A development and testing framework for simulation-based supervisory control with application to optimal zone temperature ramping demand response using a modified genetic algorithm

There is a growing demand for integrated control strategies for building systems with numerous responsive elements, such as solar shading devices, thermal storage and hybrid ventilation systems. Simulation-based supervisory control is a promising way of approaching this challenge. The essential idea is to use detailed building simulation real-time within a supervisory control system to test possible set-point configurations at each controller time-step before choosing the best configuration to use in the building. This dissertation presents a flexible software framework for simulation-based supervisory control, along with a modified genetic algorithm developed for use within it, and applies it to a case study of demand response by zone temperature ramping in an office space. Rule-based and simulation-based control variants are studied by using a two-model configuration (one acting as the 'real' building, the other being used within the simulation-based control framework). For the case study, simulation-based control was found to perform only slightly better than a logarithmic rule-based approach under ideal conditions, and worse under conditions of model or prediction inaccuracies. The results are of use in guiding further research, and the case study itself has been a good test of the software framework, which can be further developed and should be useful in future simulation-based control research