Paper Conference

Proceedings of Building Simulation 2021: 17th Conference of IBPSA


Development of Gray-box modeling technique for HVAC&R systems with automatic calibration process

Hyemi Kim 1, Sumiyoshi Daisuke 2, Youngjin Choi 3
1 Ph.D. Student, Faculty of Human-Environment Studies, Kyushu University, Japan
2 Associate Prof., Faculty of Human-Environment Studies, Kyushu University, Japan
3 Assistant Prof., Department of Architectural Engineering, Kyonggi University, Repubpic of Korea

Abstract: A gray-box model overcomes the limitations of black-box building modeling. However, existing gray-box models have many parameters and calculate an optimized objective function from the data. We developed gray-box techniques for heating, ventilation, air-conditioning, and refrigeration (HVAC&R) equipment to understand the status of the system operation. The parameters of the individual equipment model are minimized and fitted for interval data by a machine learning model to automatically perform the calibration process. The power consumption of the individual models was approximately 14% higher than that of TRNSYS, and the power consumption of the entire system model was approximately 7% lower.
Keywords: Automatized, Calibrate, Building Performacne Simulation, ANN, Grey-box modeling
Pages: 1625 - 1632