Intelligent supervisory-level control of industrial processes

Faced by intense competition, the pulp and paper industry is aiming for higher profitability by raising productivity and lowering costs. On the global scale, the industry is having to meet increasing market demands for higher product quality, more-specialty products, greater production flexibility and a higher standard of environmental protection. For these reasons, extensive research is being conducted to improve existing processes. One option that is gaining increasing attention within the industry is to improve control of existing processes by means of intelligent systems. In this paper;an intelligent kiln control system is presented. The system consists of two parts: the stabilizing control is based on a new LECont concept which uses a novel linguistic equation approach, while the targets are determined by a StabOpti concept on the optimization level. Nonlinear models obtained from process data are used for optimization and compensation of load disturbances. According to test results from an industrial pulp mill, operational flexibility was improved and both economic and ecological benefits were obtained. The overall control system provides a better insight into the process and extends the scope of both fuzzy logic control and advanced extensions of classical control.