1 documents found
Information × Registration Number 0221U104526, 0120U002158 , R & D reports Title Develop effective methods for modeling, solving and investigating problems of trans-computational complexity on multiprocessor computing complexes popup.stage_title Head Sergienko Ivan V, д.ф.-м.н. Registration Date 29-03-2021 Organization V.M. Glushkov Institute of Cybernetics of National Academy of Sciences of Ukraine popup.description2 The research object is the problems of transcomputing complexity, e.g., discrete optimization, mathematical physics and bioinformatics. The purpose of the work is to develop and prove the methods of modeling and solving these problems on the multiprocessor computer systems. Two modifications of the iterative taboo algorithm are developed and analyzed for solving the quadratic assignment problem. Portfolios of algorithms are used for solving the quadratic assignment problem on the supercomputer. New necessary and sufficient stability conditions of three types are found for the problem of finding Pareto optimal solutions for a vector partially integer optimization problem with quadratic criterion functions. The decomposition method is developed and analyzed for solving the mixed integer bilinear programming problem with the preservation of the necessary information for the postoptimal analysis. New variational problems are formulated for modeling established temperature processes for bodies where the heat flux density and the elastically deformed state of spatial bodies are specified. A discrete version of the Bayesian procedure is developed for diagnosing brain pathologies based on indicators of surface plasma resonance. The developed mathematical models of the thermal-spring state of spatial bodies and their discrete analogs could form the basis for creating a software-algorithmic complex for designing the objects experiencing significant temperature and force effects (space objects, mechanical engineering objects, chemical industry object, nuclear reactors, etc.) Product Description popup.authors Gupal Anatoliy M Semenova Nataliia V Shylo Volodymyr P popup.nrat_date 2021-03-29 Close
R & D report
Head: Sergienko Ivan V. Develop effective methods for modeling, solving and investigating problems of trans-computational complexity on multiprocessor computing complexes. (popup.stage: ). V.M. Glushkov Institute of Cybernetics of National Academy of Sciences of Ukraine. № 0221U104526
1 documents found

Updated: 2026-03-25