离散非线性规划问题的改进遗传算法

针对实际离散非线性规划问题，分析了离散与连续变量优化问题和求解方法的不同及特性．根据离散变量与遗传算法的特点，将单纯形搜索与算术交叉思想相结合，提出离散单纯形交叉算子以提高遗传算法的局部寻优能力，将种群逐步向离散极值点进行引导，实现算法的快速离散寻优．同时，设计了离散变异算子，使遗传算子真正在离散空同中进行搜索．基于梯度下降思想提出离散修复算子，提高算法对非线性约束的处理能力．实际离散非线性规划问题的应用研究验证了方法的有效性．

Improved Genetic Algorithm in Discrete Variable Non-linear Programming Problems

For discrete variable non-linear programming problems in practice, characteristics of discrete variable optimization and continuous variable optimization and their solving methods are analyzed. Based on features of discrete variables and genetic algorithm Simplex searching and arithmetic crossover are combined and a discrete Simplex crossover operator is presented to improve the local searching capability of the genetic algorithm. This method leads the population to local optimization and produces rapid discrete searching. At the same time, discrete mutation operator is presented to make the genetic operator search in discrete space. Discrete repair operator is presented on the basis of the theory of gradient to improve the performance of the algorithm in solving non-linear constraint. The application of practical discrete variable non-linear programming problems shows the validity of this algorithm.