Optimization-Based Data-Enabled Modeling Technique for HVAC Systems Components

Most of the energy consumed by the residential and commercial buildings in the U.S. is dedicated to space cooling and heating systems, according to the U.S. Energy Information Administration. Therefore, the need for better operation mechanisms of those existing systems become more crucial. The most vital factor for that is the need for accurate models that can accurately predict the system component performance. Therefore, this paper’s primary goal is to develop a new accurate data-driven modeling and optimization technique that can accurately predict the performance of the selected system components. Several data-enabled modeling techniques such as artificial neural networks (ANN), support vector machine (SVM), and aggregated bootstrapping (BSA) are investigated, and model improvements through model structure optimization proposed. The optimization algorithm will determine the optimal model structures and automate the process of the parametric study. The optimization problem is solved using a genetic algorithm (GA) to reduce the error between the simulated and actual data for the testing period. The models predicted the performance of the chilled water variable air volume (VAV) system’s main components of cooling coil and fan power as a function of multiple inputs. Additionally, the packaged DX system compressor modeled, and the compressor power was predicted. The testing results held a low coefficient of variation (CV%) values of 1.22% for the cooling coil, and for the fan model, it was found to be 9.04%. The testing results showed that the proposed modeling and optimization technique could accurately predict the system components’ performance.

This creative work has been published under the Creative Commons Attribution 4.0 International (CC-BY 4.0) license which allows copying, and redistribution as well as adaptation of the original work provided appropriate credit is given to the original author and the conditions of the license are met.