Evaluation of optimal control for ice storage systems

Ice storage systems are effective in reducing the operating cost of cooling equipment. By operating the refrigeration equipment during off-peak hours to recharge the storage tank and discharge it during on-peak hours, a significant fraction of on-peak electrical demand and energy consumption is shifted to off-peak periods. Cost savings are obtained because utility rates favor leveled energy consumption patterns.; To evaluate the potential of ice storage systems in reducing operating cost, a simulation environment was developed which determines the optimal control strategy that minimizes the total operating cost. The total operating cost combines demand and energy charges as billed to commercial customers. A dynamic programming (DP) based search algorithm was developed to determine the optimal control strategy. Mixed integer programming was applied to validate the DP-based heuristic. To expedite the search procedure, a second heuristic was proposed to find the approximate value of the two search parameters.; The optimal control was compared to three conventional control strategies: chiller-priority, constant-proportion, and storage-priority control. Using three cooling plant models of increasing complexity (basic, refined, and realistic), the effects of various factors including storage losses, utility rate structure, load forecasting performance, and design parameters were evaluated. The realistic model requires adjustable plant parameters to be optimized to allow for a true optimization. A function minimization algorithm was applied to find the minimal plant power consumption under all conditions. To expedite this dual optimization task regarding power consumption and operating cost, the minimal power values are stored in a database and retrieved by the cost optimization algorithm.; Guidelines for optimal control were developed by investigating three ice storage systems, three chiller types, five building and weather types, and eight utility rate structures. Over 1400 cases were analyzed to determine common features and to quantify the energy cost savings for various control strategies. General heuristics are suggested to improve the performance of these three conventional controls.