Energy exchange efficiency prediction from non-linear regression for membrane-based energy-recovery ventilator cores

Abstract For the design and optimization of membrane-based energy-recovery ventilators (MERVs) it is essential to know the energy-exchange efficiency. Unfortunately, performing a detailed and comprehensive measurement of the energy-exchange efficiency for all operating conditions is time-consuming and costly. In this paper, a method is proposed that makes it possible to predict the efficiency of MERV cores even if only a limited set of experimental data is available. By using a nonlinear regression model, which was derived from experimentally tested data, the efficiency of cross-flow air-to-air ventilator cores can successfully be predicted for different core-sizes and air-flow rates. As verified by experimental data, the relative error of the sensible- and enthalpy-heat efficiency prediction can be controlled within about ±7.0%, while the relative error of the latent heat efficiency prediction can be controlled within about ±8.0%. The method is suitable for both winter and summer operating conditions, and there is no need to test the characteristics of the membrane materials. Overall, the proposed method is practical and can be used to guide the design and operation of future MERVs.