给出了多连接查询优化问题的计算模型,分析了免疫遗传算法的基本原理,提出将免疫遗传算法应用于多连接查询优化问题。针对多连接查询优化问题的具体特点,给出了免疫遗传算法的设计,包括亲和度、适应度函数的设计,基于抗体浓度的选择算子、交叉算子、变异算子的设计,免疫算子的设计。

Gives a multi-join query optimization calculation model,analysis of the basic principles of immune genetic algorithm,propose immune genetic algorithm applied to multi-join query optimization problem. Against multi-join query optimization problem specific characteristics,gives the immune genetic algorithm design,including affinity, fitness function design,concentration of antibody-based selection operator, crossover operator, mutation operator design ,immune operator design.

针对线材排样优化问题，在分析线材排样方案数学模型的基础上，提出了基于改进遗传算法的优化求解方案。该算法设计了一种新颖的遗传算子，包括顺序交叉算子、线性变异算子、扩展选择算子。实验表明，该算法逼近理论最优值，而且收敛速度快，较好地解决了线材排样问题。

Basing on the analysis of wire layout mathematical model ,aiming at the wire layout optimization problem , this paper presents an optimization scheme on the basis of genetic algorithm .The algorithm designs a kind of new genetic operator ,including ordered crossover operator ,linear mutation operator and expand‐selection operator .The experiment shows that the algorithm approaches the theoretical optimal value ,its convergence rate is quick ,and is efficient to solve the wire layout problem .

文中针对函数优化方面遗传算法(GA)存在的"早熟"与收敛速度慢的问题,设计了一种基于交叉和变异算子改进的遗传算法。通过研究分析GA,根据交叉算子和变异算子的特点,在现有的GA基础上,引入拉普拉斯算子改进交叉算子以及结合黄金分割法对变异算子做了进一步改进。通过3个测试函数对该算法与标准遗传算法,以及其他两种算法加以对比,仿真结果表明文中的算法不仅增加了个体多样性,防止了"早熟",且比其他三种算法获得了更优解和更快的收敛速度。理论分析和实验表明,提出的算法是可行有效的。

Aiming at problems of genetic algorithm in terms of function optimization exists the"premature" and slow convergence,a kind of improved genetic algorithm based on crossover and mutation operators is designed. Through research and analysis of GA,according to the characteristics of the crossover operator and mutation operator,on the basis of existing GA,introduce the Laplacian operator to im-prove crossover operator and further improve mutation operator combined with golden section method. Through the three test functions to compare this algorithm with standard Genetic Algorithm ( GA) ,and the other two algorithms,simulation results show that the proposed algorithm can not only increase the diversity of individuals,preventing the "premature",and more than the other three algorithms gain better solution and faster convergence. Theoretical analysis and experimental results show that the proposed method is feasible and effec-tive.

建立了弹药运输车辆调度问题的数学模型，针对传统遗传算法求解该问题具有收敛速度慢、易陷入局部极小的缺点，提出了一种改进的遗传算法予以求解。在改进算法中引入一种基于信息素的遗传交叉算子，该算子能利用以信息素形式保存的全局信息，从而提高收敛速度；算法中的变异算子采用 Relocation、Ex-change、2-opt*及2-opt 4种启发式搜索算法，尽可能扩大搜索范围。算例分析表明了所提改进遗传算法求解弹药运输车辆调度问题的有效性和可行性。

The mathematic model is built for scheduling problem of ammunition transport vehi-cles ,and an improved genetic algorithm is put forward to solve the problem considering the defects of the traditional genetic algorithm in solving the problem w hich include the slow convergent speed and easy convergence to a local minimum point .A genetic crossover operator based on pheromone is intro-duced in the improved algorithm ,and the operator can utilize the global information saved by the form of pheromone to improve the convergence speed .The mutation operator in the algorithm adopts four heuristic search algorithms to extend the search scope as much as possible which are Relocation ,Ex-change ,2-opt* and 2-opt .The example analysis shows that the proposed and improved genetic algo-rithm is effective and feasible in solving the scheduling problem of ammunition transport vehicles .

本文首先对于罚函数遗传算法构造了合适的适应度计算方式,其次将适当的修补算子加入修补遗传算法中,保证修补的随机性和有效性;然后在两者的交叉、变异操作之后都加入进化突变算子,增强了他们的局部搜索能力;最后针对不同规模的基站选址问题,分别采用加入进化突变前后的罚函数遗传算法和修补遗传算法进行仿真,结果验证加入进化突变的修补遗传算法在求解大规模的基站选址问题时效率最高。

In the light of the characteristics of base station location for constrained optimization problems, this paper uses the penalty function genetic algorithm and repair genetic algorithm to solve the problem in compare.First of all, appropriate iftness calculation was constructed by the penalty function genetic algorithm , not only reduces the amount of calculation, also improves its generality.Next, it guarantees the randomness and effectiveness of repair by adding the appropriate repair operator to repair genetic algorithms.Then, they can enhance their ability of local search by adding evolution mutation operator after crossover and mutation operation. Finally, according to different sizes of base station location problems, it uses repair genetic algorithm and penalty function genetic algorithm before and after evolutionary mutation to simulation respectively, it veriifes the efifciency of repair genetic algorithm that adds evolutionary mutation in solving large-scale base sta