RULE-BASED SYNTHESIS OF SEPARATION SYSTEMS BY PREDICTIVE BEST FIRST SEARCH WITH RULES REPRESENTED AS TRAPEZOIDAL NUMBERS

This paper presents a new rule based systematic screening procedure fo r synthesis of separation systems. Rules are represented as trapezoida l numbers, enabling partial pattern matching and graded outcome of rul es containing continuous variables. Instead of performing conventional conflict resolution on a rule set in order to determine which rule to execute next, it is here proposed simultaneously to evaluate the grad ed outcome of all rules in the rule base and feed the result of this e valuation to a predictive best-first search mechanism. Clean separatio n between pattern matching and search is thus achieved. An admissible prediction function has been found, making it possible to guarantee th at the search algorithm terminates with optimal solutions when the sha rp split search space is restricted to simple separators. The proposed approach is based in part on qualitative knowledge about separation p roblems. The search algorithm has, therefore, been designed to produce a user specified number of alternative best solutions, which may subs equently be evaluated in more detail using more rigorous models. Examp les are included to illustrate the suggested approach.