Intelligent control system based on blackboard concept for wastewater treatment processes

An intelligent control system for wastewater treatment processes has been developed and applied to fullscale, high-rate, activated sludge process control. In this control system, multiple software agents that model the target system using their own modeling method collaborate by using data stored in an abstracted database named ‘blackboard’. The software agents, which are called ‘expert modules’, include a fuzzy expert system, a fuzzy controller, a theoretical activated sludge model, and evaluators of raw data acquired by various online sensors including a respirometer. In this paper, the difficulties of controlling an activated sludge system by using a single conventional strategy are briefly reviewed, then our approach to overcome these difficulties by using multiple modeling methods in the framework of an ‘intelligent control system’ is proposed. Case studies of applications to a high-rate activated sludge process that treats BOD and nitrogen of human excrement are also presented.