Semi-empirical mapping method for energy recovery wheel performance simulation

•Training data sets are generated by a validated finite difference method model.•Mapping method predictions agree well with data sets from independent sources.•The new method predicts performance of wheels with different physical geometries.•The new approach method requires fewer training data points than current models. Energy recovery ventilators are often employed in buildings to decrease the energy consumed by their HVAC systems, and to improve overall indoor air quality. Several studies in the open literature have developed physical or empirical models to simulate the performance of energy recovery wheels. However, these models are often computationally intensive, time-consuming, or require numerous experimental data points to train and execute the performance predictions. These models also tend to lack flexibility for varying all of the available wheel design parameters. Hence, developing a mapping method with better computational efficiency and flexibility is the goal of this research. A finite difference numerical model for simulating the performance of an energy wheel has been developed and validated using experimental test results from independent laboratories. This model was then employed to provide an extensive data set for the development of an energy wheel performance mapping method. After validating this new mapping approach, the method predictions were compared against independent data sets from two different laboratories, and additional sources available in the literature, to identify its universality. The mapping method delivers good agreement between the predictions and validation data, and requires only a small number of data points to train, which is one of its novel contributions. Another unique contribution of the proposed mapping method is that once the model is trained it can predict the performance characteristics for other wheels with different physical design geometries and operating conditions, provided only that the desiccant material is the same.