The development of controllers for a multiple-effect evaporator in sugar industry

One of the most important task in sugar production process is to ensure an optimum working regime for the multiple-effect evaporator. In this paper we present the development of a physical model of a five-effect evaporator process for analysis and synthesis of different control schemes. A number of feedback and feedforward strategies are investigated. These include model-based nonlinear and linear approaches. A structural analysis shows that the physical model of the evaporation process cannot be classified into different control classes. Model analysis of the process is carried out by employing the frequency domain and state space. The technique of linearization by static state feedback is studied by using differential geometric approaches. The generic model control (GMC) is also used for control of evaporation process. Three types of model-based linear approaches are applied, the generalized predictive control (GPC), the linear quadratic gaussian (LQG) and the internal model control (IMC). Comprehensive comparative studies are carried out.