Integrated framework for modeling and optimization of commercial district cooling systems

Abstract

The cooling of buildings drives the peak electric demand in the summer, and the cost associated with this peak demand can significantly increase the summer electric bill for many commercial buildings. Modeling and optimization of heating, ventilating, and air-conditioning (HVAC) systems in commercial districts could provide significant benefits in terms of minimizing a) energy-related costs, b) carbon-dioxide emissions, and c) enduring natural or artificial disturbances. In addition, optimization of HVAC systems could provide different scales and degrees of demand flexibility that could make buildings and districts grid-friendly. However, district-scale modeling and optimization of HVAC systems requires a framework to automate the optimization process for connected buildings. While there are co-simulation approaches that could enable optimization of HVAC systems at the district-scale, previous studies have mostly conducted this optimization at the building level or at the standalone component level (e.g., chillers) for buildings and district cooling systems. Many of these studies use simplified models for HVAC components that do not fully represent the actual performance of air- and water- side components. These studies also neglect some of the utilized components for the air- and water- side. This thesis develops an open-source, non-linear approach to optimize the air- and water- side HVAC systems for existing cooling systems of commercial districts. Two demand scenarios (i.e., rigid or flexible) are analyzed for the waterside of the HVAC system and the savings projected by the flexible demand optimization of the waterside are justified by the airside components integration. Overall, this thesis provides an integrated framework for modeling and optimization of HVAC systems in existing commercial districts. The newly developed framework is demonstrated using the Colorado School of Mines’ main chilled water plant, developing data-driven performance models with as-operated data to realistically simulate district-scale HVAC components for both the air- and water- side of the HVAC system.