Building energy model calibration using automated optimization-based algorithm

Abstract

Multiple numbers of Building Energy Simulation (BES) programs have been improved and implemented during the last decades. BES models play a crucial role in understanding building energy demands and accelerating the malfunction diagnosis. However, due to the very high number of interacting parameters, most of the developed energy simulation programs do not accurately predict building energy performance under a known condition. Even the energy models which are developed with the very precise assignment of parameters, there is always significant discrepancies between the simulation results and the real-time data measurements. Current study develops an optimization-based framework to calibrate the whole building energy model. The optimization algorithm attempts to set the identified parameters to minimize the error between the simulation results and the real-time measurements. Due to the high number of parameters, the developed optimization algorithm utilizes a Harmony Search algorithm as its search engine coupled with the energy simulation model to accelerate the calibration process. Moreover, to illustrate the efficiency of using the developed framework, a case study of the office building is modeled and calibrated and the statistical analysis was conducted to assess the accuracy of the results. The results of the calibration process show the reliability of the framework.