Parameter selection using an approximated performance map
James J. Alpigini · 2002
The performance map employs intuitive color-coding to visualize the behavior of system dynamics resulting from variations in system parameters. The resulting image is developed algorithmically, via digital computation, requiring only moderate a priori knowledge and mathematical analysis provides an immediate wealth of information to the knowledgeable viewer. The computational overhead of these maps is considerable commonly requiring more than 4/spl times/10/sup 8/ integration steps. A modification to the performance map technique is presented. The new visualization, called an approximated performance map, is generated with significant computational savings. While approximation of any form will result in loss of information, such losses are indeed acceptable in many circumstances, such as defining stable versus unstable manifolds.