An Innovative Clustering Technique to Generate Hybrid Modeling of Cooling Coils for Energy Analysis : A Case Study for Control Performance in Hvac Systems

What dynamic modeling of cooling coils still lacks, despite past studies, no definitive evidence regarding creating models or empirical correlations that cover all aspects of the flow regime (laminar, transition, and turbulent). The cooling coil is characterized by a highly nonlinear dynamic subject to multiple inputs, coupling between the latent and sensible heat transfer modes, uncertain disturbances, nonlinear constraints, and multivariate systems, all of which are among the significant challenges when it comes to modelling. Therefore, a hybrid structure model was adopted in this study to overcome these challenges. In addition, two different optimization methods, Neural Networks' Weights and Takagi–Sugeno fuzzy (TSF) model, were used. The hybridization of model layers using two different parameter types for nonlinear forms provides great flexibility to handle complex multi-input physical variables and strongly nonlinear trend signals. In each layer, the Gauss-Newton algorithm (GNA) was employed to tune the weight parameters, such that the output response obtained by the nonlinear regression of clusters model fits its target. The proposed model represents three different types of fluid flow to get the dynamic behavior of the waterside and airside heat transfer coefficients, each of which is divided into seven clusters and has its unique TSF consequence. The overall fitness test of the proposed structure reveals significant and meaningful differences in the responses of the eleven independent variables that serve as its inputs. This model is used to investigate the cooling energy differences between three different types of controllers