SOLUTION OF DYNAMIC OPTIMIZATION PROBLEMS BY SUCCESSIVE QUADRATIC-PROGRAMMING AND ORTHOGONAL COLLOCATION

Current strategies for optimization of dynamic systems usually require repeated solution of the differential equation model and may therefore be inefficient. This work explores the use of orthogonal collocation to reduce the dynamic optimization problem to an equality constrained nonlinear program (NLP). The NLP is then solved using a strategy that simultaneously converges and optimizes the algebraic model. Using a small example for comparison, significant reductions in computational effort are observed.