Online supervisory and optimal control of complex building central chilling systems

This thesis presents online supervisory and optimal control strategies for complex building central chilling systems to enhance their energy efficiency. The software tools and implementation guidelines for applying these strategies in practice are provided as well.;To test and analyze control performances and economic feasibilities of different control strategies to determine the most promising strategy prior to site implementation, a dynamic simulation platform for complex building chilling systems was developed. Three energy performance tests associated with the optimization of major control variables were then conducted to evaluate energy saving potentials in complex chilling systems.;To formulate the online supervisory and optimal control strategies, simplified models of major chilling system components (i.e., chillers, cooling towers, pumps, etc.) were developed or selected in this thesis.;To design the optimal control strategy for chilled water systems, the speed and sequence control strategies for variable speed pumps in complex air-conditioning systems were developed and presented. The performances of these strategies were tested using a simulation-assisted test method in which the control strategies were tested in a simulated virtual environment similar to the situation when they are actually implemented in practice.;Based on the simplified chiller and cooling tower models developed, an optimal control strategy for complex condenser cooling water systems is developed. This strategy consists of the performance predictor, cost estimator, optimization technique and supervisory strategy. The control and computation performances of this strategy were evaluated by comparing with that of a GA (genetic algorithm)-based strategy, while the energy performance was evaluated by comparing with that of the conventional strategies. The results show that this optimal control strategy has satisfactory performance and is suitable for online applications.;An optimal control strategy for complex chilled water systems is also developed. The performance of this strategy was evaluated using the simulation-assisted test method by comparing it with that of other strategies. The results show that about 1.28%∼2.63% energy in the system under investigation can be saved thanks to the use of this optimal control strategy.;Lastly, the software tools and implementation guidelines for applying these online supervisory and optimal control strategies in practice are presented.