios are developed considering different configurations of the EVSE, PV and BESS and an in-depth economic analysis is conducted to analyse the economic feasibility of the configurations. The impact on the electricity grid is also analysed in this thesis by conducting a load flow analysis on the CIGRE Low voltage network for Europe using Python.The proposed design enables selection of techno-economically feasible configurations of EVSE, BESS and PV. The results of the design are explained with the UK as a case study. It is observed that the configurations with 3 EVSE, BESS and 8 hours and the configuration with 3 EVSE, 1 BESS and 1 PV system for 8 hours of operation are economically viable. The proposed design shows that though the connection cost is the dominant factor affecting the feasibility, use of BESS with or without PV can reduce the connection cost by almost 90% depending on the number of BESS. Load flow analysis is conducted for the different configurations of EVSE, BESS and PV on the CIGRE LV network on Pandapower in Python. The results indicate that the existing network needs to be reinforced to facilitate the connection of EV fast chargers into the grid. Upgrading the network cables and increasing the slack voltage to a value of 1.05 or 1.1 Volts per unit, are the two strategies that have been suggested in this study to prevent any undervoltage that may occur as a result of connecting the EVSE to the electricity grid. The simulations conducted for the two strategies highlight that by implementing these strategies into the electricity grid network, the undervoltage issues in the transmission network can be mitigated. / Det ökande antalet elfordon inom transportsektorn har gjort de konventionella bränslebaserade fordonen föråldrade tillsammans med bränslepåfyllningsstationerna. Med ökningen av elbilar har det skett en ökning av den offentliga laddningsinfrastrukturen med snabbladdningsutrustning som används för att ladda elbilarna på åtminstone möjlig tid och också ta itu med frågan om ’range anxiety’ bland elägare. Många länder som Sydkorea och Tyskland har sett politik införas för att installera snabbladdare för elbilar i befintliga bensinstationer. Denna studie syftar till att genomföra en teknisk-ekonomisk genomförbarhet för att analysera potentialen för att implementera elfordonstillförselutrustning (EVSE) med snabb laddningskapacitet i befintliga bensinstationer. Potentialen med att använda solcellssystem (PV) och batterilagringssystem (BESS) för att minska belastningen från nätet undersöks också. Scenarier utvecklas med beaktande av olika konfigurationer av EVSE, PV och BESS och en djupgående ekonomisk analys genomförs för att analysera konfigurationernas ekonomiska genomförbarhet. Effekten på elnätet analyseras också i denna avhandling genom att genomföra en belastningsflödesanalys på CIGRE lågspänningsnät för Europa med Python.Den föreslagna designen möjliggör val av tekno-ekonomiskt genomförbara konfigurationer av EVSE, BESS och PV. Resultaten av designen förklaras med Storbritannien som en fallstudie. Det observeras att konfigurationerna med 3 EVSE, BESS och 8 timmar och konfigurationen med 3 EVSE, 1 BESS och 1 PV-system för 8 timmars drift är ekonomiskt lönsamma. Den föreslagna designen visar att även om anslutningskostnaden är den dominerande faktorn som påverkar genomförbarheten, kan användning av BESS med eller utan solceller minska anslutningskostnaden med nästan 90% beroende på antalet BESS. Lastflödesanalys utförs för de olika konfigurationerna av EVSE, BESS och PV på CIGRE LV-nätverket på Pandapower i Python. Resultaten visar att det befintliga nätverket måste förstärkas för att underlätta anslutningen av EV-snabbladdare till nätet. Uppgradering av nätverkskablarna och ökning av spänningen till 1,05 eller 1,1 volt per enhet är de två strategier som har föreslagits i denna studie för att förhindra underspänning som kan uppstå till följd av att EVSE ansluts till elnätet. Simuleringarna för de två strategierna lyfter fram att genom att implementera dessa strategier i elnätet kan underspänningsfrågorna i överföringsnätet mildras.

La forte dépendance du pétrole et les contraintes écologiques et environnementales obligent de nombreux constructeurs automobiles à développer des programmes de recherche importants dans le développement des véhicules électriques (VEs) et des infrastructures associées. Les batteries embarquées dans les VEs peuvent être rechargées par le réseau électrique ou par une autre source d'énergie renouvelable. Dans ce contexte, cette thèse a pour but d'étudier un système de recharge de VEs à partir de l'énergie solaire. Pour ce faire, une étude de dimensionnement du système a été proposée afin d'évaluer les besoins énergétiques des VEs et déterminer les quantités d'énergie nécessaires pour les propulser. Les éléments principaux du système ont été dimensionnés. Il s'agit de panneaux photovoltaïques, de la batterie de stockage (au plomb), et de la batterie de traction (Li-ion). A partir de cette étude de dimensionnement, nous avons pu déterminer la puissance et la quantité d'énergienécessaire pour propulser un VE compte tenu de ses caractéristiques. Pour étudier le comportement des différents éléments du système et analyser leur adéquation avec le processus de recharge, une étude de modélisation a été réalisée, et chaque élément est représenté par un modèle mathématique. L'analyse et la comparaison des résultats obtenus (résultats de simulation et résultats expérimentaux) ont permis de valider les modèles proposés. De même, cette étude de modélisation a permis de valider le choix des composants du système de recharge. Dans le but de gérer efficacement les processus et les planifications de recharge, une étude d'optimisation d'affectation de VEs aux stations de recharge a été effectuée. Pour ce faire, le problème a été formalisé par un programme linéaire avec un objectif à atteindre et des contraintes, liées aux caractéristiques des VEs, des stations de recharge et des conditions de circulation, à satisfaire. Cette approche a permis d'affecter, de manière adéquate et optimale, les VEs aux stations de recharge. / The strong dependence on oil and ecological and environmental constraints force many car manufacturers to develop new research programs for the promotion of electric vehicles (EVs) and associated infrastructures. The embedded batteries into the EVs can be charged by the electrical network or by another source of renewable energy. In this context, the work presented in this thesis aims to study a system of EVs charging using solar energy through photovoltaic panels. To do so, a sizing study of the system has been proposed in order to evaluate the energy needs for an EV and determine the quantity of required energy for its propel. The key elements of the system have been sized: photovoltaic panels, storage battery (Lead-acid) and traction battery (Li-ion). From this sizing study and considering the EV characteristics, we determined the energy quantity required for itspropel. With the aim to study the behavior of each system component and analyze its adequacy with the charging process, a modeling study was conducted, and each element is represented by a mathematical model. The performed analysis and comparison of obtained results (simulation results and experimental results) allowed us to validate the developed models. In addition, this modeling study, allowed the validation of the choice of all components of charging system. In fact, the problem was formalized by a linear program with the aim to assign each EV to an adequate charging station. The assignment takes into account various constraints and characteristics of EVs, as well as those of charging stations, traffic conditions, interest points of drivers, etc. The proposed approach allowed to assign adequately and optimally EVs to charging stations while satisfying all constraints.

Submitted by Igoor Morro Mello (igoor.mello@unesp.br) on 2018-09-03T21:44:57Z No. of bitstreams: 1 dissertação de mestrado 2018 - Igoor Morro Mello.pdf: 10593338 bytes, checksum: 0253d1ce16d1ea2f863390cd1d68fd1e (MD5) / Approved for entry into archive by Cristina Alexandra de Godoy null (cristina@adm.feis.unesp.br) on 2018-09-04T17:09:34Z (GMT) No. of bitstreams: 1 mello_im_me_ilha.pdf: 10593338 bytes, checksum: 0253d1ce16d1ea2f863390cd1d68fd1e (MD5) / Made available in DSpace on 2018-09-04T17:09:34Z (GMT). No. of bitstreams: 1 mello_im_me_ilha.pdf: 10593338 bytes, checksum: 0253d1ce16d1ea2f863390cd1d68fd1e (MD5) Previous issue date: 2018-08-24 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Nos últimos anos, o uso dos veículos elétricos nas zonas urbanas tem se intensificado. Como política para o aumento na penetração de veículos elétricos e reduzir a poluição do ar, os táxis elétricos vem sendo introduzidos nos sistemas de transporte. Eles necessitam de atenção especial devido aos seus diferentes padrões de condução. Em contraste com os veículos elétricos privados, que podem ser recarregados por um longo período, táxis elétricos necessitam de recarga em um curto período de tempo devido a sua constante operação. Portanto, estações de recarga rápida são necessárias para receber a demanda de recarga dos táxis elétricos e devem estar localizadas em locais estratégicos. Além disso, uma análise deve ser realizada para a conexão destas estações com a rede elétrica. Para melhorar sua alocação e conectividade, este trabalho apresenta uma metodologia para auxiliar na tomada de decisão da instalação de estações de recarga rápida considerando como critérios: locais com maior fluxo de táxis elétricos e baixo nível de carga nas baterias, espaço físico disponível para realizar o carregamento e funções de custo para a conexão das estações de recarga. Os resultados da proposta são mapas com a localização das estações de recarga rápida e análise dos locais de menor custo para a conexão com a rede elétrica. A metodologia é testada em uma cidade de médio porte no Brasil, mostrando a importância dos mapas e funções de custo na tomada de decisão. A proposta é comparada com outras metodologias, mostrando que esta metodologia proposta considera diferentes critérios e cria uma melhor distribuição espacial para as estações de recarga, dando melhores opções aos donos dos táxis elétricos. / In recent years, the use of electric vehicles in urban zones has been intensified. As a policy of increasing the penetration of electric vehicles and reducing air pollution, electric taxis have been introduced into transportation systems. They need special attention because of its different driving patterns. In contrast to private electric vehicles, which can be recharged for a long period, electric taxis need to recharge only for a short time due to their constant operation. Therefore, fast charging stations are required to meet the demand for recharging electric taxis and should be located at strategic places. In addition, an analysis must be performed to connect these stations in the electric network. To improve their allocation and connectivity, this work presents a methodology to help in decision making for installing fast charging stations, considering as criteria: locations with greater flow of electric taxis and low level of state of charge, the available physical space to carry out their recharge and cost functions for the connection of charging stations. The result of the proposal is a map with the location of fast charging stations and analysis of the lowest cost places for connection to the network. The methodology is tested in a medium-sized city in Brazil, showing the importance of this map and cost functions in decision making. The proposal is compared with another methodology, showing that the proposed method considers different criteria and creates a better spatial distribution of charging stations giving better options to the owners of electric taxis. / CAPES: 1667419

Orientador: Antonio Padilha Feltrin / Abstract: In recent years, the use of electric vehicles in urban zones has been intensified. As a policy of increasing the penetration of electric vehicles and reducing air pollution, electric taxis have been introduced into transportation systems. They need special attention because of its different driving patterns. In contrast to private electric vehicles, which can be recharged for a long period, electric taxis need to recharge only for a short time due to their constant operation. Therefore, fast charging stations are required to meet the demand for recharging electric taxis and should be located at strategic places. In addition, an analysis must be performed to connect these stations in the electric network. To improve their allocation and connectivity, this work presents a methodology to help in decision making for installing fast charging stations, considering as criteria: locations with greater flow of electric taxis and low level of state of charge, the available physical space to carry out their recharge and cost functions for the connection of charging stations. The result of the proposal is a map with the location of fast charging stations and analysis of the lowest cost places for connection to the network. The methodology is tested in a medium-sized city in Brazil, showing the importance of this map and cost functions in decision making. The proposal is compared with another methodology, showing that the proposed method considers different criteria and creates a better s... (Complete abstract click electronic access below) / Resumo: Nos últimos anos, o uso dos veículos elétricos nas zonas urbanas tem se intensificado. Como política para o aumento na penetração de veículos elétricos e reduzir a poluição do ar, os táxis elétricos vem sendo introduzidos nos sistemas de transporte. Eles necessitam de atenção especial devido aos seus diferentes padrões de condução. Em contraste com os veículos elétricos privados, que podem ser recarregados por um longo período, táxis elétricos necessitam de recarga em um curto período de tempo devido a sua constante operação. Portanto, estações de recarga rápida são necessárias para receber a demanda de recarga dos táxis elétricos e devem estar localizadas em locais estratégicos. Além disso, uma análise deve ser realizada para a conexão destas estações com a rede elétrica. Para melhorar sua alocação e conectividade, este trabalho apresenta uma metodologia para auxiliar na tomada de decisão da instalação de estações de recarga rápida considerando como critérios: locais com maior fluxo de táxis elétricos e baixo nível de carga nas baterias, espaço físico disponível para realizar o carregamento e funções de custo para a conexão das estações de recarga. Os resultados da proposta são mapas com a localização das estações de recarga rápida e análise dos locais de menor custo para a conexão com a rede elétrica. A metodologia é testada em uma cidade de médio porte no Brasil, mostrando a importância dos mapas e funções de custo na tomada de decisão. A proposta é comparada com outras me... (Resumo completo, clicar acesso eletrônico abaixo) / Mestre

No description available.

Battery electric vehicles and fuel cell vehicles, i.e. hydrogen vehicles, are promising alternatives to internal combustion engine vehicles to reduce GHG emissions from the transport sector. EV charging and hydrogen refuelling infrastructure is crucial to the deployment of alternative fuels in transport. Although several studies have analyzed electric public buses infrastructure, fuel cell buses have not been the target of such extensive analyses. Additionally, there is a gap in the literature regarding the comparison of infrastructure for these two types of vehicles and their cost and refuelling schedule differences. The study aims to conduct a techno-economic analysis of electricity versus hydrogen refuelling infrastructure to decarbonize public buses, using renewable sources to produce renewable electricity and green hydrogen. The outcome is a proposed system design regarding the size of the refuelling station, storage system capacity, renewable energy capacity, on-site hydrogen production system size, and the optimized refuelling schedule. The system is modelled to minimize the overall system cost while maintaining the current bus service level. The impact of electricity market prices, demand charges and varying bus energy demand in the optimal system configuration and schedule is also addressed. Scenarios are developed to study different levels of new installed renewable capacity integration and how these affect the cost, bus refuelling schedules and infrastructure design. The mixed-integer linear programming problem was modelled using Python. The model is applied to the case study of one bus line in Umeå. One terminal station was chosen to place the refuelling stations. The results show that the most economical option is electrifying the line with electricity supply only from the grid. For scenarios with additional renewable energy capacity installed, the option with 50% integration of new installed capacity is the most economically viable. In both these cases, there is no installation of BESS at the charging station. Electric buses infrastructure is cheaper than hydrogen infrastructure in all scenarios, but these values converge as renewable energy integration increases. For hydrogen infrastructure, the scenario with 50% renewable energy integration is the least costly. Although electric bus infrastructure is more economical than hydrogen infrastructure, hydrogen buses present advantages in terms of significantly higher range and thus higher flexibility for refuelling. Therefore, in the decision-making process to replace a fossil fuel bus line with an alternative fuel bus line, one must consider the multi-dimensional level of the different options. / Batterielektriska fordon och bränslecellsfordon, dvs. vätgasfordon, är lovande alternativ till fordon med förbränningsmotorer för att minska växthusgasutsläppen från transportsektorn. Infrastruktur för laddning av elfordon och tankning av vätgas är avgörande för att alternativa bränslen ska kunna användas inom transportsektorn. Även om flera studier har analyserat infrastrukturen för offentliga elbussar har bränslecellsbussar inte varit föremål för sådana omfattande analyser. Dessutom finns det en lucka i litteraturen när det gäller jämförelsen av infrastruktur för dessa två typer av fordon och deras skillnader i fråga om kostnader och tankningsschema. Syftet med studien är att genomföra en teknisk-ekonomisk analys av infrastruktur för tankning av el respektive vätgas för att avkarbonisera offentliga bussar, med hjälp av förnybara källor för att producera förnybar el och grön vätgas. Resultatet är ett förslag till systemutformning med avseende på tankstationens storlek, lagringssystemets kapacitet, kapaciteten för förnybar energi, storleken på systemet för vätgasproduktion på plats och det optimerade tankningsschemat. Systemet modelleras för att minimera den totala systemkostnaden samtidigt som den nuvarande service nivån förbussarna bibehålls. Effekten av elmarknadspriser, efterfrågeavgifter och varierande energiefterfrågan från bussarna på den optimala systemkonfigurationen och schemat behandlas också. Scenarier utvecklas för att studera olika nivåer av nyinstallerad förnybar kapacitet och hur  dessa påverkar kostnaden, bussarnas tankningsscheman och infrastrukturens utformning. Det linjära programmeringsproblemet med blandade heltal modellerades med hjälp av Python. Modellen tillämpas på fallstudien av en busslinje i Umeå. En ändstation valdes ut för att placera tankstationerna. Resultaten visar att det mest ekonomiska alternativet är att elektrifiera linjen med elförsörjning endast från nätet. För scenarier med ytterligare installerad kapacitet för förnybar energi är alternativet med 50 % integrering av ny installerad kapacitet det mest ekonomiskt lönsamma. I båda dessa fall finns det ingen installation av BESS vid laddningsstationen. Infrastrukturen för elbussar är billigare än infrastrukturen för vätgas i alla scenarier, men dessa värden närmar sig varandra när integrationen av förnybar energi ökar. När det gäller vätgasinfrastruktur är scenariot med 50 % integrering av förnybar energi det minst kostsamma. Även om infrastrukturen för elbussar är billigare än infrastrukturen för vätgasbussar har vätgasbussar fördelar i form av betydligt större räckvidd och därmed större flexibilitet när det gäller tankning. I beslutsprocessen för att ersätta en busslinje med fossila bränslen med en busslinje med alternativa bränslen måste man därför ta hänsyn till de olika alternativens flerdimensionella nivå.

Dans des régions avec beaucoup d'électricité renouvelable, comme le Québec, une augmentation du nombre de Véhicules Électriques (VE) peut réduire les gaz à effet de serre. Par contre, l'autonomie réduite des VE et la présence limitée d'infrastructure publique pour recharger les véhicules peuvent contribuer à un phénomène nommé anxiété de l'autonomie, où les usagers n'achètent pas des VE par peur qu'ils tombent en panne. On peut alors planifier l'emplacement de l'infrastructure publique de recharge de manière stratégique pour combattre cet effet, menant alors à un taux d'adoption plus élevé pour les VE. En utilisant des modèles de choix discret, nous incorporons des modèles économétriques de demande avancés capturant les préférences hétérogènes des usagers à l'intérieur de l'optimisation. En particulier, comme nous le démontrerons, ceci permet l'inclusion de nouveaux facteurs importants, tels qu'une disponibilité de la recharge à domicile et des effets de distance plus granulaire. Par contre, la méthodologie existante pour ce processus crée un modèle de programmation linéaire mixte en nombres entiers qui ne peut pas être résolue, même pour des instances de taille modeste. Nous développons alors une reformulation efficace en problème de couverture maximum qui, comme nous le démontrerons, permet une amélioration de plusieurs ordres de magnitude pour le temps de calcul. Bien que cette reformulation dans un problème de couverture maximum améliore grandement la capacité à résoudre le modèle, celui-ci demeure difficile à résoudre pour des problèmes de grandes tailles, nécessitant des heuristiques pour obtenir des solutions de haute qualité. Nous développons alors deux méthodes de décomposition de Benders spécialisées pour cette application. La première est une méthode de décomposition de Benders accélérée, qui se spécialise à réduire l'écart d'optimalité et à la résolution de problèmes de petite taille ou de taille modeste. La deuxième approche rajoute un branchement local à la méthode de décomposition de Benders accélérée, qui sacrifie de l'efficacité lors de la résolution de problèmes de plus petite taille pour une capacité augmentée afin d'obtenir des solutions réalisables de haute qualité. Finalement, nous présentons une méthode pour dériver des valeurs de paramètres autrement difficiles à obtenir pour le modèle de choix discrets dans le modèle d'optimisation. Ces paramètres dictent les effets de l'infrastructure publique de recharge sur l'adoption des VE. Pour ce processus, nous regardons les facteurs qui encouragent les usagers courants des VE à utiliser l'infrastructure existante. De manière plus précise, nous utilisons des données de recharge réelles de la ville de Montréal (Québec) pour estimer les impacts des caractéristiques des stations, tels que la distance des usagers, le nombre de bornes de recharge, et les installations à proximité. Différents types d'infrastructure sont considérés, de manière parallèle avec des modèles de choix discrets qui peuvent tenir compte de plusieurs observations pour chaque individu. Les contributions de cette thèse sont plus générales que simplement l'adoption de VE, étant applicable, par exemple, au problème de capture maximum, au problème de couverture maximum à multiples périodes, et à la prédiction de la station de recharge choisie par les conducteurs de VE. / In areas with large amounts of clean renewable electricity, such as Quebec, an increase to the number of electric vehicles (EVs) can reduce greenhouse gas emissions. However, the reduced range of EVs and the limited public charging infrastructure can contribute to a phenomenon known as range anxiety, where users do not purchase EVs out of concern they run out of charge while driving. We can strategically optimise the placement of public EV charging infrastructure to combat this effect, thus leading to increased EV adoption. By utilising discrete choice models, we incorporate advanced econometric demand models capturing heterogeneous user preferences within the optimisation framework. In particular, as we demonstrate, this allows for the inclusion of new, important attributes, such as a more granular home charging availability and a continuous degradation of quality based on the distance. However, existing methodologies for this optimisation framework result in a mixed-integer linear program which cannot be solved for even moderately sized instances. We thus develop an efficient reformulation into a maximum covering location problem which, as we show experimentally, allows for multiple orders of magnitude of improved solving time. While the reformulation into a maximum covering location problem greatly improves the solving capabilities for the model, it remains intractable for large-scale instances, relying on heuristics to obtain high-quality solutions. As such, we then develop two specialised Benders decomposition methods for this application. The first is an accelerated branch-and-Benders-cut method, which excels at solving small or medium-scale instances and at decreasing the optimality gap. The second approach incorporates a local branching scheme to the accelerated branch-and-Benders-cut method, which sacrifices some efficiency in solving smaller instances for an increased ability to obtain high-quality feasible solutions. Finally, we discuss a method for deriving difficult-to-obtain parameter values of the discrete choice model in the optimisation framework. These parameter values dictate the effects of the public charging infrastructure on EV adoption and, as such, play a crucial role in the optimisation model. For this process, we investigate the attributes that encourage current EV owners to utilise existing infrastructure. More specifically, we use real charging session data from the city of Montreal (Quebec) to determine the impacts of station characteristics such as the distance to the users, the number of outlets, and the nearby amenities. Different types of charging infrastructure are considered alongside discrete choice models which take into account multiple observations from individual users. The contributions of this thesis lie more broadly than simply EV adoption, being applicable to, e.g., the maximum capture problem, the multi-period maximum covering location problem, and the prediction of the charging station selected by EV drivers.