First Results on Modelling of a Plate Heat Exchanger of a District Heating System

This paper describes the first results of the modelling of a plate heat exchanger which is installed in the VUB district heating network. A properly controlled energy exchange through the heat exchanger improves the economy of the entire system, minimizes the pollutant emissions and fossil fuel consumption, in accordance with the main goals of the European energy plan till 2030. To achieve these goals, an accurate model of the plate heat exchanger is required. In this article two approaches are considered to accurately model the heat exchanger: a physics-based white-box and a data-based black-box modelling technique. The white-box modelling approach uses the unsteady energy balance and heat transfer equations to predict the temperature evolutions in the plate heat exchanger. For the (data-driven) black-box modelling approach, there are three interrelated steps. Step 1 is the (semi-automatic) preprocessing of the measurements and the building of a nonparametric frequency response function (FRF) model. Step 2 is the building of a linear parametric (state-space) model, based on the FRF estimate. Step 3 is building a polynomial nonlinear state-space model that relies on the original linear state-space model. The proposed black-box method outperforms the white-box techniques in terms of prediction capabilities and computational needs.