A global optimization algorithm using linear relaxation (Q2507847)

A global optimization algorithm is presented for non-convex quadratically constrained quadratic programming. A linear method is applied to the initial problem. The algorithm is shown to attain the global minimum through the successive refinement of a linear relaxation of the feasible region and of the objective function and the subsequent solution of a series of linear programming. The proposed approach is applied to the three problems, and convergence of the global minimum is achieved. Numerical results are given to illustrate the feasibility and effectiveness of the present algorithm.