MODELING AND OPTIMIZATION OF AN ELECTRIC RESIDENTIAL DOMESTIC WATER HEATER MODEL WITH CONTROL

Restricted
Author:
Hodge, Eric Alan
Area of Honors:
Mechanical Engineering
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
  • Dr. Stephaie Stockar, Thesis Supervisor
  • Dr. Jacqueline O’connor, Honors Advisor
Keywords:
  • Demand Response
  • Renewable integration
  • Model Predictive Control
  • Domestic water heater
Abstract:
Extensive flexible electricity loads are capable of offering grid stability services in both a traditional fossil fuel grid mix to reduce the necessary peak generation capacity requirements, and a renewable mix to scale demand with instantaneous generation capacity. Electric domestic water heaters such as electric resistance and heat pump storage water heaters offer favorable prospects for useful utility scale demand regulation in a time of use pricing grid due to their inherent energy storage capability, relative control simplicity and ubiquity. This thesis develops an electric resistance and heat pump water heater models and develops a model predictive controller scenario with particle swarm optimization which generates a prediction of a near-optimal load scheduling and control solution. The water heater model and controller are linked to a preexisting whole-home model which also has a model predictive controller. The electricity consumption of the whole-home, including the water heater, is simulated with and without optimal control, and in single home and multi-home scenarios. The results of these simulations show that there is potential to use the proposed controller for demand side load management. However, the extent to which peak load reduction or other control objectives can be achieved is dependent on further development and tuning of the controller parameters and relative importance of control objectives and consumer inconvenience.