Model-based Control of Heat Exchange Process

Control of cascade-connected heat exchangers aiming to firmly maintain the value of output variable within set boundaries has been researched in this paper. Traditional control realized through automatic stabilization of the temperatures has been extended with originally designed control system by setting the set points using model based control. A warm liquid with a low boiling point is fed into a distillation column where precise maintenance of the temperature of the fed liquid close to boiling point is required. In the first stage, the liquid is heated up to a temperature close to the desired output value, and in the second stage the final value is obtained within a permitted range of +/- 0.3 degrees C around the set-point. When the pressure change occurs in the distillation column, the disturbance produces undesirable material and heat imbalance. In order to solve the task, a traditional process control has been conducted, thus gathering the data needed for neural network application. Improvement in traditional process control has been made through an aplication of neural network that represents an inverse process model behaving in a feed forward principle. In order to realize improvement in performance of the control system, a process control method has been developed using generic model control based on the neural network model of the process.