研究非线性均衡问题,引入一个磨光算子将原问题转化为光滑问题,并用此光滑问题来逼近原来的问题而求解.在每步迭代中,通过转轴运算,求解一个线性约束二次规划问题和显式修正方向来得到主方向,并通过一个显式公式来得到高阶修正方向使得算法避免Maratos效应.在不需要上层互补备件下证明了算法具有全局收敛性和强收敛性且具有超线性收敛速度.
In this paper,mathematical programs with nonlinear equilibrium constraints are considered.By means of F-B function,the equilibrium constraints are transformed into a nonsmooth equation.Combining the idea of approximation with SQP method,a new approximate SQP method with explicit modify direction for mathematical programs with nonlinear equilibrium constraints is presented,and its global convergence and superlinear convergence are obtained without the upper-level strict complementarity condition.