LAUSR

Abstract Mathematical models of heating, ventilation and air conditioning (HVAC) systems is of great significance for optimizing power consumptions in large buildings. This paper proposes a semi-physical model to describe static characters of a typical HVAC system. Major components of the HVAC system are modeled individually according to their thermodynamic principles or input–output relations, and are integrated into an overall model by choosing input and output variables from a perspective of optimizing power consumptions. One set of experimental data samples is used to estimate unknown model coefficients, and more data sets are taken to validate the estimated model. Based on the estimated model, genetic algorithms are exploited to find the optimal operating point of the HAVC system for minimizing power consumptions under constraints on indoor temperatures.