基于人工免疫的机车二系调簧方法

为进一步提高电力机车二系支承载荷调整优化算法的性能,解决已有算法在加垫数量控制上的欠缺,并避免冗余计算,提出了一种基于人工免疫算法的两级结构机车二系载荷调整方法:免疫优势克隆选择多目标算法。针对调簧问题的偏好结构特点,设计具有调簧寻优阶段和可控加垫量寻优阶段的两级寻优算法,分别定义了两个阶段的寻优目标和可行域,进一步将调簧问题的先验知识作为免疫优势引入算法设计中,并结合调簧问题的具体特点以及人工免疫知识设计了特定的抗体编码形式、变异策略和免疫优势获得算法。对不同车型的多次应用试验统计结果表明:该算法不仅具有良好的优化载荷分布特性,还能稳定地控制调簧产生的加垫数量,并且能够避免冗余计算,极大地提高了调簧效率和实际应用价值。     

High-Integrated-Photosensitivity Negative-Electron-Affinity GaAs Photocathodes with Multilayer Be-Doping Structures

The effect of changing Be doping concentration in GaAs layer on the integrated photosensitivity for negative-electron-affinity GaAs photocathodes is investigated.Two GaAs samples with the monolayer structure and the multilayer structure are grown by molecular beam epitaxy.The former has a constant Be concentration of 1e19cm-3,while the latter includes four layers with Be doping concentrations of 1e19,7e18,4e18,and 1e18cm-3 from the bottom to the surface.Negative-electron-affinity GaAs photocathodes are fabricated by exciting the sample surfaces with alternating input of Cs and O in the high vacuum system.The spectral response results measured by the on-line spectral response measurement system show that the integrated photosensitivity of the photocathode with the multilayer structure enhanced by at least 50% as compared to that of the monolayer structure.This attributes to the improvement in the crystal quality and the increase in the surface escape probability.Different stress situations are observed on GaAs samples with monolayer structure and multilayer structure,respectively.     

基于免疫的并行单亲遗传算法研究

在分析了单亲遗传算法的优越性与存在不足的基础上，借鉴生物免疫概念与理论并结合并行计算的思想，提出了一种新的遗传算法——基于免疫的并行单亲遗传算法。理论分析和仿真结果表明，该算法不仅能够有效地保持群体多样性，而且减轻了遗传算法后期的波动现象，同时收敛速度有明显的提高。     

Recent Advances of Cell Membrane‐Coated Nanomaterials for Biomedical Applications

Surface modification of nanomaterials is essential for their biomedical applications owing to their passive immune clearance and damage to reticuloendothelial systems. Recently, a cell membrane‐coating technology has been proposed as an ideal approach to modify nanomaterials owing to its facile functionalized process and good biocompatibility for improving performances of synthetic nanomaterials. Here, recent advances of cell membrane‐coated nanomaterials are reviewed based on the main biological functions of the cell membrane in living cells. An overview of the cell membrane is introduced to understand its functions and potential applications. Then, the applications of cell membrane‐coated nanomaterials based on the functions of the cell membrane are summarized, including physical barrier with selective permeability and cellular communication via information transmission and reception processes. Finally, perspectives of biomedical applications and challenges about cell membrane‐coated nanomaterials are discussed.     

智能优化技术的免疫算法研究

介绍了最优化理论中智能技术的发展及免疫计算模型，对基于人工免疫的智能优化方法进行了讨论。最后通过实例说明了免疫计算求解优化问题的实现。     

一种基于梯度免疫算法的参差MTI滤波器设计

为了解决动目标检测过程中的盲速问题,通常采用参差周期法设计参差滤波器组来达到同时抑制杂波和推远盲速的效果。提出了一种基于梯度免疫算法的参差MTI滤波器优化设计。该算法利用免疫算法搜索最优参差码,有效地避免了传统遗传算法中的早熟问题。针对传统的遗传算法运算量较大的问题,引入了梯度算子缩小初始种群规模,使运算量大大减少。仿真结果验证了算法的可行性和有效性。     

电子材料分析中的能谱干扰峰

主要阐述了在电子元器件分析中用能谱仪作定性分析时常见的一些干扰峰和容易混淆及误判的一些谱峰。主要有和峰、逃逸峰以及一些在元器件材料分析中常遇到的元素特征峰之间的交错重叠的识别和判定方法，并把这些容易误判的谱线整理列成3种表格，以供参考，这些数据基本覆盖了在X射线能谱仪中可能出现的所有相关的谱峰。     

基于免疫算法的空战目标分配

目标分配是现代空战系统对抗中要首先面对的问题,它是一个组合优化问题.首先用优势函数来表明敌我态势,建立机群对抗优势函数矩阵.然后将一种新型的启发式算法--免疫算法用于解决目标分配问题.免疫算法具有很好的全局搜索能力和收敛性,能够很好地解决这类问题,实例也证明了该方法是可行和有效的.     

基于免疫优化算法的军用中心仓库选址

为了最大程度地降低军用中心仓库的综合成本,优化其选址位置,针对从众多的军用仓库选取最佳中心仓库的问题进行研究.在层次分析法的基础上提出综合成本的概念,并以最小综合成本为目标函数,利用免疫优化算法建立模型,利用Matlab软件确定实例中军用中心仓库的选址位置,并通过适应度曲线确定其适应度收敛情况.验证结果表明:该算法计算简单、效果好,能提供高效有力的物流保证,对军队的军用中心仓库选址具有较强的参考意义,具有极高的推广价值.     

Rate adaptive resource allocation in orthogonal frequency division multiple access system using multi‐objective immune algorithm

Rate adaptive downlink resource allocation in orthogonal frequency division multiple access system is a constraint optimization problem, which is to maximize the minimum data rate of the user subject to constraint that total power cannot exceed a given value. How to handle the constraint is a key issue for constrained optimization problem. Different with the available schemes on constraint handling, the proposed algorithm converts the constraint into an objective. Then, the resource allocation is combined into a multi‐objective optimization problem. An improved multi‐objective optimization algorithm based on artificial immune system is proposed to solve it. The simulation results show that, compared with previous schemes, the proposed algorithm performs remarkable improvement in sum capacity. Copyright © 2013 John Wiley & Sons, Ltd.