Verification of a Neural Network-Based Controller for Commercial Ice Storage Systems

摘要

This paper describes the validation and performance of an optimal neural network-based controller for an ice thermal storage system. The controller self-learns equipment responses to the environment and then determines the control settings that should be used. As such, there is minimal need to calibrate the controller to installed equipment. Results are verified using computer simulation as well as with the operation of a full-scale HVAC laboratory. These results demonstrate the robustness of a neural network-based controller and its ability to develop an optimal solution with minimal human interaction.