| dc.description.abstract | Ground Source Heat Pump (GSHP) systems have grown in popularity and acceptance worldwide as an attractive option to replace conventional Heating Ventilation and Air Conditioning (HVAC) technologies due to their capacity to provide space heating and cooling in buildings and houses. Such GSHP systems may participate as a price-taker in electricity markets through a load aggregator to optimize their load demand, being able to provide grid services, such as load shifting. Therefore, aggregated GSHP systems have the potential, if properly designed, integrated, and applied, to yield energy and carbon savings in the energy market. However, the integration of such aggregated GSHP systems brings new challenges to operators, as it involves uncertainties on ambient temperature and electricity price forecasts, which can be highly volatile and thus impact the GSHP system operation and its participation in electricity markets. From a detailed literature review of GSHP applications for load management for residential users, it can be concluded that there are no works that discuss the operational performance of large-scale GSHP systems, modeled in detail, and their integration in electricity markets; additionally, none of the existing works have considered uncertainties in terms of ambient temperature and electricity price forecasts for the optimal operation of aggregated GSHP systems.
After a comprehensive review of the relevant background related to GSHP systems, aggregator strategies in the electricity market, and optimization in the presence of uncertainties, in this thesis, a detailed mathematical model is presented of a GSHP with a vertical U-pipe Ground Heat eXchanger (GHX) configuration to provide residential space heating/cooling, integrating them into a load aggregator model. Based on this model, a two-stage operational strategy for the GSHP price-taker aggregator participating in Day-Ahead Market (DAM) and Real-Time Market (RTM) is proposed, to determine the optimal annual heating/cooling load dispatch to control the temperatures for a community of houses that minimizes the aggregator’s cost. Simulations are presented then of an aggregator’s optimal load dispatch with a conventional HVAC and the proposed GSHP alternative, considering comfort maximization vis-a-vis minimization of electricity costs, and showing the impact of each objective with respect to the dispatch of controllable loads, in-house temperature, and total procurement costs.
Finally, a novel model based on Robust Optimization (RO) is proposed and developed, considering uncertainties in terms of the DAM and RTM electricity prices and hourly ambient temperature forecasts, which yields an optimum schedule that protects against the worst-case scenario for a given level of conservatism. The RO model is compared and validated in a realistic test system with respect to Model Predictive Control (MPC) and Monte Carlo Simulations (MCS) approaches that are traditionally used to manage uncertainty. It is shown that the proposed RO approach is computationally efficient compared to the MPC and MCS approaches, and properly accounts for the considered uncertainties, demonstrating the advantage of the presented RO technique for GSHP dispatch by aggregators. | en |
