遗传算法的一种特例——正交试验设计法

简要介绍正交试验设计法与遗传算法的基本原理,分析它们之间的内在关系,指出正交试验设计法可以认为是遗传算法的一种特例,即它是一种初始种群固定的、只使用定向变异算子的、只进化一代的遗传算法.计算结果表明,正交试验设计法可以解决一般遗传算法中的最小欺骗问题.     

一种改进的遗传算法Scatter GA

介绍一种改进的扩散式遗传算法 Scatter GA(Sc GA)。在原简单遗传算法 Sim ple GA(SGA)的基础上进行局部改进 ,采用更接近问题实际的实数编码方式 ,增加了第 2个变异算子 ,取消了选择算子。使用改进的算法对 4个典型函数进行计算 ,并与 SGA,m icro GA(m GA) ,Steady State GA(SSGA)的结果进行比较 ,可以看出 ,改进算法在收敛速度和精度上均优于其它同类算法。     

双变异算子遗传算法的应用

针对简单遗传算法(SGA)所存在的缺点和不足,提出了一种新的改进遗传算法一双变异算子GA.该算想法通过将所有产生的子代个体与父代个体混合作为下一代种群,在种群选择前对适应度值较低的个体进行一次变异,然后通过选择、交叉,再一次变异产生新种群,再利用自适应算法改变交叉和变异率及最优保存策略保护历代最优个体,利用matlab软件编程计算,在TSP中得到了较好的优化结果.实例说明,双变异算子的遗传算法能够最大限度使种群多样性,这样最有可能得到最优解,也易突破局部收敛的局限而达到全局最优.     

基于遗传算法的立体车库车位调度研究

立体车库的车位调度是一个比较复杂的问题,遗传算法是搜索立体车库最短路径的有效方法之一.对传统的GA结构加以改进,利用一种改良的OX交叉算子加快算法的收敛速度,利用变换变异算子维持群体的多样性防止算法早熟收敛.仿真实验结果验证了算法的有效性.     

双变异算子遗传算法的应用

针对简单遗传算法（SGA）所存在的缺点和不足,提出了一种新的改进遗传算法一双变异算子GA.该算想法通过将所有产生的子代个体与父代个体混合作为下一代种群,在种群选择前对适应度值较低的个体进行一次变异,然后通过选择、交叉,再一次变异产生新种群,再利用自适应算法改变交叉和变异率及最优保存策略保护历代最优个体,利用matlab软件编程计算,在TSP中得到了较好的优化结果。实例说明,双变异算子的遗传算法能够最大限度使种群多样性,这样最有可能得到最优解,也易突破局部收敛的局限而达到全局最优。     

遗传算法在旅行商问题中的实践与探究

TSP问题是组合优化领域的经典问题之一,旨在求出遍历若干个城市的最短路径。本文通过遗传算法GA的选择和变异算子的确定和、交叉算子的改进,并在TSP问题中的实践来探索这个经典的NP(Nondeterministic Polynomial)难题。     

一种基于选择的遗传算法

鉴于标准遗传算法比较容易产生早熟现象和模式欺骗而收敛于局部最优解,论文对标准遗传算法的遗传操作进行了改进,提出了基于选择的遗传算法(GA_S)。在该算法中,首次提出了基因选择算子、广义精英算子、引进选择算子、基于精英集的成长期变异等概念,并对其进行了比较详细的描述。之后,使用7个经典测试函数对其进行了大量实验。实验表明算法对早熟和模式欺骗具有较强的突破能力。     

基于混沌“微变异”自适应遗传算法

遗传算法可以较好地解决复杂的组合优化问题,但也存在两方面不足:一是搜索效率比其他优化算法低;二是容易过早收敛,陷入局部最优.对此,提出一种混沌“微变异”遗传算法.利用混沌优化算法具有随机性和遍历性的特点,解决遗传算法容易陷入局部最优解的早熟问题,使得新算法同时具有较强的局部搜索能力和完成全局寻找最优解的能力.同时,对遗传算法的选择算子增加了混沌扰动,对交叉算子和变异算子进行自适应调整,对适应度函数进行改进,使遗传算法整体性能得到提高.最后,通过经典函数验证表明,混沌“微变异”遗传算法比一般的混沌遗传算法和经典遗传算法的进化速度更快,搜索精度更高.     

一种改进选择算子的遗传算法

遗传算法（Genetic Algorithm,GA）是一种模拟生物进化的智能算法,被广泛应用于求解各类问题。简单遗传算法（Simple GA）仅靠变异产生新的数值,常常存在搜索精确度不高的问题。针对这个问题,对SGA的选择算子进行改进,即把相似个体分在同一组中,以组为单位进行选择,并通过该组个体的特点进行高斯搜索生成新的群体。这样使得GA在搜索过程中不仅可以很好地保持个体的多样性,并且可以提高解的精确度。通过对11个函数（单峰和多峰）的仿真实验,证明了采用新的选择算子后,GA在求解问题的精确度上有了很大地改善。     

基于"优胜劣汰"原则的差异算子

借鉴生物学中“优胜劣汰”的原则,引入一种新的遗传算子,从而对传统的遗传算法（GA）进行改进.该算子的引入达到了扩大搜索空间、提高收敛速度、保持群体中个体多样性的目的.通过函数优化测试,结果表明：算子提高了GA对全局最优解的搜索能力和收敛速度.进一步对其相关参数设置的研究,将会使GA在众多实际的优化问题上具有更广泛的应用前景.     

Genetic algorithm with fuzzy logic controller for preemptive and non-preemptive job-shop scheduling problems

In this paper, we propose a new genetic algorithm (GA) with fuzzy logic controller (FLC) for dealing with preemptive job-shop scheduling problems (p-JSP) and non-preemptive job-shop scheduling problems (np-JSP). The proposed algorithm considers the preemptive cases of activities among jobs under single machine scheduling problems. For these preemptive cases, we first use constraint programming and secondly develop a new gene representation method, a new crossover and mutation operators in the proposed algorithm.However, the proposed algorithm, as conventional GA, also has a weakness that takes so much time for the fine-tuning of genetic parameters. FLC can be used for regulating these parameters.In this paper, FLC is used to adaptively regulate the crossover ratio and the mutation ratio of the proposed algorithm. To prove the performance of the proposed FLC, we divide the proposed algorithm into two cases: the GA with the FLC (pro-fGA) and the GA without the FLC (pro-GA).In numerical examples, we apply the proposed algorithms to several job-shop scheduling problems and the results applied are analyzed and compared. Various experiments show that the results of pro-fGA outperform those of pro-GA.     

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem

The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.     

基于改进遗传算法的船舶管路布局设计

针对船舶管路布局设计中的路径规划问题提出一种改进型遗传算法求解方法。建立船舶管路布局设计问题的模型空间、约束条件和优化目标;提出一种基于连接点网格的定长编码方法,结合该编码方法设计了适合改进遗传算法应用的适应度函数和交叉、变异算子,定长编码可降低遗传算子设计复杂度和非法个体修补代价;提出在进化流程中嵌入以“去折弯”和“改模式”两种改善型变异方法构建的爬山操作,以提升算法收敛性和寻优能力。通过仿真实验验证所提算法具有可行性和先进性。     

Dynamic path planning of mobile robots with improved genetic algorithm

In this study, a new mutation operator is proposed for the genetic algorithm (GA) and applied to the path planning problem of mobile robots in dynamic environments. Path planning for a mobile robot finds a feasible path from a starting node to a target node in an environment with obstacles. GA has been widely used to generate an optimal path by taking advantage of its strong optimization ability. While conventional random mutation operator in simple GA or some other improved mutation operators can cause infeasible paths, the proposed mutation operator does not and avoids premature convergence. In order to demonstrate the success of the proposed method, it is applied to two different dynamic environments and compared with previous improved GA studies in the literature. A GA with the proposed mutation operator finds the optimal path far too many times and converges more rapidly than the other methods do.     

The schema deceptiveness and deceptive problems of genetic algorithms

Genetic algorithms (GA) are a new type of global optimization methodology based on na-ture selection and heredity, and its power comes from the evolution process of the population of feasi-ble solutions by using simple genetic operators. The past two decades saw a lot of successful industrial cases of GA application, and also revealed the urgency of practical theoretic guidance. This paper sets focus on the evolution dynamics of GA based on schema theorem and building block hypothesis (Schema Theory), which we thought would form the basis of profound theory of GA. The deceptive-ness of GA in solving multi-modal optimization problems encoded on {0,1} was probed in detail. First, a series of new concepts are defined mathematically as the schemata containment, schemata compe-tence. Then, we defined the schema deceptiveness and GA deceptive problems based on primary schemata competence, including fully deceptive problem, consistently deceptive problem, chronically deceptive problem, and fundamentally decepti     

改进遗传优化的贝叶斯网络结构学习

针对贝叶斯网络结构学习提出了一种改进的遗传算法,和传统遗传算法相比,该改进算法针对贝叶斯网络结构学习问题增加了优化变异和修正非法图两个新的算子。新算子不但保持了贝叶斯网络学习的多样性和正确性,而且还能保证算法快速搜索到全局最优的网络结构。将该改进遗传算法用于贝叶斯网络结构学习的仿真结果表明,和传统K2算法、GS／GES算法、遗传算法和粒子群算法等算法相比,该算法具有更好的全局搜索能力和收敛速度。     

Iterated fast local search algorithm for solving quadratic assignment problems

This paper concentrates on multi-row machine layout problems that can be accurately formulated as quadratic assignment problems (QAPs). A genetic algorithm-based local search approach is proposed for solving QAPs. In the proposed algorithm, three different mutation operators namely adjacent, pair-wise and insertion or sliding operators are separately combined with a local search method to form a mutation cycle. Effectiveness of introducing the mutation cycle in place of mutation is studied. Performance of the proposed iterated approach is analyzed and the solution qualities obtained are reported.     

Optimisation of distributed manufacturing flexible job shop scheduling by using hybrid genetic algorithms

In contrast to traditional job-shop scheduling problems, various complex constraints must be considered in distributed manufacturing environments; therefore, developing a novel scheduling solution is necessary. This paper proposes a hybrid genetic algorithm (HGA) for solving the distributed and flexible job-shop scheduling problem (DFJSP). Compared with previous studies on HGAs, the HGA approach proposed in this study uses the Taguchi method to optimize the parameters of a genetic algorithm (GA). Furthermore, a novel encoding mechanism is proposed to solve invalid job assignments, where a GA is employed to solve complex flexible job-shop scheduling problems (FJSPs). In addition, various crossover and mutation operators are adopted for increasing the probability of finding the optimal solution and diversity of chromosomes and for refining a makespan solution. To evaluate the performance of the proposed approach, three classic DFJSP benchmarks and three virtual DFJSPs were adapted from classical FJSP benchmarks. The experimental results indicate that the proposed approach is considerably robust, outperforming previous algorithms after 50 runs.     

时滞系统控制参数优化的遗传模拟退火算法

该文基于遗传模拟退火算法，提出一种时滞系统的控制参数优化方法，同时对Matlab遗传算法工具箱GAOT进行改进，使之适用于PID参数的优化。该文所采用的算法保留了遗传算法和模拟退火算法分别在全局和局部搜索能力强的优点，能克服常规遗传算法中解的早熟现象、局部寻优能力差，难以保证对参数优化的计算效率和可靠性要求等缺陷。研究表明，改进后的遗传模拟退火算法是一种行之有效的方法，具有实用价值。     

A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming

Evolutionary programming can solve black-box function optimisation problems by evolving a population of numerical vectors. The variation component in the evolutionary process is supplied by a mutation operator, which is typically a Gaussian, Cauchy, or Lévy probability distribution. In this paper, we use genetic programming to automatically generate mutation operators for an evolutionary programming system, testing the proposed approach over a set of function classes, which represent a source of functions. The empirical results over a set of benchmark function classes illustrate that genetic programming can evolve mutation operators which generalise well from the training set to the test set on each function class. The proposed method is able to outperform existing human designed mutation operators with statistical significance in most cases, with competitive results observed for the rest.