Introducing a new nonlinear approach to model heat exchangers designed for control engineering applications

A new low-order dynamic model is proposed to predict the nonlinear heat transfer in heat exchangers, using a modeling approach that does not require detailed information about the flow arrangement. On the primary side, a model of a throttling valve is added to include the control signal's influence on the secondary side's output temperature. The model does not consider time delays but uses variable time constants that depend on the mass flow rates. A two-step procedure is proposed to estimate the model parameters: first, the complete parameter vector is estimated to solve a parameter estimation problem. Then, a subset of the estimated parameters is tuned online using an Unscented Kalman Filter to fit the model further to reality. The accuracy of the model as well as the implementation of the parameter estimation are demonstrated using an example from practice.