Download files
Abstract
The electricity industry is plagued by technical and economic challenges due to increasing demand for energy services, increasing reliability requirements, and concerns on climate change. It can be assisted in these challenges by including the potential contributions of the demand-side to optimize its operation. This may be achieved by utilizing decentralized generating, storage, and controllable load resources and by engaging in distributed decision-making.
This thesis presents a novel energy service decision-support tool (ES-DST) that consumers can use to optimize the acquisition of their energy services. The tool is composed of an energy service model and a scheduler for distributed energy resources (DER). The model is based on the consumers putting different levels of benefit to services at different times of the day, and it assigns this benefit to the energy that realizes the service. The scheduling algorithm determines how controllable DER available to the consumers may be operated to maximize their net benefit based on the energy service and DER models, and DER technical characteristics and capabilities. The ES-DST may be utilized to control DER in the household level, supporting the concept of a ‘smart’ home, or within a large building or group of buildings, supporting the concept of a microgrid or a virtual power plant.
The capabilities of the ES-DST are demonstrated using a ‘smart’ home case study. It is used to create deterministic DER schedules under different electricity tariff structures. It is also used to formulate robust day-ahead and real-time operating schedules with stochastic energy service demand, DER availability, and activation of dynamic peak pricing (DPP). The simulation results could give consumers insights on how to operate their DER to maximize their benefits, and give industry managers the potential impacts of price-based demand response programs like time-of-use rates, DPP, and net feed-in tariffs. The ES-DST is also used to determine the value added by the coordination among DER, and to identify the forecasted information that are crucial to making effective schedules. The scheduling of DER is a challenging optimization problem; hence, a heuristic simulation-based approach based on particle swarm optimization (PSO) is used. Improvements to the PSO are also presented, and are demonstrated to generate effective schedules for more complex problems within manageable computation times.