Robust model predictive control for heat exchanger network

Optimal operation of heat exchangers represents a challenging task from the control viewpoint, due to the presence of system nonlinearities, varying process parameters, internal and external disturbances and measurement noise. Various robust control strategies were developed to overcome all these problems. The robust model predictive control (RMPC) represents one of suitable approaches. It enables to design effective control algorithms for optimization of the control performance subject to the process uncertainties and the input and output constraints. The possibility to implement the RMPC for control of a heat exchanger network is investigated in this paper, where three counter-current heat exchangers with uncertain parameters connected in series represent the controlled process. The efficiency of the advanced RMPC algorithm was verified by simulation experiments realized in the MATLAB/Simulink environment. The results confirmed that using the RMPC for the controlled process modelled as a system with uncertain parameters led to less consumption of cooling medium compared with the consumption achieved by using the optimal linear quadratic (LQ) control. •The robust and the optimal controllers control the heat exchanger network.•The robust model predictive controller is a solution of linear matrix inequalities.•Robust control assures smaller offsets and lower coolant consumption.