Application of ńvidia CUDA technology in resolving complex issues of dynamic optimization

• Autorzy: Pala A.
• Języki publikacji: EN

Abstrakty

EN

This paper presents analysis of possibility to apply programmable graphics circuits based on Nvidia CUDA technology in solving problems of dynamie optimization. Application of optimization methods based on utilization of Pontryagin’s maximum principle leads to the need of solving extremely large systems of state cquations as well as adjoint equations. Using GPGPU graphics cards for this purpose can significantly speed up process of solving this kind of equations.

• Wydawca: Oficyna Wydawnicza Politechniki Opolskiej
• Czasopismo: Zeszyty Naukowe. Elektryka / Politechnika Opolska
• Rocznik: 2012
• Tom: z. 68
• Strony: 51--52
• Opis fizyczny: Bibliogr. 7 poz.

Twórcy

autor

Pala A.

• Faculty of Electrical Engineering Automatic Control and Informatics Institute of Control and Computer Engineering

Bibliografia

• [1] FINDEISEN W„ SZYMANOWSKI J., WIERZBICKI A.: Metody obliczeniowe optymalizacji, Wydawnictwo Politechniki Warszawskiej 1973
• [2] NVIDIA CORPORATION: Tesla M-Class GPU Computing Modules, Nvidia Corporation 2011
• [3] NVIDIA CORPORATION: Nvidia Tesla GPU Computing, Nvidia Corporation 2010
• [4] FORENC J.: Analiza spekulacyjna stanów nieustalonych w układach elektrycznych, Rozprawa doktorska, Politechnika Białostocka 2006
• [5] BURRAGE K.: Parallel and Sequential Methods for Ordinary Differential Equations. Clarendon Press, Oxford, 1995
• [6] PETCU D: Parallelism in Solving Ordinary Differential Equations. Mathematical Monographs 64, Timisoara University Press, 1998
• [7] GEAR C.W.: Massive Parallelism Across Space in ODEs. Applied Numerical Mathematics 11, 1993