A knowledge based planning approach for waste water treatment system

Waste water treatment design can be identified as a planning problem. The waste water consists of several contaminants(compounds or chemicals) that need to be removed during the treatment process. A treatability database has been developed containing the treatability of various compounds through different types of treatment processes. Normally several chemicals appear together and we may need to use two or more treatment processes together to meet the treatment objectives. The waste water treatment system consists of two phases. In the initial phase an inductive learning algorithm ID3 is used to extract the knowledge rules from the database. These rules are represented together to identify the effect of an individual treatment process on several compounds at various concentrations. The final phase has been formulated as a planning problem. This phase has been designed as a heuristic search problem to generate the sequence of technologies satisfying the necessary constraints.