基于改进边际优化的离散变量优化设计算法

针对传统离散变量优化方法存在的目标函数测算次数多、收敛性不佳等问题,借鉴边际优化理论和模式搜索算法,设计了一种基于改进边际优化的离散变量优化设计算法.借鉴边际效用优化原理,通过引入周围单位步长空间的概念,在初始点选择、边际增量设计、禁忌搜索策略等方面进行了改进,并设计了变异操作以跳出局部最优.实例分析表明,所提算法能够...     

A new method for uniform local heating deep in body using ultrasound phased-array system

A new method for targeted heating of deep tissue was developed by using an ultrasound phased-array system which can generate various multiple foci patterns by electronically changing its amplitude or phase pattern. This method involves using a technique of combining switching and rotating of multiple foci patterns to create a uniform temperature over tissue volumes in various size. Using this method, the target tissue deep in the body can be heated to a specified temperature, which gives conditions for thermo-sensitive liposomes release. A simulation study for a 108-element, spherically sectioned array was performed to determine an optimal heating scheme from a set of multiple focus fields which were produced by inputting different combinations of phases and amplitudes. Comparisons of a static multiple foci field, the switched fields and the switched-rotated fields indicated that the technique of combining switching and rotating of multiple foci patterns has advantages of both lowering the peak temperature and evening the temperature distribution. The simulation results also show that the therapeutic heating zones in various size employing the combined method. These results offer significant data for designing thermotherapy equipment for tumor-specific drug release with thermo-sensitive liposomes.     

基于改进势场法的智能车超车动态路径规划

为了改进人工势场法(APF)在智能车高速超车状态下规划出的轨迹规范性差、安全性得不到保障、以及路径不平滑的问题。通过设计用于指引智能车换道的换道引力势场,以保证换道轨迹的规范性;构建异向分布结构的障碍车斥力势场提高换道的安全性;并建立带有转向缺口的道路中心线斥力势场和设计平滑代价函数优化规划出的轨迹点,使规划出的路径更加平滑。基于线性二次型调节器(LQR)设计路径跟踪控制器跟踪规划出的路径。通过Simulink与Carsim联合仿真,结果表明所改进的APF算法能规划出更加合理的路径,设计的LQR路径跟踪控制器跟踪误差趋近于零,能够较好地控制车辆完成超车换道。     

基于动态种群结构的粒子群算法及仿真研究

针对标准粒子群算法易陷入局部最优而早熟的问题,提出了一种基于动态种群结构的粒子群算法。该算法在种群结构中引入小世界网络模型,由于网络模型的演化,使算法具有动态的种群结构,从而保持了种群的多样性。同时为了使粒子尽可能地分布在不同的搜索空间,在网络模型演化过程中考虑了结点的个体价值。为了加快算法的收敛速度,在进化后期采用全局模型粒子群算法。通过对三个经典测试函数优化问题的数值仿真并与其它方法进行比较,结果表明了算法的有效性和实用性。     

一种新的水平集停止项函数选取方法研究

现今的停止项采用的全局阈值法很难使经典水平集算法得到较准确的分割效果。为此,构建了一个新的停止项函数,该函数在梯度较小时为一个常数,超过某个阈值后为单调递减函数,直至到某个梯度时递减为零。这保证了采用该函数的水平集方法能够实现在同质区加速演化,在边缘区停止演化的目的。实验结果表明,采用新的停止项函数能够使水平集方法获得更好的分割结果。     

单车型确定性动态车辆调配问题

给出了单车型确定性动态车辆调配问题的定义,引入网络图描述该问题.建立了问题的线性规划模型,鉴于线性模型的缺点,利用函数逼近技术构造一个特殊的线性函数来近似目标函数中未来时段部分,从而建立起问题的时空分解模型,把问题从时间和空间上分解为多个单时段单节点问题,并根据单时段单节点问题特点设计简单的排序求解方法.最后,给出了模型的完整求解过程,从而使问题得到有效地解决.     

干线运输优化的两阶段局部搜索启发式算法

干线运输的车货匹配及车辆运输路径优化对于降低大规模、多分布网点的物流企业的运输成本有重要意义.建立了干线运输的发车费用模型及总费用模型,提出应用一种基于直达网络搜索和车辆行驶路径搜索的两阶段局部搜索启发式算法进行求解.通过对企业实际数据的计算,表明该算法优化了干线运输的运输路径,减少了运输车辆数,降低了运输总费用,具有较好的应用前景.     

基于上下层MAS的SAR图像海洋分布目标检测

针对SAR图像中的海洋分布目标,从人的认知过程出发提出了一种基于上下层多Agent系统(top bottom multi Agent system, TBMAS)检测算法。该算法针对上下层Agent定义了不同的行为。下层Agent获得初步的视觉知觉,上层Agent加入联想分析等处理获得最终检测结果。算法从分布目标的局部统计特性考虑,不需要进行统计模型和参数的估计,利用局域直方图和Kullback Leibler散度进行差异度分析。实验表明,对于多种分布目标均有较好的检测效果。     

Testing serial correlation in semiparametric varying coefficient partially linear errors-in-variables model

The authors propose a V _{N, p} test statistic for testing finite-order serial correlation in a semiparametric varying coefficient partially linear errors-in-variables model. The test statistic is shown to have asymptotic normal distribution under the null hypothesis of no serial correlation. Some Monte Carlo experiments are conducted to examine the finite sample performance of the proposed V _{N, p} test statistic. Simulation results confirm that the proposed test performs satisfactorily in estimated size and power. This research is supported by the National Natural Science Foundation of China under Grant Nos. 10871217 and 40574003; the Science and Technology Project of Chongqing Education Committee under Grant No. KJ080609; the Doctor's Start-up Research Fund under Grant No. 08-52204; and the Youth Science Research Fund of Chongqing Technology and Business University under Grant No. 0852008.     

基于脸部结构知识的三维人脸建模

将Sibson坐标用于局部特征分析,提出了一个新的局部形变模型.该模型具有较强的局部性,可以精细地刻画人脸个性特征,且能够根据人脸形状的先验信息进行整体控制.然后,在局部形变的基础上采用双重形变,实现了两种互补的形变算法在特征级的融合,进一步提高了建模精度.实验结果表明,本算法具有较小的重建误差和较快的处理速度,能够生成光滑、逼真的三维人脸模型.