Two-Phase Genetic Algorithm Applied in the Optimization of Multi-Modal Function

This paper presents a two-phase genetic algorithm (TPGA) based on the multi- parent genetic algorithm (MPGA). Through analysis we find MPGA will lead the population' s evol vement to diversity or convergence according to the population size and the crossover size, so we make it run in different forms during the global and local optimization phases and then forms TPGA. The experiment results show that TPGA is very efficient for the optimization of low-dimension multi-modal functions, usually we can obtain all the global optimal solutions.

A Two-Level Subspace Evolutionary Algorithm for Solving Multi-Modal Function Optimization Problems

In this paper, a new algorithm for solving multi-modal 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.

Elitism-based immune genetic algorithm and its application to optimization of complex multi-modal functions

A novel immune genetic algorithm with the elitist selection and elitist crossover was proposed, which is called the immune genetic algorithm with the elitism (IGAE). In IGAE, the new methods for computing antibody similarity, expected reproduction probability, and clonal selection probability were given. IGAE has three features. The first is that the similarities of two antibodies in structure and quality are all defined in the form of percentage, which helps to describe the similarity of two antibodies more accurately and to reduce the computational burden effectively. The second is that with the elitist selection and elitist crossover strategy IGAE is able to find the globally optimal solution of a given problem. The third is that the formula of expected reproduction probability of antibody can be adjusted through a parameter β, which helps to balance the population diversity and the convergence speed of IGAE so that IGAE can find the globally optimal solution of a given problem more rapidly. Two different complex multi-modal functions were selected to test the validity of IGAE. The experimental results show that IGAE can find the globally maximum/minimum values of the two functions rapidly. The experimental results also confirm that IGAE is of better performance in convergence speed, solution variation behavior, and computational efficiency compared with the canonical genetic algorithm with the elitism and the immune genetic algorithm with the information entropy and elitism.

基于NSGA-Ⅱ传感位置优化的曲面重构及误差补偿方法

通过优化光纤布拉格光栅形状传感技术中传感点位置和补偿重构结果来提高薄层合金板三维形状重构精度。通过ANSYS workbench建立合金板仿真模型,提取应变和位移模态振型,根据模态置信准则、转换矩阵稳定性和模态振型相似性分别设计了三个目标函数,采用快速和精英机制的多目标遗传算法优化传感器位置。将镍钛合金板弯曲成不同曲率半径的弧形,利用光纤布拉格光栅中心波长漂移量和线性插值算法计算得出不同形状下的结构应变,重构合金板形状,均方根误差和最大误差相较于单目标优化算法分别减小30%和15%。利用粒子群优化径向基函数神经网络算法拟合误差与位移的关系实现误差补偿,均方根误差和最大误差比无补偿时分别减小了90%和70%,最大相对百分比误差仅为5%,提高了三维形状重构算法精度。

基于VMD-PSO-多核极限学习机的短期负荷预测

为提高短期负荷预测精度,解决核极限学习机单一核函数难以适应负荷多数据特征的问题,提出了一种基于变分模态分解与粒子群优化的多核极限学习机模型。该模型采用变分模态分解技术将原始负荷序列分解为具有不同特征频率的子序列,并对每个子序列建立预测模型。负荷预测模型采用粒子群优化的多核极限学习机,其使用混合核函数代替单一的核函数,使其在不同的参数下不仅有良好的局部搜索能力,同时也加强了全局搜索能力。实验表明,该模型拥有更好的回归精度和泛化能力,能够得到更精确的预测结果。

Functional imaging and endoscopy

The emergence of endoscopy for the diagnosis of gastrointestinal diseases and the treatment of gastrointestinal diseases has brought great changes.The mere observation of anatomy with the imaging mode using modern endoscopy has played a significant role in this regard.However,increasing numbers of endoscopies have exposed additional deficiencies and defects such as anatomically similar diseases.Endoscopy can be used to examine lesions that are difficult to identify and diagnose.Early disease detection requires that substantive changes in biological function should be observed,but in the absence of marked morphological changes,endoscopic detection and diagnosis are difficult.Disease detection requires not only anatomic but also functional imaging to achieve a comprehensive interpretation and understanding.Therefore,we must ask if endoscopic examination can be integrated with both anatomic imaging and functional imaging.In recent years,as molecular biology and medical imaging technology have further developed,more functional imaging methods have emerged.This paper is a review of the literature related to endoscopic optical imaging methods in the hopes of initiating integration of functional imaging and anatomical imaging to yield a new and more effective type of endoscopy.

ARTIFICIAL IMMUNE ALGORITHM OF MULTICELLULAR GROUP AND ITS CONVERGENCE

Objective To find out more extrema simultaneously including global optimum and multiple local optima existed in multi-modal functions. Methods Germinal center is the generator and selector of high-affinity B cells, a multicellular group's artificial immune algorithm was proposed based on the germinal center reaction mechanism of natural immune systems. Main steps of the algorithm were given, including hyper-mutation, selection, memory, similarity suppression and recruitment of B cells and the convergence of it was proved. Results The algorithm has been tested to optimize various multi-modal functions, and the simulation results show that the artificial immune algorithm proposed here can find multiple extremum of these functions with lower computational cost. Conclusion The algorithm is valid and can converge on the satisfactory solution set D with probability 1 and approach to global solution and many local optimal solutions existed.

基于超图正则化的多模态信息融合算法

多模态数据融合方法通过学习多个数据集间的关联信息和互补信息,提高了数据分类或预测的性能。但现有的数据融合方法大都基于单独数据集自身的特征模式进行学习,不同异构数据之间的结构信息往往被忽略。因此,文中提出了一种基于超图正则化的多模态信息融合算法(sHMF),通过超图和流行正则项的方法结合表示模态内样本间的高阶关系和模态间的关系,即得到同构和异构的高阶网络。其中,采用超图稀疏表达学习超图,减少冗余边。为了验证所提算法的性能,在模拟数据和影响遗传学真实数据下进行实验,结果表明,sHMF算法在模拟数据和真实数据上均优于多任务学习、多邻域分类等流行算法对精神分裂症的分类精度。同时,sHMF在真实数据上得出的实验结果进一步揭示了一些与精神分裂症显著相关的生物标记物以及风险基因、甲基化因子和异常脑区之间潜在的联系。

基于模态注意力图卷积特征融合的EEG和fNIRS情感识别

为了提升情感识别的准确率,从情绪视频引起的脑电(EEG)和功能近红外(fNIRS)数据中提取每个通道的信号之间的联系,并提出基于模态注意力多路图卷积神经网络(MA-MP-GF)的特征融合情感识别方法.将EEG和fNIRS数据构建为图结构数据,通过多路图卷积分别对每种模态的信号进行特征提取;利用模态注意力图卷积层融合不同模态通道间的连接信息.模态注意力机制可以赋予不同模态节点不同权重,使得图卷积层能够更加充分提取不同模态节点间连接关系.对采集的30个被试的4类情感数据进行实验测试,与仅EEG和仅fNIRS单模态识别结果相比,所提出的图卷积融合方法能够获得更高的识别准确率,分别提升了8.06%、22.90%;与当前常用的EEG-fNIRS融合方法相比,所提出的图卷积融合方法的平均识别准确率提升了2.76%~7.36%;图卷积融合方法在加入模态注意力后识别率最高提升了1.68%.

国际多式联运电子单证法律规制初析

多式联运一体化是现代运输组织形式的重大变革。电子单证与纸质单证的理论分野与实践差异深刻反映出信息化与工业化时代的更迭,多式联运单证统一化与数字化是商事实践发展的必然趋势。现有法律规则实践价值与技术属性的功能性缺失成为国际多式联运发展的现实障碍,多式联运一单制的渐进改革与电子单证的法律功能受到明显制约。当前,应在对国际多式联运电子单证的维度解构与逻辑聚合以及规则构建认识论进行重新诠释、明确国际多式联运电子单证规则构建的价值导向的基础上,通过法律规范化、技术数字化与实践国际化来构建国际多式联运电子单证法律制度。

巷修机器人多功能机械臂结构的设计与仿真

为实现巷修机器人在煤矿井下狭小空间内的巷道修复作业,设计了一种大折展比多功能机械臂,通过3节机械臂折展来适应巷道条件,并完成受限空间内大范围的巷道修复作业。利用三维建模软件辅助设计并建立机械臂模型。为了验证该机械臂结构合理性及工作过程中的安全稳定性,利用三维仿真软件对机械臂模型进行了静力学分析,并对其进行模态分析。分析结果表明:机械臂的强度满足要求;机械臂前6阶模态振幅偏大,需要对其加强设计,以提高其稳定性。

哈尔滨市中央大街历史文化街区官方历史建筑语言景观调查研究

该文从语言服务、旅游语言景观的视角出发,实地调查了哈尔滨市中央大街历史文化街区官方的历史建筑语言景观现状。调查结果显示,官方设计建筑标牌时将多模态融入历史建筑语言景观的构建中,哈尔滨市官方历史建筑语言景观具有多元化及动态化的发展趋势,但总体上仍表现为静态化特点。中央大街官方的历史建筑语言景观具有信息功能和象征功能,信息功能直观传递老建筑的基本概况信息,象征功能主要指文化象征功能,哈尔滨市中央大街历史文化街区的历史建筑语言景观映射出独具地域特色的城市异域文化及本土红色文化。最后,针对历史建筑语言景观建构中的静态化特点,提出文旅融合时代背景下的历史建筑语言景观的发展建议及构建策略。

Can multi-modal neuroimaging evidence from hippocampus provide biomarkers for the progression of amnestic mild cognitive impairment?

Impaired structure and function of the hippocampus is a valuable predictor of progression from amnestic mild cognitive impairment（a MCI） to Alzheimer＇s disease（AD）. As a part of the medial temporal lobe memory system,the hippocampus is one of the brain regions affected earliest by AD neuropathology,and shows progressive degeneration as a MCI progresses to AD. Currently,no validated biomarkers can precisely predict the conversion from a MCI to AD. Therefore,there is a great need of sensitive tools for the early detection of AD progression. In this review,we summarize the specifi c structural and functional changes in the hippocampus from recent a MCI studies using neurophysiological and neuroimaging data. We suggest that a combination of advanced multi-modal neuroimaging measures in discovering biomarkers will provide more precise and sensitive measures of hippocampal changes than using only one of them. These will potentially affect early diagnosis and disease-modifying treatments. We propose a new sequential and progressive framework in which the impairment spreads from the integrity of fibers to volume and then to function in hippocampal subregions. Meanwhile,this is likely to be accompanied by progressive impairment of behavioral and neuropsychological performance in the progression of a MCI to AD.

粒子群优化算法的收敛性分析及其混沌改进算法

分析了粒子群优化算法的收敛性,指出它在满足收敛性的前提下种群多样性趋于减小,粒子将会因速度降低而失去继续搜索可行解的能力;提出混沌粒子群优化算法,该算法在满足收敛性的条件下利用混沌特性提高种群的多样性和粒子搜索的遍历性,将混沌状态引入到优化变量使粒子获得持续搜索的能力.实验结果表明混沌粒子群优化算法是有效的,与粒子群优化算法、遗传算法、模拟退火相比,特别是针对高维、多模态函数优化问题取得了明显改善.

求解多峰函数优化的小生境克隆选择算法

分析了传统的用于多峰函数优化问题的小生境遗传算法的特点和不足,基于免疫系统中的克隆选择原理,运用记忆算子、抑制算子和重组算子等技术对克隆选择算法进行了改造,并引入一种新的小生境技术,提出了一种解决多峰函数优化问题的小生境克隆选择算法。最后,实现了该算法对单无人作战飞机(unmanned combat air vehicle,UCAV)多航路规划这类多峰函数优化问题的优化仿真,结果表明该算法简单有效。

基于多种群的改进粒子群算法多模态优化

针对多模态函数寻优过程中开发与探索能力难以平衡的问题,提出一种基于多种群的改进粒子群算法(EMSPSO)。该算法在基于种群的粒子群算法(SPSO)的基础上改进了种群生成策略,通过在个体最优值中选择种子,将粒子群分为若干独立进化的种群,增强了算法收敛的稳定性;为了提高粒子的利用率、算法的全局搜索能力和搜索效率,引入冗余粒子重新初始化策略;同时为了防止算法在寻优的过程中遗漏适应度较优的极值点,对速度更新公式进行改进,使算法的开发与探索能力得到了有效的均衡。最后选用6个典型的测试函数进行对比实验,实验结果表明,EMSPSO具有较高的多模态寻优成功率与较优的全局极值搜索性能。

演化计算中的种群隔离与自聚集

通过对局部最优吸引域的分析,说明了种群隔离的作用.以局部演化的方式,实现了种群间分离与种群内自聚集,使多峰函数优化问题转化为单峰函数优化问题.结合在单峰函数优化中收敛速度较快的、基于空间划分的（μ+1）演化方法作为基本的演化操作,提出了种群按照吸引域自动分离与聚集的演化算法.最后给出了运算实验的结果.

带拥塞控制的多种群二元蚁群算法

二元蚁群算法在函数优化中有着良好的表现,但仍存在易陷入局部最优和在多峰函数求解中无法同时得到多个解的缺陷.使用拥塞控制策略改善算法的全局寻优能力,同时引入多种群的思想,提出了带拥塞控制多种群二元蚁群算法.通过对几个不同函数(包括单峰与多峰)的测试,实验结果表明该改进算法在保证较好的全局搜索能力的基础上,拥有很好的多目标求解能力。

免疫算法

首先对免疫算法作了介绍 ,并与其它非确定性算法进行了比较 ,随后介绍了现有的几种免疫算法及其实现步骤。

基于混沌序列的多峰函数微粒群寻优算法

基于混沌序列的多峰函数微粒群寻优算法的目标就是找到多峰函数的所有局部优化峰值。在分析微粒群优化算法中各个参数对微粒运动影响的基础上,对微粒群算法进行改造,让微粒运动从初始位置沿优化函数曲线向优化峰值方向爬行,直至找到所在区域的局部优化峰值;要想求得尽可能多的局部优化峰值,就要求微粒群中微粒的初始位置分布具有随机性和遍历性,为此采用混沌序列设置微粒初始位置;为使每一个局部最优值点都可能有微粒群中的微粒经过,采用变步长的迭代计算;为防止优化函数曲线的某些局部峰附近没有微粒分布,从而漏掉该局部峰值,对计算进行重复,直至两轮求得的优化函数的局部峰值之差小于给定阈值。仿真结果表明,该算法具有很好的局部寻优特性,计算过程简捷,寻优效果良好,可有效地应用于多峰函数的局部寻优并求取全局最优值。