With growing concerns on global energy demand and climate change, it is important to focus on efficient utilization of electricity in commercial buildings, which contribute significantly to the overall electricity consumption. Accordingly, there has been a number of Building Energy Management (BEM) software/hardware solutions to monitor energy consumption and other measurements of individual building loads. BEM software serves as a platform to implement smart control strategies and stores historical data. Although BEM software provides such lucrative benefits to building operators, in terms of energy savings and personalized control, these benefits are not harnessed by most small to mid-sized buildings due to a high cost of deployment and maintenance. A cloud-based BEM system can offer a low-cost solution to promote ease of use and support a maintenance-free installation. In a typical building, a conventional router has a public address and assigns private addresses to all devices connected to it. This led to a network topology, where the router is the only device in the Internet space with all other devices forming an isolated local area network behind the router. Due to this scenario, a cloud-based BEM software needs to pass through the router to access devices in a local area network. To address this issue, some devices, during operation, make an outbound connection to traverse through the router and provide an interface to itself on the Internet. Hence, based on their capability to traverse through the router, devices in a local area network can be distinguished as cloud and non-cloud devices. Cloud-based BEM software with sufficient authorization can access cloud devices. In order to access devices adhering to non-cloud protocols, cloud-based BEM software requires a device in the local area network which can perform traversal through the router on behalf of all non-cloud devices. Such a device acts as an IoT gateway, to securely interconnect devices in a local area network with cloud-based BEM software. This thesis focuses towards architecting, designing and prototyping an Internet-of-Things (IoT) gateway which can perform traversal on behalf of non-cloud devices. This IoT gateway enables cloud-based BEM software to have a comprehensive access to supported non-cloud devices. The IoT gateway has been designed to support BACnet, Modbus and HTTP RESTful, which are the three widely adopted communication protocols in the building automation and control domain. The developed software executes these three communication protocols concurrently to address requests from cloud-based BEM system. The performance of the designed architecture is independent of the number of devices supported by the IoT gateway software. / Master of Science

The focus of this study was to investigate an alternative and more cost effective solution for occupancy sensing in commercial office buildings. The intended purpose of this solution is to aid in efficient energy management. The main requirements were that the proposed solution made use of existing infrastructure only, and provided a means to focus on occupant location. This research was undertaken due to current solutions making use of custom occupancy sensors that are relatively costly and troublesome to implement. These solutions focus mainly on monitoring environmental changes, and not the physical locations of the occupants themselves. Furthermore, current occupancy sensing solutions are unable to provide proximity and timing information that indicate how far an occupant is located from a specific area, or how long the occupant resided there. The research question was answered by conducting a proof of concept study with data simulated in the OMNeT++ environment in conjunction with the MiXiM framework for wireless networks. The proposed solution investigated the fidelity of existing WiFi infrastructure for occupancy sensing, this entailed the creation of a Virtual Occupancy Sensor (VOS) that implemented RSS-based localisation for an occupant’s WiFi devices. Localisation was implemented with three different location estimation techniques; these were trilateration, constrained nearest neighbour RF mapping and unconstrained nearest neighbour RF mapping. The obtained positioning data was interpreted by a developed intelligent agent that was able to transform this regular position data into relevant occupancy information. This information included a distance from office measurement and an occupancy result that can be interpreted by existing energy management systems. The accuracy and operational behaviour of the developed VOS were tested with various scenarios. Sensitivity analysis and extreme condition testing were also conducted. Results showed that the constrained nearest neighbour RF mapping approach is the most accurate, and is best suited for occupancy determination. The created VOS system can function correctly with various tested sensitivities and device loads. Furthermore results indicated that the VOS is very accurate in determining room level occupancy although the accuracy of the position coordinate estimations fluctuated considerably. The operational behaviour of the VOS could be validated for all investigated scenarios. It was determined that the developed VOS can be deemed fit for its intended purpose, and is able to give indication to occupant proximity and movement timing. The conducted research confirmed the fidelity of WiFi infrastructure for occupancy sensing, and that the developed VOS can be considered a viable and cost effective alternative to current occupancy sensing solutions. / MIng (Computer and Electronic Engineering), North-West University, Potchefstroom Campus, 2014

The focus of this study was to investigate an alternative and more cost effective solution for occupancy sensing in commercial office buildings. The intended purpose of this solution is to aid in efficient energy management. The main requirements were that the proposed solution made use of existing infrastructure only, and provided a means to focus on occupant location. This research was undertaken due to current solutions making use of custom occupancy sensors that are relatively costly and troublesome to implement. These solutions focus mainly on monitoring environmental changes, and not the physical locations of the occupants themselves. Furthermore, current occupancy sensing solutions are unable to provide proximity and timing information that indicate how far an occupant is located from a specific area, or how long the occupant resided there. The research question was answered by conducting a proof of concept study with data simulated in the OMNeT++ environment in conjunction with the MiXiM framework for wireless networks. The proposed solution investigated the fidelity of existing WiFi infrastructure for occupancy sensing, this entailed the creation of a Virtual Occupancy Sensor (VOS) that implemented RSS-based localisation for an occupant’s WiFi devices. Localisation was implemented with three different location estimation techniques; these were trilateration, constrained nearest neighbour RF mapping and unconstrained nearest neighbour RF mapping. The obtained positioning data was interpreted by a developed intelligent agent that was able to transform this regular position data into relevant occupancy information. This information included a distance from office measurement and an occupancy result that can be interpreted by existing energy management systems. The accuracy and operational behaviour of the developed VOS were tested with various scenarios. Sensitivity analysis and extreme condition testing were also conducted. Results showed that the constrained nearest neighbour RF mapping approach is the most accurate, and is best suited for occupancy determination. The created VOS system can function correctly with various tested sensitivities and device loads. Furthermore results indicated that the VOS is very accurate in determining room level occupancy although the accuracy of the position coordinate estimations fluctuated considerably. The operational behaviour of the VOS could be validated for all investigated scenarios. It was determined that the developed VOS can be deemed fit for its intended purpose, and is able to give indication to occupant proximity and movement timing. The conducted research confirmed the fidelity of WiFi infrastructure for occupancy sensing, and that the developed VOS can be considered a viable and cost effective alternative to current occupancy sensing solutions. / MIng (Computer and Electronic Engineering), North-West University, Potchefstroom Campus, 2014

Buildings are responsible for a significant portion of energy consumption worldwide. Intelligent buildings have been devised as a potential solution, where energy consumption and building use are harmonised. At the heart of the intelligent building is the building energy management system (BEMS), the central platform which manages and coordinates all the building monitoring and control subsystems, such as heating and lighting loads. There is often a disconnect between the BEMS and the building it is installed in, leading to inefficient operation, due to incongruous commissioning of sensors and control systems. In these cases, the BEMS has a lack of knowledge of the building form and function, requiring further complex optimisation, to facilitate efficient all year round operation. Flawed BEMS configurations can then lead to ‘sick buildings’. Recently, building energy performance simulation (BEPS) has been viewed as a conceptual solution to assist in efficient building control. Building energy simulation models offer a virtual environment to test many scenarios of BEMS operation strategies and the ability to quickly evaluate their effects on energy consumption and occupant comfort. Challenges include having an accurate building model, but recent advances in building information modelling (BIM) offer the chance to leverage existing building data, which can be translated into a form understood by the building simulator. This study will address these challenges, by developing and integrating a BEMS, with a BIM for BEPS assisted predictive control, and assessing the outcome and potential of the integration.

Balancing indoor comfort and energy consumption is crucial to building energy efficiency. Occupancy information is a vital aspect in this process, as it determines the energy demand. Although there are various sensors used to gather occupancy information, environmental sensors stand out due to their low cost and privacy benefits. Machine learning algorithms play a critical role in estimating the relationship between occupancy levels and environmental data. To improve performance, more complex models such as deep learning algorithms are necessary. Long Short-Term Memory (LSTM) is a powerful deep learning algorithm that has been utilized in occupancy estimation. However, recently, an algorithm named Attention has emerged with improved performance. The study proposes a more effective model for occupancy level estimation by incorporating Attention into the existing Long Short-Term Memory algorithm. The results show that the proposed model is more accurate than using a single algorithm and has the potential to be integrated into building energy control systems to conserve even more energy. / Master of Science / The motivation for energy conservation and sustainable development is rapidly increasing, and building energy consumption is a significant part of overall energy use. In order to make buildings more energy efficient, it is necessary to obtain information on the occupancy level of rooms in the building. Environmental sensors are used to measure factors such as humidity and sound to determine occupancy information. However, the relationship between sensor readings and occupancy levels is complex, making it necessary to use machine learning algorithms to establish a connection. As a subfield of machine learning, deep learning is capable of processing complex data. This research aims to utilize advanced deep learning algorithms to estimate building occupancy levels based on environmental sensor data.

The efficiency, security and resiliency are very important factors for the operation of a distribution power system. Taking into account customer demand and energy resource constraints, electric utilities not only need to provide reliable services but also need to operate a power grid as efficiently as possible. The objective of this dissertation is to design, develop and deploy the Multi-Agent Systems (MAS) - together with control algorithms - that enable demand response (DR) implementation at the customer level, focusing on both residential and commercial customers. For residential applications, the main objective is to propose an approach for a smart distribution transformer management. The DR objective at a distribution transformer is to ensure that the instantaneous power demand at a distribution transformer is kept below a certain demand limit while impacts of demand restrike are minimized. The DR objectives at residential homes are to secure critical loads, mitigate occupant comfort violation, and minimize appliance run-time after a DR event. For commercial applications, the goal is to propose a MAS architecture and platform that help facilitate the implementation of a Critical Peak Pricing (CPP) program. Main objectives of the proposed DR algorithm are to minimize power demand and energy consumption during a period that a CPP event is called out, to minimize occupant comfort violation, to minimize impacts of demand restrike after a CPP event, as well as to control the device operation to avoid restrikes. Overall, this study provides an insight into the design and implementation of MAS, together with associated control algorithms for DR implementation in smart buildings. The proposed approaches can serve as alternative solutions to the current practices of electric utilities to engage end-use customers to participate in DR programs where occupancy level, tenant comfort condition and preference, as well as controllable devices and sensors are taken into account in both simulated and real-world environments. Research findings show that the proposed DR algorithms can perform effectively and efficiently during a DR event in residential homes and during the CPP event in commercial buildings. / Ph. D.

Commercial buildings consume more than 40% of the total energy consumption in the United States. Almost 90% of these buildings are small- and medium-sized buildings that do not have a Building Energy Management (BEM) system. The reasons behind this are – lack of awareness, unavailability of inexpensive packaged solutions, and disincentive to invest in a BEM system if the tenant is not the owner. Several open source tools and technologies have emerged recently that can be used for building automation and energy management. However, none of these systems is turnkey and deployment ready. They also lack consistent and intuitive navigation, security, and performance required for a BEM system. The overall project - of which this thesis research is a part - addresses the design and implementation of an open source secure web based user platform to monitor, schedule, control, and perform functions needed for a BEM system serving small and medium-size buildings. The focus of this work are: principles of intuitive graphical user interface design, abstracting device functions into a comprehensive data model, identifying threats and vulnerabilities, and implementing a security framework for the web platform. Monitor and control solutions for devices such as load controllers and sensors are abstracted and their decentralized control strategies are proposed and implemented using an open source robust scalable user platform accessible locally and remotely. The user platform is open-source, scalable, provides role-based access, dynamic, and modular in design. The comprehensive data model includes a user management model, device model, session model, and a scheduling model. The data model is designed to be flexible, robust and can be extended for any new device type. Security risks are analyzed using a threat model to identify security goals. The proposed security framework includes user authentication, device approval, role-based access, secure information exchange protocols, and web platform security. Performance of the user interface platform is evaluated for responsiveness in different screen sizes, page response times, throughput, and the performance of client side entities. / Master of Science

In the United States, over 40% of the country's total energy consumption is in buildings, most of which are either small-sized (<5,000 sqft) or medium-sized (5,000-50,000 sqft). These buildings offer excellent opportunities for energy saving and demand response (DR), but these opportunities are rarely utilized due to lack of effective building energy management systems and automated algorithms that can assist a building to participate in a DR program. Considering the low load factor in US and many other countries, DR can serve as an effective tool to reduce peak demand through demand-side load curtailment. A convenient option for the customer to benefit from a DR program is to use automated DR algorithms within a software that can learn user comfort preferences for the building loads and make automated load curtailment decisions without affecting customer comfort. The objective of this dissertation is to provide such a solution. First, this dissertation contributes to the development of key features of a building energy management open source software platform that enable ease-of-use through plug and play and interoperability of devices in a building, cost-effectiveness through deployment in a low-cost computer, and DR through communication infrastructure between building and utility and among multiple buildings, while ensuring security of the platform. Second, a set of reinforcement learning (RL) based algorithms is proposed for the three main types of loads in a building: heating, ventilation and air conditioning (HVAC) loads, lighting loads and plug loads. In absence of a DR program, these distributed agent-based learning algorithms are designed to learn the user comfort ranges through explorative interaction with the environment and accumulating user feedback, and then operate through policies that favor maximum user benefit in terms of saving energy while ensuring comfort. Third, two sets of DR algorithms are proposed for an incentive-based DR program in a building. A user-defined priority based DR algorithm with smart thermostat control and utilization of distributed energy resources (DER) is proposed for residential buildings. For commercial buildings, a learning-based algorithm is proposed that utilizes the learning from the RL algorithms to use a pre-cooling/pre-heating based load reduction method for HVAC loads and a mixed integer linear programming (MILP) based optimization method for other loads to dynamically maintain total building demand below a demand limit set by the utility during a DR event, while minimizing total user discomfort. A user defined priority based DR algorithm is also proposed for multiple buildings in a community so that they can participate in realizing combined DR objectives. The software solution proposed in this dissertation is expected to encourage increased participation of smaller and medium-sized buildings in demand response and energy saving activities. This will help in alleviating power system stress conditions by employing the untapped DR potential in such buildings. / Ph. D.

Energy and environment have become hot topics in the world. The building sector accounts for a high proportion of energy consumption, with over one-third of energy use globally. A variety of optimization methods have been proposed for building energy management, which are mainly divided into two types: model-based and model-free. Model Predictive Control is a model-based method but is not widely adopted by the building industry as it requires too much expertise and time to develop a model. Model-free Deep Reinforcement Learning(DRL) has successful applications in game-playing and robotics control. Therefore, we explored the effectiveness of the DRL algorithms applied to building control and investigated which DRL algorithm performs best. Three DRL algorithms were implemented, namely, Deep Deterministic Policy Gradient(DDPG), Double Deep Q learning(DDQN) and Soft Actor Critic(SAC). We used the building optimization testing (BOPTEST) framework, a standardized virtual testbed, to test the DRL algorithms. The performance is evaluated by two Key Performance Indicators(KPIs): thermal discomfort and operational cost. The results show that the DDPG agent performs best, and outperforms the baseline with the saving of thermal discomfort by 91.5% and 18.3%, and the saving of the operational cost by 11.0% and 14.6% during the peak and typical heating periods, respectively. DDQN and SAC agents do not show a clear advantage of performance over the baseline. This research highlights the excellent control performance of the DDPG agent, suggesting that the application of DRL in building control can achieve a better performance than the conventional control method. / Energi och miljö blir heta ämnen i världen. Byggsektorn står för en hög andel av energiförbrukningen, med över en tredjedel av energianvändningen globalt. En mängd olika optimeringsmetoder har föreslagits för Building Energy Management, vilka huvudsakligen är uppdelade i två typer: modellbaserade och modellfria. Model Predictive Control är en modellbaserad metod men är inte allmänt antagen av byggbranschen eftersom det kräver för mycket expertis och tid för att utveckla en modell. Modellfri Deep Reinforcement Learning (DRL) har framgångsrika tillämpningar inom spel och robotstyrning. Därför undersökte vi effektiviteten av DRL-algoritmerna som tillämpas på byggnadskontroll och undersökte vilken DRL-algoritm som presterar bäst. Tre DRL-algoritmer implementerades, nämligen Deep Deterministic Policy Gradient (DDPG), Double Deep Q Learning (DDQN) och Soft Actor Critic (SAC). Vi använde ramverket Building Optimization Testing (BOPTEST), en standardiserad virtuell testbädd, för att testa DRL-algoritmerna. Prestandan utvärderas av två Key Performance Indicators (KPIs): termiskt obehag och driftskostnad. Resultaten visar att DDPG-medlet presterar bäst och överträffar baslinjen med besparingen av termiskt obehag med 91.5% och 18.3%, och besparingen av driftskostnaden med 11.0% och 14.6% under topp och typisk uppvärmning perioder, respektive. DDQN- och SAC-agenter visar inte en klar fördel i prestanda jämfört med baslinjen. Denna forskning belyser DDPG-medlets utmärkta prestanda, vilket tyder på att tillämpningen av DRL i byggnadskontroll kan uppnå bättre prestanda än den konventionella metoden.

Being located in subtropical area, the weather in Taiwan is constantly hot and humid which imposes huge cooling load on buildings. Especially, the economic booms in Taiwan further boosted power demand, and worsened the power shortage situation. Dr. H.T. Lin and Dr. K.H. Yang had conducted systematic research since mid-1980s, which constructed a solid ground in this field in Taiwan. Among these results, the ENVLOAD index has become legal binding since 1997 while the PACS index is now under investigation. However, it is in short of analysis and full-scale experimental investigation on energy use of hotels in Taiwan. Therefore, the establishment of the EUI and DUI indexes in Taiwan is the goal of this study. A simplified calculation method has been established in analyzing the energy use and demand use of hotels in Taiwan, by normalizing experimental data from full-scale tests. The result can be drawn accurately based on a few terms, which are available from daily building operations such as occupancy, and is thus practically straightforward and easy to use. In addition, the accuracy was validated by experiments performed and data collected through information technology with Internet access in 4 different forms, which yielded successful results. It is anticipated that the calculation methodology developed in this study on EUI and DUI, and the experimental validation would provide a foundation for the establishment of hotel building energy codes in Taiwan in the future.