Accession Number : ADA322778

Title :   Self-Optimization in Simulation-Aided High Power Microwave Source Design.

Descriptive Note : Final rept. 1 Oct 94-30 Sep 96,

Corporate Author : PHILLIPS LAB KIRTLAND AFB NM

Personal Author(s) : Arman, M. J. ; Havranek, James ; Dietz, David

PDF Url : ADA322778

Report Date : 28 FEB 1997

Pagination or Media Count : 22

Abstract : Computer-aided simulation techniques are essential in the design and development of new High Power Microwave sources. Using simulation codes to optimize a process is very slow, however, due to the iterative nature of the optimization, and the true optimal values are never found because the parameters are optimized one at a time instead of the simultaneous minimization process needed for multivariable systems. We have developed a new method that automatically finds the real optimal values by simultaneously minimizing the proper physical quantity versus the relevant variables before the simulation is carried out. There will be no need for repeated execution of nearly identical simulations. The proposed approach will speed up the process at least ten times. A similar approach can be used to speed up non-Particle-In-Cell simulation-aided optimization processes that will benefit the research community in all sectors.

Descriptors :   *COMPUTERIZED SIMULATION, ALGORITHMS, OPTIMIZATION, COMPUTATIONS, MONTE CARLO METHOD, COMPUTER APPLICATIONS, MAN COMPUTER INTERFACE, APPLIED MATHEMATICS, PLASMA GENERATORS, MICROWAVE TRANSMISSION, RADIOFREQUENCY OSCILLATORS.

Subject Categories : Computer Programming and Software

Distribution Statement : APPROVED FOR PUBLIC RELEASE