Accession Number : ADA310373
Title : Memory-Based Computational Intelligence for Materials Processing and Design.
Descriptive Note : Final rept. 1 Dec 92-1 Dec 95,
Corporate Author : CASE WESTERN RESERVE UNIV CLEVELAND OH
Personal Author(s) : Pao, Y. H.
PDF Url : ADA310373
Report Date : APR 1996
Pagination or Media Count : 56
Abstract : The work reported in this document is concerned with the efficient use of computers in materials research and in applications of the results of that research. Emphasis is on the development of computational methodologies which can facilitate the innovative design of materials and of materials processing, for high performance materials and for composite materials structures. Basic advances have been made in three areas of adaptive computing: in establishing the practice of functional-link neural-net computing for learning models of material behavior, in developing a parallel processing evolutionary search paradigm for optimization and in exploring various of establishing and using associative memories. This report also describes how these ovations in conceptual and theoretical matters can be used to effect in dealing with materials processing and design tasks. Applications include in-situ - real-time interpretation of ellipsometry data, optimal formulation of material composition, control of nonlinear dynamic systems, memory-based design and the implementation of visual displays for multi-dimensional data.
Descriptors : *NEURAL NETS, *PRODUCTION ENGINEERING, ALGORITHMS, OPTIMIZATION, DATA MANAGEMENT, COMPUTER AIDED DESIGN, DISTRIBUTED DATA PROCESSING, MONITORING, REAL TIME, COMPOSITE MATERIALS, EFFICIENCY, INPUT OUTPUT PROCESSING, PARALLEL PROCESSING, COMPOSITE STRUCTURES, NONLINEAR SYSTEMS, DATA REDUCTION, INFORMATION RETRIEVAL, ADAPTIVE SYSTEMS, ASSOCIATIVE PROCESSING, PATTERN RECOGNITION, ARTIFICIAL INTELLIGENCE, SYSTEMS ANALYSIS, COMPUTER AIDED MANUFACTURING, CONTROL THEORY, STOCHASTIC CONTROL, SELF ORGANIZING SYSTEMS, FRESNEL INTEGRALS.
Subject Categories : Cybernetics
Mfg & Industrial Eng & Control of Product Sys
Distribution Statement : APPROVED FOR PUBLIC RELEASE