An SQP Method with a Revised Non--Monotone Line Search
一种修改的非单调线搜索SQP算法

In the paper introduces a kind of revised non--monotone line search SQP method for constrained NLP problems is introduced, and the global convergence of this method is proved without using a penalty function as a merit function, a filter or the feasibility restoration phase. This method is based on the concept of multi--objective optimization: a trial point can be accepted if and only if either object function value decreases or the measure of violation constraints decreases. Numerical results, compared with LANCELOT, show that the approach is effective.