Thesis (MTech(Electrical Engineering)) -- Cape Peninsula University of Technology, 2014 / Today the building energy modeling industry is facing a number of challenges, the advanced programs or methods developed for building energy modeling, are very technical and complex to be used, especially for earlier designs, and the easy programs or methods are not accurate. Moreover, more than a hundred programs developed for energy modeling, have been used in the same building, but most of the time the results differed by about 30%. That is why this thesis has developed a new building energy model in System Modeling Language (SysML), in order to meet, at the same time, the accuracy and the simplicity to be used for future and existing buildings. In this thesis, SysML has been used to develop an energy model and to set up an automation system to the existing building. SysML can do more than simulations, but this thesis is limited to only the simulations steps by using easy applications of SysML and fewer diagrams which could develop in a complete building energy model. SysML is the extension of Unified modeling Language (UML), which uses fewer diagrams than UML. SysML is simple, open and more flexible to be used in any Engineering System. The previous chapter describes SysML and gives the overview and the platform of SysML. The simulations of SysML in this project have been developed through Enterprise Architect and Mat lab software. The inputs used to simulate the program are the parameters of the existing building chosen for modeling that is a student residential building complex located in Stellenbosch, Western Cape in South Africa. Automation system program used in this thesis was based on the norms and building standards of South Africa, renewable energy and the requirements of the buildings’ occupants, in order to meet energy efficiency and safety of the occupants.

Le secteur du bâtiment est considéré comme un gros consommateur d'énergie et une source de pollution majeure parmi tous les secteurs économiques. Il représente entre 16 et 50 pour cent des consommations nationales d'énergie. La réduction de ces consommations et des émissions est donc une étape importante vers un développement durable. Récemment, la transition vers la construction des bâtiments à faible consommation d’énergie a conduit à de nouvelles exigences en matière de performance et de durabilité, et ainsi encore complexifié le processus de conception des bâtiments. Le comportement des occupants est maintenant considéré comme un facteur déterminant de la performance énergétique d’un bâtiment, particulièrement dans le cas des bâtiments basse consommation (BBC). Pourtant, les outils de simulation utilisés dans l'industrie des bâtiments ne sont pas aujourd'hui en mesure de fournir des estimations fiables de la demande d'énergie des occupants. Par conséquent, les experts en énergie et bâtiments portent une grande attention à développer des méthodes plus précises pour la modélisation et la prévision de l’influence des occupants sur la performance du bâtiment. Ces modèles doivent pouvoir fournir des estimations plus précises des consommations d’énergie et évaluer la variabilité de ces consommations. En conséquence, l’objectif visé est de permettre aux experts en construction d’améliorer leurs solutions techniques, améliorer la performance de leurs services, et promouvoir des incitations mieux ciblées vers les usagers afin de réduire leurs consommations énergétiques. L'objectif de cette thèse est de proposer un modèle pour estimer la consommation d'énergie liée aux comportements des occupants de bâtiments résidentiels, en prenant en compte la variabilité des modes de consommation au travers de la diversité des profils socio-démographiques et économiques des occupants. Une approche stochastique basée sur la notion d’activité est donc adoptée. Avec ce modèle, la consommation d'énergie d'un ménage est estimée en additionnant la consommation d'énergie des différentes activités domestiques (comme faire la cuisine, le lavage du linge, etc.). La nature stochastique du modèle est due aux relations probabilistes établies entre les attributs des ménages d'une part (type de ménage, nombre d'occupants, etc.) et la possession des équipements domestiques, les caractéristiques des appareils, leur puissance, et les quantités d'activité d’autre part. Afin d'établir ces relations stochastiques, un nombre suffisant d'attributs est pris en compte pour caractériser un ménage. Le modèle proposé a été appliqué pour deux activités domestiques, à savoir regarder la télévision et laver le linge. Des simulations de Monte Carlo sont effectuées pour fournir des estimations de consommation d'énergie pour ces deux activités dans trois cas de figure : pour un ménage spécifique, pour des ménages générés aléatoirement avec des contraintes sur leurs attributs, et pour des ménages totalement aléatoires représentatifs de la population française. Une comparaison entre les résultats de la simulation de modèle d’une part et des données de consommation d'énergie réelle d’autre part, a permis de valider le modèle pour les deux activités considérées. Un cadre de généralisation du modèle pour d'autres activités domestiques a été introduit, et sa possible intégration dans le processus de conception des bâtiments a été discutée et illustrée au travers d’un certain nombre d’exemples. / Résumé en Anglais : The building sector is considered as a major energy consumer and pollution source among all economic sectors. It accounts for important shares, ranging between 16 and 50 percent, of national energy consumption worldwide. Reducing these consumptions and emissions is thus an important step towards sustainable development. Recently, the shift towards constructing low-consuming and nearly zero-energy buildings lead to further requirements with regard to performance and sustainability, and thus caused the design process of buildings to be more complex. Occupants’ behavior is now considered as a key determinant of building’s energy performance especially in the case of green buildings. Yet, energy simulation tools used in buildings industry nowadays are not capable of providing accurate estimations of occupant-related energy demands. Therefore, buildings and energy experts are devoting considerable efforts on developing more precise methods for modeling and forecasting occupants influence on whole building performance. Such models can provide accurate energy estimates and can assess future consumption variability. Consequently, building experts may improve their technical solutions, ameliorate their service performances, and promote targeted incentives. The objective of this dissertation is to propose a model for forecasting occupant-related energy consumption in residential buildings, while accounting for variability in consumption patterns due to diversity in occupants’ socio-demographic and economic profiles. A stochastic activity-based approach is thus adopted. By activity-based, it means that energy consumption of a household is estimated by summing up the energy use of different activities performed (such as cooking, washing clothes, etc.). The stochastic nature of the model is due to the probabilistic mapping established between household attributes from one side (household type, number of occupants, etc.) and the corresponding appliance ownership, appliance characteristics and power rating, and activity quantities from the other side. In order to establish these stochastic relations, a fairly sufficient number of households’ characterizing attributes is taken into account. The proposed model is applied for two domestic activities, namely watching TV and washing laundry. Three types of Monte Carlo simulations are performed to provide energy estimates for these two activities: for a given specified household, for randomly generated households with constraints, and for totally random population-wise households. A comparison between model’s simulation results and real measured energy consumption data enables validating the model for the two considered activities. A generalization framework of the modeling approach for other domestic activities is sketched, and its possible integration into buildings design process is discussed and illustrated through a number of examples.

A production industry with many robots working 24 hours a day, 7 days a week consumes a lot of energy. Industries aim to reduce the energy consumed per machine so as to support their financial budgets and also to be a more sustainable, energy efficient entity. Energy models can be used to predict the energy consumed by robot(s) for optimising the input parameters which determine robot motion and task execution. This work presents an ener-gy model to predict the energy consumption of 2D belt robots used for press line tending. Based on the components' specifications and the trajectory, an estimation of the energy consumption is computed. As part of this work, the proposed energy model is formulated, implemented in MATLAB and experimentally validated. The energy model is further used to investigate the effect of tool weight on energy consumption which includes predicting potential energy reductions achieved by reducing the weight of the gripper tools. Further, investigation of potential energy savings which can be achieved when mechanical brakes are used when the robot is idle is also presented. This illustrates the purpose and usefulness of the proposed energy model.

Wireless Sensor Networks (WSNs) have attracted growing interest from both realcustomers and the scientific community in the recent years due to their powerfulcapabilities and varied applications. Each wireless sensor node relays data to thebase station (BS) directly in the direct communication protocol. However, there aresome applications that require some of the nodes to be placed at vast distances fromthe BS, which leads to rapid reduction in the total energy of all nodes. Since energyconsumption in sensor networks is critical, many new protocols have beendeveloped to reduce losses. The Low-energy adaptive clustering hierarchy(LEACH) is one of the most popular protocols that increase network lifetime byreducing losses. In this report, I survey and compare some algorithms and propose a new enhancedalgorithm for the LEACH protocol, thus saving energy and prolonging the networklifespan. MATLAB is used for simulation.

The phase transition in the number partitioning problem (NPP), i.e., the transition from a region in the space of control parameters in which almost all instances have many solutions to a region in which almost all instances have no solution, is investigated by examining the energy landscape of this classic optimization problem. This is achieved by coding the information about the minimum energy paths connecting pairs of minima into a tree structure, termed a barrier tree, the leaves and internal nodes of which represent, respectively, the minima and the lowest energy saddles connecting those minima. Here we apply several measures of shape (balance and symmetry) as well as of branch lengths (barrier heights) to the barrier trees that result from the landscape of the NPP, aiming at identifying traces of the easy-hard transition. We find that it is not possible to tell the easy regime from the hard one by visual inspection of the trees or by measuring the barrier heights. Only the difficulty measure, given by the maximum value of the ratio between the barrier height and the energy surplus of local minima, succeeded in detecting traces of the phase transition in the tree. In addition, we show that the barrier trees associated with the NPP are very similar to random trees, contrasting dramatically with trees associated with the p spin-glass and random energy models. We also examine critically a recent conjecture on the equivalence between the NPP and a truncated random energy model.

During the construction of the new 3 story, 25,000+ square foot police station, a decision was made to participate in the LEED program to ensure the building had low operating costs, reduced emissions, conserved water and overall energy. The design of the building includes a primary-secondary ground source heat pump (GSHP) loop, a Dedicated Outside Air System (DOAS) with Energy Recovery Ventilation (ERV) wheel, all controlled by CO2 monitoring through Demand Control Ventilation (DCV) to supply heat pumps located in each space; all monitored by a Building Automation System (BAS). The building’s future energy performance was predicted through an energy simulation model (ESM) software. Measurement and verification (M&V) was then performed on the building to determine its actual energy performance. Data was collected through the building’s electrical meters, the building automation system (BAS), and other techniques to determine discrepancies. Installed electrical submetering along with ESM results helped identify faults on a subcomponent level. This bottom up approach helped drive a successful retro-commissioning of the building systems reducing energy consumption. This thesis will detail a methodology for retro-commissioning of underperforming new-construction buildings. Optimization of the building’s systems will be facilitated through utilization of the M&V and ESM data. Discussed will be techniques and strategies to benchmark the building’s systems, providing utility from the retro-commissioning and M&V results, to determine the value of the ESM. Last will be to discuss and demonstrate the future benefits of utilizing this real-time data to help building operators reduce, manage, and sustain their energy consumption profiles.

Planar defects such as free surfaces and grain boundaries in metals and alloys play important roles affecting many material properties such as fracture toughness, corrosion resistance, wetting, and catalysis. Their interactions with point defects and solute elements also play critical roles on governing the microstructural evolution and associated property changes in materials. This work seeks to use atomistic modeling to obtain a fundamental understanding of many surface and interface related properties and phenomena, namely: orientation-dependent surface energy of elemental metals and alloys, segregation of solute elements at grain boundaries and their impact on grain boundary cohesive strength, and the controversial sluggish diffusion in both the bulk and grain boundaries of high entropy alloys. First, an analytical formula is derived, which can predict the surface energy of any arbitrary (h k l) crystallographic orientation in both body-centered-cubic (BCC) and face-centered-cubic (FCC) pure metals, using only two or three low-index (e.g., (100), (110), (111)) surface energies as input. This analytical formula is validated against 4357 independent single element surface energies reported in literature or calculated by the present author, and it proves to be highly accurate but easy to use. This formula is then expanded to include the simple-cubic (SC) structure and tested against 4542 surface energies of metallic alloys of different cubic structures, and good agreement is achieved for most cases. Second, the effect of segregation of substitutional solute elements on grain boundary cohesive strength in BCC Fe is studied. It is found that the bulk substitution energy can be used as an effective indicator to predict the embrittlement or strengthening potency induced by the solute segregation at grain boundaries. Third, the controversial vacancy-mediated sluggish diffusion in an equiatomic FeNiCrCoCu FCC high entropy alloy is studied. Many literature studies have postulated that the compositional complexity in high entropy alloys could lead to sluggish diffusion. To test this hypothesis, this work compares the vacancy-mediated self-diffusion in this model high entropy alloy with a hypothetical single-element material (called average-atom material) that has similar average properties as the high entropy alloy but without the compositional complexity. The results show that the self-diffusivities in the two bulk systems are very similar, suggesting that the compositional complexity in the high entropy alloy may not be sufficient to induce sluggish diffusion in bulk high entropy alloys. Based on the knowledge learned from the bulk alloy, the exploration of the possible sluggish diffusion has been extended to grain boundaries, using a similar approach as in the study of self-diffusion in bulk. Interestingly, the results show that sluggish diffusion is evident at a Σ5(210) grain boundary in the high entropy alloy due to the compositional complexity, especially in the low temperature regime, which is different from the bulk diffusion. The underlying mechanisms for the sluggish diffusion at this grain boundary is discussed. / Doctor of Philosophy / Human beings have utilized metals and alloys for over ten millennia and learned much from them. Based on the accumulated knowledge, they have countless applications in our current daily life. However, there is still much to learn for improving our current technology and even opening new opportunities. Throughout most of history, our understanding of these materials was largely obtained through empirical experimentation and refining them into theories and scientific laws. Nowadays, due to the advancements in computer simulations, we can learn more by modeling the behaviors of metals and alloys at the length and time scales that are either be too arduous, costly, or currently impossible experimentally. This work aims at using computer modeling to study some important surface/interface related physical behaviors and properties in metals and alloys at the atomistic scale. First, this work intends to develop a robust surface energy model in an analytical form for any crystallographic orientation. Surface energy is an important material property for many surface-related processes such as fracturing, wetting, sintering, catalysis, and crystalline particle shape. Surface energy is different at different surface orientations, and predicting this difference is important for understanding these surface phenomena. Second, the effect of solute segregation on grain boundary cohesive strength is studied. Most commonly used metallic materials consist of many small crystalline grains and the borders between them are called grain boundaries, which are weak spots for fracture. The minimum energy required to split a boundary is called the grain boundary cohesive strength. The presence of solutes or impurities at grain boundaries can further alter the cohesive strength. A better understanding of this phenomena will eventually help us develop more fracture-resistant materials. The third project deals with the possible sluggish/retarded diffusion in high entropy alloys, which contain five or more principal alloying elements and have many unique mechanical, radiation-resistant, and corrosion-resistant properties. Many researchers attribute these unique properties to the slow species diffusion in these alloys, but its existence is still controversial. This work studies the atomic-level diffusion mechanisms in an FeNiCrCoCu high entropy alloy both in bulk (grain interior) and at grain boundaries in order to determine if sluggish diffusion is present and its causes.

Ag thin film dewetting upon high temperature annealing is a non-trivial problem for its application in the semiconductor industry as an ohmic contact metal. Thus, preventing Ag thin film from dewetting is of great importance. Typically, adhesion-promoting layers of chromium are deposited to prevent dewetting, but this deposition has its own process optimization parameters. In this thesis, we introduce an alternative, novel strategy for dewetting prevention via Si or In ion implantation. Electron microscopy including SEM, AFM and AES are conducted to characterize changes in film morphology after ion implantation. Thermodynamic simulation is established to better understand the mechanism of this anti-dewetting approach as well as to predict the performance of doped Ag thin films. It is found that Ag films implanted by a trace amount of Si dopants remain intact after 24 h annealing at 530℃ rather than break down into isolated particles as pure Ag film did. Furthermore, Ag grains in doped samples are much smaller than that in non-doped samples and higher Si or In doses contribute to smaller grains, indicating that a retarding force against film grain growth is introduced by the implanting species. Fortunately, electrical conductivity and optical reflectivity of doped films change trivially, suggesting an insignificant influence of external species on the film performance. The retarding force suppressing film grain growth is demonstrated to be solute drag, which will introduce a size limit towards Ag grain growth. A grain growth model including the solute drag effect is established here to describe the grain growth process. Combining our thermodynamic simulation with our grain growth model in the presence of the solute drag effect, the critical grain diameter to initiate agglomeration of 100 nm thick Ag thin film is calculated to be 350 nm and the critical Si dose to prevent 100 nm thick Ag thin film from dewetting is predicted to be 2.0×1013 per cm2. Finally, we successfully synthesize ultrathin Si thin films via ion implantation and pattern as-deposited Si films by implanting through a hard mask in order to identify some steps towards synthesizing 2-D silicon, or silicene. The as-achieved pattern has an identical shape as that of mask, suggesting Si atoms only diffuse within the implanting regions during thermal annealing. Even though only amorphous Si films are prepared at present, this novel strategy possesses potential to fabricated CMOS-compatible 2-D silicon films for semiconductor industry. / Thesis / Master of Applied Science (MASc)

This work aims to develop a model capable of calculating the energy consumption of an electric car. The goal is a method of calculating the energy needs for a certain trip and presenting recommendations for a more efficient driving mode. The purpose of the work is to create a tool to help drivers use less energy and identify problems when developing such a tool. The model is based on a power equation which considers air resistance, rolling resistance, height differences and acceleration. A model of regenerative brakes is also developed in order to account for regained kinetic energy. Using map data, a routing tool is developed to allow input of a trip which the model can calculate energy needs for. The model shows for an example car Porsche Taycan 4S over a specific test-trip an energy consumption of 128.65 Wh/km and a 3% energy savings in relation to energy efficient driving. When evaluating the model against the Urban Dynamometer Driving Schedule (UDDS) and the New European Driving Cycle (NEDC) estimated driving ranges corresponding to 493.3 km and 521.2 km respectively, in relation to official statistics of 396 km. Finally, findings during development, problems with the model and recommendations are discussed.

Cities play a crucial role in sustainable energy system transformation. Urban energy systems account for 75% of global primary energy use, and 70% of global greenhouse gas (GHG) emissions (IEA, 2021). There is currently a large, untapped potential for reducing both energy demand and emissions by focusing measures on one of the largest consumers of energy: buildings. In North Macedonia, there is an estimated energy savings potential of 57% in the residential sector, and 29% in the public service sector (Apostolska et al., 2020). In the midst of the country’s ambitious targets of decreasing energy demand and GHG reductions, the city of Bitola is in the process of developing an action plan for a sustainable transition of the city. For this purpose, there is a need to investigate the current challenges in the energy system of the city and to evaluate potential future pathways to address these challenges, with a focus on the built environment. In this thesis, a city-scale urban building energy model (UBEM) of the city of Bitola was developed using the software City Energy Analyst (CEA). This involved modeling a total of 14 024 buildings in the city ranging from residential buildings to commercial and industrial facilities. Out of these 14 024 buildings, 10 792 were included in the analysis after excluding abandoned buildings which account for an estimated 25% of the total residential building stock. One Baseline scenario based on the current energy use in the built environment in the city, and four scenarios investigating building retrofit measures and alternative heating solutions were developed for the time period 2023-2040 which were then assessed based on three key performance indicators (KPIs). A 2% implementation rate was used for the measures included in the scenarios, resulting in a total of 34% of the buildings being included in the scenario assessment. The scenarios included in the analysis are Business-as-Usual (BAU), decentralized natural gas boilers (NGB), district heating (DH) and decentralized heat pumps (HP). The KPIs include the total primary energy demand, the total operational CO2 emissions, and the economic performance of the system, measured as a net present value (NPV). All scenarios were also evaluated with and without solar photovoltaic (PV). The results showed the BAU scenario to be the lowest performing scenario for all three KPIs, while the HP scenario showed to be the best-performing scenario regarding the reduction of energy demand and CO2 emissions, with a 99% reduction of CO2 emissions and a 65% lower energy demand than in the baseline year. However, this comes at a relatively high cost compared to the other scenarios. The DH and NGB scenarios performed moderately regarding demand and CO2 emission savings while performing better from an economic standpoint. All scenarios showed a low share of buildings on an individual level having a positive NPV, thus failing to reach a positive total NPV for the entire system. On the other hand, the sensitivity analysis demonstrated how a reduction of the capital expenditure (CAPEX) led to a positive NPV for all scenarios with PV, and for all scenarios except BAU without PV. This indicates that subsidies provided by local or national stakeholders could result in a profitable investment. Two important conclusions can be drawn from the results: firstly, taking any action and implementing either of the HP, NGB and DH scenarios will be more beneficial than taking no action, and secondly, the sustainable development of the city needs to be led by the local municipality, as well as national stakeholders to enable a long-lasting transition. / Städer spelar en avgörande roll för omställningen till hållbara energisystem. Energisystem i städer står för 75% av den globala primära energianvändningen och 70% av de globala växthusgasutsläppen (IEA, 2021). För närvarande finns det en stor, outnyttjad potential för minskning av både energibehov och utsläpp genom att fokusera på åtgärder för en av de största energikonsumenterna: byggnader. I Nordmakedonien uppskattas det finnas potential för energibesparingar på 57% i bostadssektorn och 29% i offentlig sektor (Apostolska et al., 2020). I samband med landets ambitiösa mål om att minska energianvändning och växthusgasutsläpp genomgår staden Bitola för närvarande en process för att utveckla en handlingsplan för en hållbar omställning av staden. För detta ändamål krävs en undersökning av de aktuella utmaningarna i stadens energisystem och utvärdering av potentiella framtida riktningar för att möta dessa utmaningar, med fokus på den bebyggda miljön. I detta examensarbete utvecklades en modell i stadsskala av energianvändningen i byggnader för staden Bitola i Nordmakedonien med hjälp av programvaran City Energy Analyst (CEA). Modellen omfattade totalt 14 024 byggnader, från bostadshus till kommersiella och industriella fastigheter. Då 25% av stadens bostadsbyggnader uppskattas vara övergivna ingick totalt 10 792 byggnader i den slutgiltiga analysen. Ett basscenario som beskriver dagens energianvändning i byggnaderna, och fyra framtida scenarier, som omfattar energieffektiviseringsåtgärder och alternativa värmesystem, utvecklades för tidsperioden 2023-2040. En implementeringstakt om 2% av byggnadsbeståndet, vilket resulterade i att totalt 34% av byggnadsbeståndet inkluderades i scenarioanalysen. De fyra framtida scenarierna som ingick i analysen är Business-as-Usual (BAU), decentraliserade gasvärmepannor (NGB), fjärrvärme (DH) och decentraliserade värmepumpar (HP). Scenarierna bedömdes med hjälp av tre nyckeltal (KPI:er): den totala primärenergianvändningen, de totala operativa CO2 utsläppen och den ekonomiska prestandan, mätt som investeringens nu värde (NPV). Samtliga scenarier utvärderades med och utan implementering av solceller. Resultaten visade att scenariot BAU presterade sämst för alla tre KPI:er, medanHP-scenariot visade sig vara det bäst presterande scenariot för minskning avenergibehovet och CO2-utsläppen, med 99% minskning av CO2-utsläpp och 65%lägre energianvändning jämfört med basscenariot. Dock är detta förknippat medrelativt höga kostnader jämfört med de andra scenarierna. DH- och NGB-scenariotpresterade måttligt gällande besparing av energibehov och CO2-utsläpp, samtidigt somde presterade bättre ur ett ekonomiskt perspektiv. Alla scenarier resulterade i en lågandel av byggnader på individuell nivå med ett positivt NPV, vilket innebär att demisslyckas med att nå ett positivt totalt NPV för hela systemet. Å andra sidan visadekänslighetsanalysen att en minskning av investerings kostnaderna (CAPEX) ledde tillett positivt NPV för alla scenarier med solceller, och för alla scenarier utom BAU utan solceller. Detta indikerar att subventioner från lokala och nationella aktörer kan leda till en lönsam investering. Två viktiga slutsatser kan dras från dessa resultat: för det första, att vidta åtgärder och implementera något av HP-, NGB- eller DH-scenariot är mer fördelaktigt än att inte vidta några åtgärder, och för det andra, behöver den hållbara utvecklingen av staden ledas av den lokala kommunen samt nationella aktörer för att möjliggöra en långvarig omställning.