Control-oriented identification using maximum likelihood for a district heating plant

The rapid development of model-based control strategies, such as Model Predictive Control (MPC), has risen the need to develop control-oriented models. In this paper, we develop a control-oriented model for a district heating plant, in the city of Tallinn, Estonia. The development of this model is carried out using Maximum likelihood estimation. To that end, we develop a linear model with all the available data, which has been taken at different seasons during the year. The model consider all the inputs to the heating plant, even though we have two uncontrollable inputs. Simulation results show that the proposed control strategy works well to control the aforementioned plant, and that the models obtained with ML estimation overperforms the ones obtained by using standard estimation techniques.