C. A. BAKER, L. T. WATSON, B. GROSSMAN, W. H. MASON, S. E. COX, AND R. T. HAFTKA, Study of a global design space exploration method for aerospace vehicles. preprint, 1999.  Home/Search   Document Details and Download   Summary   Related Articles

....version are sequences of these sweeps. There is little convergence theory for DIRECT beyond the observation from [14] that the search will eventually sample arbitrarily near every point in Omega . The method has been applied to optimal design of gas pipe lines [3 5] and aerospace engineering [1, 2] and seems to perform well, especially in the early stages of an optimization. In this paper we quantify how the subdivisions cluster near a global minimizer and use this result to motivate an alternative version of DIRECT [8] which is different from that in [13, 14] in that ffl = 0, the l 1 ....

....the search more toward local improvement can reduce the cost of a sweep and more rapidly identify the global minimum. Such a modification is likely to be more useful for small N , as for larger N more work will be needed in the global search even to explore the design space at a coarse level [1, 2]. A globally biased version of direct, called aggressive DIRECT, was proposed in [1] In that approach the potential optimality condition is abandoned and the hyperrectangles with the lowest function value in each group are all subdivided. The formulation from [8] differs from that in [13, 14] in ....

C. A. BAKER, L. T. WATSON, B. GROSSMAN, W. H. MASON, S. E. COX, AND R. T. HAFTKA, Study of a global design space exploration method for aerospace vehicles. preprint, 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC