一类约束优化问题的改进遗传算法

Modified Genetic Algorithm and Its Applications in Optimization Problems with Constraints

遗传算法是一种解决非线性无约束优化问题的搜索算法，对于约束优化问题通常采用罚函数法；所提出的改进算法很好地解决了一类带有凸约束的非线性优化问题。数值结果说明该方法效果较好。

Genetic algorithm(GA)is used for solving unconstrained optimization problems. Constraints may be handled indirectly through penalty functions. In this paper, a modified GA by a direct way to incorporate one kind of convex constraints is proposed.The fitness, crossover and mutation are designed especially for handling constraints. Numerical resultes show the algorithm is effective.