基于评价函数的遗传算法求解非线性规划问题

针对具有待式约束和非等式约束的非线性规划问题，通过引进准可行方向、主导准可行方向和可行度等概念，提出描述和度量非可行点（梁色体）的新方法；通过嵌入非可行染色体的信息于评价函数，提出３种改进的评价非可行染色体的新方法；基于新的评价函数方法，提出一种沿权重梯度方向变异的遗传算法（ＥＧＡ）。对测试问题的仿真结果表明了ＥＧＡ算法的有效性。

Evaluation-based Genetic Algorithm for Non-linear Programming Problem

By embedding the information of infeasible points/chromosomes into the evaluation function, three improved evaluation functions are designed to formulate and evaluate the infeasible chromosomes. On the basis of introducing concepts of improved version of semi feasible direction, dominated semi feasible direction and feasibility degree etc, an evaluation based genetic algorithm (EGA) is developed for solving non linear programming (NLP) problems with inequality and equality constraints. Simulation of test problems shows that this algorithm is efficient.