Abstract
An optimizing, adaptive process control system for a phosphate processing plant has been designed and tested using actual data from an operating phosphate processing plant. The level-2 intelligent control system consists of three key elements: (1) a control element, (2) an adaptive element, and (3) a learning element. The controller employs a neural network model of the phosphate plant, a fuzzy logic control element for manipulating the plant, and a genetic algorithm-based search engine. This report presents key background information regarding intelligent control systems, describes the overall architecture of the control system developed, designed, and tested, and presents results demonstrating the effectiveness and potential of the control system. Finally, recommendations are made regarding directions for future development and implementation of optimizing, adaptive control systems in the phosphate industry.
Charles L. Karr and Bernard J. Scheiner - BCD Technologies