A new evolutionary algorithm for function optimization

A new algorithm based on genetic algorithm(GA) is developed for solving function optimization problems with inequality constraints. This algorithm has been used to a series of standard test problems and exhibited good performance. The computation results show that its generality, precision, robustness, simplicity and performance are all satisfactory. Foundation item: Supported by the National Natural Science Foundation of China (No. 69635030), National 863 High Technology Project of China, the Key Scientific Technology Development Project of Hubei Province. Biography: GUO Tao(1971-), male, Ph D, research interests are in evolutionary computation and network computing.     

多维函数优化的遗传算法研究

本文研究了求解多变量函数优化问题的遗传算法,在此算法中采用了十进制浮点数基因表示方法,并相应地提出了一种叠加零均值Gauss随机扰动的变异方法,研究表明,对于满足组件假说的多维函数优化问题,这种遗传算法具有较高的搜索效率．     

面向多模函数优化的改进网格优化算法

利用网格优化算法（COA）编码简单、收敛速度快、不宜陷入局部最优等特点,针对多模态函数优化问题,对GOA算法进行了改进,扩大了优化搜索范围,保持了父本种群的多样性,增强了全局搜索能力。对典型多模态函数问题的测试结果表明,改进的网格优化算法在解决多模态函数优化问题方面具有很强的全局搜索能力和很高的搜索效率。     

A two-level subspace evolutionary algorithm for solving multi-modal function optimization problems

In this paper, a new algorithm for solving multimodal function optimization problems-two-level subspace evolutionary algorithm is proposed. In the first level, the improved GT algorithm is used to do global recombination search so that the whole population can be separated into several niches according to the position of solutions; then, in the second level, the niche evolutionary strategy is used for local search in the subspaces gotten in the first level till solutions of the problem are found. The new algorithm has been tested on some hard problems and some good results are obtained. Foundation item: Supported by the National Natural Science Foundation of China (70071042, 60073043, 60133010). Biography: Li Yan( 1974-), female, Ph. D candidate, research interest: evolutionary computation.     

求解约束优化问题的一种新的进化算法

分析了现有的约束优化进化算法的一些不足之处，提出了一种处理约束优化问题的新算法。新算法将多目标优化思想与全局搜索和局部搜索机制有机地结合起来；在全局搜索过程中，作为一种小生态遗传算法，排挤操作利用Pareto优劣关系比较个体并接受具有相似性的父代个体和予代个体中的优胜者；在局部搜索过程中，首先对局部群体中的个体赋予Pareto强度，然后根据Pareto强度选择个体。通过一个复杂高维多峰测试函数验证了新算法的有效性。     

Asynchronous parallel evolutionary algorithms for constrained optimizations

Recently Guo Tao proposed a stochastic search algorithm in his PhD thesis for solving function optimization problems. He combined the subspace search method (a general multi-parent recombination strategy) with the population hill-climbing method. The former keeps a global search for overall situation, and the latter keeps the convergence of the algorithm. Guo's algorithm has many advantages, such as the simplicity of its structure, the higher accuracy of its results, the wide range of its applications, and the robustness of its use. In this paper a preliminary theoretical analysis of the algorithm is given and some numerical experiments has been done by using Guo's algorithm for demonstrating the theoretical results. Three asynchronous parallel evolutionary algorithms with different granularities for MIMD machines are designed by parallelizing Guo's Algorithm. National Laboratory for Parallel and Distributed Processing Foundation item: Supported by the Natonal Natural Science Foundation of China (No. 70071042, 50073043), the National 863 Hi-Tech Project of China (No. 863-306-ZT06-06-3) and the National Laboratory for Parallel and Distributed Processing. Biography: Kang Li-shan (1934-), male, Professor, research interests: parallel computing and evolutionary computation.     

基于KFDD与进化算法的可逆电路优化算法

为降低由Kronecker功能决策图(Kronecker functional decision diagram,KFDD)综合所得可逆电路的成本,提出一种基于进化算法的可逆电路优化算法.该算法基于遗传算法模型进行设计,分别采用离散值和整型值编码KFDD输入变量的分解类型和顺序,使用所设计的遗传算子,将量子成本作为主要目标、量子位数作为次要目标进行可逆电路的优化.为解决过早收敛问题,该算法在搜索过程的前期阶段利用多个子群搜索解空间中的不同区域,在搜索过程的后期阶段将多个子群合并为整体种群,利用整体种群进行集中搜索.使用基准函数对算法进行验证的结果表明,所提出算法具有较强的全局寻优能力,有较好的结果稳定性,能够降低可逆电路的量子成本.     

多峰函数寻优的微粒群算法

某些实际问题的优化目标是求所有的局部最优解，即求解多峰寻优问题，为了求解多峰优化问题，提出了改造的微粒群优化算法．尽量减少微粒群算法中的全局因素，从而增大其局部因素，同时采用变步长方法增加微粒的多样性．并给出了该算法的原理和步骤．仿真实验表明该算法概念清楚，计算简单，具有很好的局部寻优特性，可应用求解于多峰寻优问题．另外还给出了几个运算实例和与其它优化算法的比较．图表，表1，参9．     

Novel evolutionary algorithm for global numerical optimization with continuous variables

Evolutionary algorithms (EA) are a class of general optimization algorithms which are applicable to functions that are multimodal, non-differentiable, or even discontinuous. In this paper, a novel evolutionary algorithm is proposed to solve global numerical optimization with continuous variables. In order to make the algorithm more robust, the initial population is generated by combining determinate factors with random ones. And a decent scale function is designed to tailor the crossover operator so that it can not only find the decent direction quickly but also keep scanning evenly in the whole feasible space. In addition, to improve the performance of the algorithm, a mutation operator which increases the convergence-rate and ensures the convergence of the proposed algorithm is designed. Then, the global convergence of the presented algorithm is proved at length. Finally, the presented algorithm is executed to solve 24 benchmark problems. And the results show that the convergence-rate is noticeably increased by our algorithm.     

Novel evolutionary algorithm for global numerical optimization with continuous variables

Evolutionary algorithms (EA) are a class of general optimization algorithms which are applicable to functions that are multimodal, non-differentiable, or even discontinuous. In this paper, a novel evolutionary algorithm is proposed to solve global numerical optimization with continuous variables. In order to make the algorithm more robust, the initial population is generated by combining determinate factors with random ones. And a decent scale function is designed to tailor the crossover operator so that it can not only find the decent direction quickly but also keep scanning evenly in the whole feasible space. In addition, to improve the performance of the algorithm, a mutation operator which increases the convergence-rate and ensures the convergence of the proposed algorithm is designed. Then, the global convergence of the presented algorithm is proved at length. Finally, the presented algorithm is executed to solve 24 benchmark problems. And the results show that the convergence-rate is noticeably increased by our algorithm.     

求解QoS路由优化的一种新进化算法

对网络中支持多个QoS参数路由的数学模型进行了形式化分析,提出了一种多目标进化算法（QMOEA）。该算法能有效地将多个优化目标统一起来,并在此基础之上引入“自适应退避”机制与贪心策略,保证了群体的多样性和快速收敛。仿真结果与理论分析验证了该算法的有效性与正确性。     

关于热传导方程有限差分区域分解并行算法精度的注记

以一维热传导模型方程为例来说明用有限差分区域分解算法求解热传导方程中的三个现象.     

Parameter optimization of electric bus transmission system based on dynamical evolutionary algorithm

The transmission ratio is the key parameters influence power performance and economic performance of electric vehicle (EV). As a class of heuristic algorithms, Dynamical Evolutionary Algorithm (DEA) is suitable to solve multi-objective optimization problems. This paper presents a new method to optimize the transmission ratio using DEA. The fuzzy constraints and objective function of transmission ratio are established for parameter optimization problem of electric bus transmission. DEA is used to solve the optimization problem. The transmission system is also designed based on the optimization result. Optimization and test results show that the dynamical evolutionary algorithm is an effective method to solve transmission parameter optimization problems.     

A new evolutionary algorithm based on the decimal coding

Traditional Evolutionary Algorithm (EAs) is based on the binary code, real number code, structure code and so on. But these coding strategies have their own advantages and disadvantages for the optimization of functions. In this paper a new Decimal Coding Strategy (DCS), which is convenient for space division and alterable precision, was proposed, and the theory analysis of its implicit parallelism and convergence was also discussed. We also redesign several genetic operators for the decimal code. In order to utilize the historial information of the existing individuals in the process of evolution and avoid repeated exploring, the strategies of space shrinking and precision alterable, are adopted. Finally, the evolutionary algorithm based on decimal coding (DCEAs) was applied to the optimization of functions, the optimization of parameter, mixed-integer nonlinear programming. Comparison with traditional GAs was made and the experimental results show that the performances of DCEAS are better than the tradition GAs. Foundation item: Supported by the National Natural Science Foundation of China (No. 69703011) Biography: Dong Wen-yong (1973-), male Ph. D. candidate, research direction: parallel algorithms, evolutionary computation, computer simulation.     

基于混合搜索算法的图像稀疏分解

为了提高图像稀疏分解的效果,降低其计算时间,提出一种基于量子进化算法(quantum-inspired evolutionaryalgorithm,QIEA)和改进差分进化算法(improved differential evolution,IDE)的混合搜索算法,并应用到图像稀疏分解中.该方法将1DE引入到QIEA中...     

一种求解多目标优化问题的混合演化算法

提出了一种混合演化算法求解多目标优化问题.演化算法是解决多目标优化问题的有效方法,在全局优化问题中具有很好的鲁棒性,但其局部搜索性能有待改善.Hooke and Jeeves方法是一经典的局部搜索算法,将其与演化算法结合求解多目标优化问题,提高了解的收敛质量,因而从整体上提高了算法的性能,并且测试结果也说明了该算法的可行性.     

A new evolutionary algorithm for solving multi-objective optimization problems

Multi-objective optimization is a new focus of evolutionary computation research. This paper puts forward a new algorithm, which can not only converge quickly, but also keep diversity among population efficiently, in order to find the Pareto-optimal set. This new algorithm replaces the worst individual with a newly-created one by “multi-parent crossover”. so that the population could converge near the true Pareto-optimal solutions in the end. At the same time, this new algorithm adopts niching and fitness-sharing techniques to keep the population in a good distribution. Numerical experiments show that the algorithm is rather effective in solving some Benchmarks. No matter whether the Pareto front of problems is convex or non-convex, continuous or discontinuous, and the problems are with constraints or not, the program turns out to do well. Foundation item: Supported by the National Natural Science Foundation of China(60133010, 60073043, 70071042) Biography: Chen Wen-ping ( 1977-), female, Master candidate, research direction: evolutionary computation.     

一种求解约束函数优化问题的遗传算法

遗传算子和种群更新策略在遗传算法全局寻优过程中发挥着重要作用，通过多父体杂交算子使产生的后代更具多样性和采用最小代数代沟种群替换模型有效地均衡算法对问题解空间的探索和开发能力提高算法的性能，给出了一种求解约束函数优化问题的遗传算法。对两个典型约束函数优化问题进行了数值实验，实验结果表明了该算法的有效性和稳健性。     

波动方程的重叠型区域分解并行有限差分算法

提出了一类新的计算波动方程数值解的并行差分算法. 算法基于区域分解和子区域校正,在每个子域上进行残量修正,各子域之间可以并行计算. 证明了算法的收敛性,并且理论分析表明,在每一时间步,只需校正一或两次,即可达到最优的收敛阶. 数值试验表明了算法的有效性和优越性.     

一种函数优化问题的混沌遗传算法

混沌作为一种非线性动态系统中的现象，受到了人们的重视，给出一种结合混沌优化方法和遗传算法的混沌算法，实验表明，它通过对问题的解空间交替进行全局和局部搜索，能更有效地求解函数优化问题。