A Compound Scaling Algorithm for Mathematical Optimization
Vipperla B. Venkayya, Victoria Tischler · 1989
Abstract : This paper derives a compound scaling algorithm from a single scaling algorithm for use with general mathematical optimization problems. Compound scaling is necessary when the constraints are nonlinear and the variables do not all belong to the active set. Compound scaling in conjunction with the optimality criteria method has practical applications in large scaled multidisciplinary design. It will extend the scope of the optimality criteria method to handle problems with thousands of variables and constraints. The development of simple scaling is included as a basis for the development of the compound scaling algorithm. Five problems were selected to demonstrate the effectiveness of the simple and compound scaling algorithms.