自适应协同进化算法及其在创新设计中的应用研究

提出一种自适应协同进化算法,对其进行了数学描述。设计了一个支持该算法的创新设计系统,为分布式环境下设计人员的协作和创新思路的开拓提供了支撑平台。算法中自适应学习的引入为在设计中自动而有效地使用先验智能提供了可行性。最后以一个建筑实例的设计为例对所述的方法和系统加以描述。     

具有参数自适应机制的改进离散差分进化算法

在研究和分析离散差分进化算法的基础上,提出了一种具有参数自适应机制的改进离散差分进化算法(PADDE)。该算法首先对连续域进化过程中的参数进行自适应调整,以平衡全局搜索与局部搜索,协调种群多样性和收敛速度间的矛盾,其次根据对应离散域上成功进化的个体的离散编码反馈信息引导算法协同进化。通过对背包问题进行的实验表明,该算法具有良好的收敛效率和稳定性。     

基于双层演化的多人囚徒博弈研究

针对传统的多人重复囚徒博弈（NIPD）难以在大N值时涌现高合作率的问题进行研究,分析了NIPD模型在自由竞争模式和协议竞争模式下的博弈情况,类比2-IPD问题的“针锋相对”策略（TFT）,提出了“类TFT”的策略思想,并结合协同进化的理论,提出Agent及其聚集体Group分层演化的思想,建立了双层演化的仿真模型DL-NIPD。实验结果表明,自由竞争模式只适合小N值的合作,要从根本上保证任何N值下系统都能涌现很高的合作率,必须建立起双层的演化模式,通过显式的协议和团队的竞争,来促进微观主体的合作。     

基于多粒度共进化功能能推理的机械运动方案设计新方法

提出了基于多粒度共进化功能推理的机械运动方案设计方法.首先分析了共进化功能推理模型和多粒度设计模型的特点,并结合二者的优点构造了一种多粒度的共进化功能推理模型;然后将该推理模型应用于机械运动方案设计,提出了一种面向机械运动方案设计的共进化功能推理方法,该方法采用分类功能来描述机构单元及机械系统的运动特征信息,将机构单元和机械系统都采用运动功能变换函数进行抽象表达,通过功能推理来生成机械运动变换单元的串联组合方案;随即给出了相应的功能推理算法流程,通过与已有算法的比较详细分析了该算法的特性,并讨论说明了该算法所具有的效率高、可精确描述运动功能变换特性等优点;最后通过电线进给机构运动方案设计实例验证了该方法的有效性.     

A genetic algorithm for the simultaneous delivery and pickup problems with time window

This paper concerns a Simultaneous Delivery and Pickup Problem with Time Windows (SDPPTW). A mixed binary integer programming model was developed for the problem and was validated. Due to its NP nature, a co-evolution genetic algorithm with variants of the cheapest insertion method was proposed to speed up the solution procedure. Since there were no existing benchmarks, this study generated some test problems which revised from the well-known Solomon’s benchmark for Vehicle Routing Problem with Time Windows (VRPTW). From the comparison with the results of Cplex software and the basic genetic algorithm, the proposed algorithm showed that it can provide better solutions within a comparatively shorter period of time.     

基于食物链的生态进化算法

借鉴生物在生态环境中的能量传递方式一食物链,提出了一种多模式协同的生态进化算法。同时将该算法应用于典型优化问题求解,实验结果表明此算法具有良好的性能。     

Modelling sustainability

The article presents a general classification of the models being developed in the area of sustainability arguing that the existing models represent the historical conceptualisation of sustainability starting from environmental constraints and moving towards economic valuation and social behaviour and policies. Coupled with computer power, sophisticated models with a varying levels of complexity have also been developed (static/dynamic; local/global; specific/general). However as any model is a simplification of the complex reality, the main purpose of any sustainability modelling (and the newly emerging area of sustainometrics) should be to allow dynamic representation, including the co-evolution of the sustainability systems and the role of humans as sustainability guardians.     

基于自适应协同进化粒子群算法的虚拟网节能映射研究

该文针对虚拟网节能映射问题提出自适应的协同进化粒子群算法。首先,为虚拟网节能映射问题设置了聚合度,该聚合度被用于自适应地选择粒子的搜索方式,即随机搜索、种内搜索或种外搜索。其次,根据粒子群的进化结果,自适应地确定是否终止对子群的搜索。最后,在常用的测试环境下进行了仿真实验,对映射的能耗效果对比了结果,实验结果表明了所提算法的高效性。     

Advances in evolutionary feature selection neural networks with co-evolution learning

Training neural networks is a complex task provided that many algorithms are combined to find best solutions to the classification problem. In this work, we point out the evolutionary computing to minimize a neural configuration. For this purpose, a distribution estimation framework is performed to select relevant features, which lead to classification accuracy with a lower complexity in computational time. Primarily, a pruning strategy-based score function is applied to decide the network relevance in the genetic population. Since the complexity of the network (connections, weights, and biases) is most important, the cooling state of the system will strongly relate to the entropy as a minimization function to reach the desired solution. Also, the framework proposes coevolution learning (with discrete and continuous representations) to improve the behavior of the evolutionary neural learning. The results obtained after simulations show that the proposed work is a promising way to extend its usability to other classes of neural networks.     

Gause竞争型协同进化算法在FNN中的应用

自从60年代J.霍兰提出遗传算法以来,模拟进化算法得到了很大的发展和应用。协同进化算法是针对遗传算法的不足提出,还处于研究初步阶段。该文在竞争型协同进化的基础上,借鉴生态学中物种竞争模型,提出了基于Gause竞争方程的竞争型协同进化算法,并将该算法应用于模糊神经系统的辨识问题上。实验证明,该算法比标准遗传算法、典型竞争型协同进化算法和BP学习算法具有更好的全局收敛性和更快的收敛速度,在一定程度上解决了标准遗传算法的不足。