초록

용어보기논문에서 용어와 풀이말을 자동 추출한 결과로, 시범 서비스 중입니다. AI-Helper

Abstract ▼ AI-Helper

Quadratic programming problems(QP) have been widely used as a direction-finding subproblem in the engineering and structural design optimization. To develop an efficient solution algorithm for the QP subproblems, theoretical aspects and numerical behavior of mathematical programming methods that can...

Quadratic programming problems(QP) have been widely used as a direction-finding subproblem in the engineering and structural design optimization. To develop an efficient solution algorithm for the QP subproblems, theoretical aspects and numerical behavior of mathematical programming methods that can be used as QP solver are studied and compared. For the solution of both primal and dual QP, Simplex, gradient projection(GRP), and augmented Lagrange multiplier algorithms are investigated and coded. From the numerical study, it is found that the primal GRP algorithm with potential constraint strategy and the dual Simplex algorithm are more attractive and effective than the others. They have theoretical robustness as well. Moreover, primal GRP algorithm is preferable in case the number of constraints is larger than the number of design variables. Favorable features of GRP and Simplex algorithm are merged into a combined algorithm, which is useful in the structural design optimization.