Dynamic demand modeling, energy management, and investment strategies in uncertain energy markets

This thesis is focused on development of an integrated decision making support framework to assist with the design of a sustainable community that has access to secure clean energy and utilizes innovative technologies and strategies. Innovative technologies include but not limited to renewable generation/storage resources, Plug-In Electric Vehicle (PEV), Building Monitoring Systems (BMS), Programmable Communication Devices (PCD), and etc. This framework has 3 unique components: i. energy dynamic demand modeling, ii. investment strategies for Distributed Energy Resources (DER), and iii. demand side energy management in uncertain markets. In the existing litrature and industry practices, energy load profile is considered as an input to a decision making support tools and its dynamics is ignored which can lead to unreliable and less cost effective investment decisions in the long run. The analysis of such dynamics is not possible with existing demand forecast models, which are built based on time series forecasts relying on historical data. Therefore, in this thesis a bottom-up demand forecasting model entitled High Resolution Adaptive Model (Hi-RAM) is integrated with DER investment model. Hi-RAM provides compelling results concerning the potential load shifting of PEVs, as well as how advanced energy management systems enable response to Electric Distribution Companies (EDC) price signals. Hi-RAM is a bottom-up stochastic demand model consisting of: 1- Markovian stochastic process for simulating human activities, and buildings occupany profiles. 2- Probabilistic Bayesian and Logistic technology adoption models and 3- Optimization and rule-based energy management models for building end-uses e.g., Heating, Ventilating, and Air Conditioning (HVAC), Lighting, and PEV charging which enable them to respond to EDC price signals without compromising users' comfort. The DER investment model is built as a non-linear stochastic mixed integer programming to maximize the cash flow over the planning horizon considering long-term market variations, short-term operational volatilities, and the dynamics of underlying demand. Integration of Hi-RAM and the DER investment model offers a novel analytical framework which can be used at the design stage of new products to assess their effectiveness. Such a framework can also assist decision makers to investigate investment strategies on community's DER e.g., PV solar, wind turbines, electric storages, and etc. taking into account the specifics of building behavioral and physical characteristics as well as the emergence of new end-uses and Demand-Side Management (DSM) capabilities. While developing this tool is an essential step towards sustainable and efficient energy solutions in the planning stage of new buildings and communities, attention must also be paid to the existing building stock where the U.S. buildings sector alone accounted for 7% of global primary energy consumption. Recognizing its importance, this thesis also investigates advanced energy magement and control policies for buildings within a constrained peak demand envelope while ensuring that custom climate conditions are facilitated. This will mitigate service disruption and high cost of energy production and distribution.Ph.D.Includes bibliographical referencesby Farbod Farza