A Novel Mernetic Algorithm for Nonlinear Bilevel Programming with Inequality Constraints for the Upper Level

摘要

Nonlinear bilevel programming(NBP) is a nested optimization problem with hierarchical structure. In this paper, a special kind of (NBP) is studied. The lower level of (NBP) is transformed into an equivalent unconstrained optimization problem. To solve the equivalent problem effectively, a Hooke-Jeeves method to solve the lower level programming is presented once the leader makes a decision. By designing the crossover, mutation operator and selection strategy, a novel memetic algorithm is proposed. Simulations on 10 benchmark problems are made, effectiveness of the proposed algorithm is proven by comparing with other algorithms.