The Use of Quantitative Databases in Aladin, an Alloy Design System.

摘要: Abstract : The Aluminum Alloy Design Inventor (Aladin) is an expert system that aids in the design of new alloys. uses traditional, rule based, qualitative methods as well quantitative calculations. Our approach to model process a collaboration among several different bodies expertise: composition terms elements be added; properties known alloys; internal sturcture, ranging from solid phases and microstructure down crystal lattice structure; thermo-mechanical processing metal. inference mechanism variant hypothesize-and-test. present implementation combines Carnegie Representation Language (CRL) OPS5 along with Lisp. A database metallurgical knowledge data on alloys built form CRL schemata. execution controlled by OPS productions most produced specific run represented working memory elements. This paper gives preliminary overview architecture Aladin emphasis couplng symbolic numeric computations. We breifly describe representation Alloys, Phase diagrams Microstructure information. explain how reasoning used define variables dealt quantitatively regression used. multidimensional constraints they are deal interaction between variables, implement least commitment strategy at same time provide criteria for search termination backtracking.