A fuzzy process controller for in situ groundwater bioremediation

Dynamic operation control for in situ groundwater bioremediation requires expertise in both areas of bioremediation and biochemistry. This paper presents an artificial intelligence aided process control system that incorporates an on-line expert system. The fuzzy controller controls the pumping rate based on the measured pollution level. Through analysis of knowledge from domain experts, a model of interval-parameter has been developed from which fuzzy information and fuzzy rules can be derived. There are two levels of fuzzy rules in the hierarchical controller. The lower level can be further divided into two branches. Each branch is divided into three layers based on the employed artificial intelligence techniques and the system inputs and outputs. The system was applied to a real-world case study in western Canada. The results indicated that the developed artificial intelligence aided control system could help improve efficiencies of in situ bioremediation at petroleum-contaminated groundwater systems. (C) 2003 Elsevier Ltd. All rights reserved.