A continuously differentiable exact penalty function for nonlinear programming problems with unbounded feasible set (Q1319684): Difference between revisions

A new continuously differentiable exact penalty function for the solution of nonlinear programming problems is defined. The structure of the penalty function is derived from the penalty functions proposed by the second author and \textit{L. Grippo} [Math. Program. 36, 1-18 (1986; Zbl 0631.90061)] and the third author [SIAM J. Optim. 2, No. 4, 558-574 (1992; Zbl 0761.90089)]. The authors give exactness properties of the new penalty function and some numerical results.