数字波束形成算法仿真实现

从讨论波束形成方法在声系统中的主要作用和主要形成过程入手，介绍几种具体实现波束形成的算法。主要讨论一种能够降低硬件要求的波束形成算法——正交波束形成算法。此种方法不但适用于声系统，对于其他检测系统也有一定的实际意义。给出几种主要波束形成算法的Matlab仿真实现程序，将按照此类方法仿真所得到的波束形成指向性图与理论指向性图比较，两者非常类似。     

模拟退火算法在化学自由能模型中的应用

对化学自由能模型进行系统性的研究，着重分析化学反应动态平衡条件下粒子组分的求解方案，提出应用模拟退火算法寻找自由能密度函数极小值点的求解方案.该方案同时解决了两个难题：1）在一级相变区化学势平衡方程组可能遇到多个解而无法甄别其物理意义.通过模拟退火算法定位到自由能密度函数曲面的最低点，因而可从多个解中甄别出稳定的热力学平衡态.2）模拟退火算法用随机的"热涨落"消除初值敏感性，因而可采用同一套初值计算不同的温度密度点，为实现宽区域上大量温度密度点的连续快速计算奠定基础.作为该平衡态求解方案的应用，计算氦流体在"等离子相变"区的物态方程，揭示了丰富的"等离子相变"现象，并与第一性原理计算揭示的氢流体"液液相变"现象进行类比.     

结合精确阵列响应控制的时域宽带波束图综合

提出了采用时域结构的低计算复杂度宽带波束图综合方法。首先,引入自适应零陷波束设计思想,通过在特定频率与角度设置虚拟干扰源,实现对旁瓣区域阵列响应的精确控制,多次迭代控制后满足波束设计要求。其次,所提方法能够在现有波束设计基础上实现灵活调整,根据旁瓣级等设计参数的变化快速更新波束图。理论研究表明,相较现有采用内点法的波束图综合方法,所提方法降低了单次迭代计算复杂度,有效提升了波束设计效率。仿真实验与半消声室实验结果表明,该方法所设计波束形成器相较传统固定波束形成器具有更低的旁瓣级,并且对方向性干扰有更好的抑制效果。     

模拟退火算法在无端偏振控制器中的应用

构造了由四个相移范围在0~2π的电光晶体波片组成的偏振控制器,利用模拟退火法作为其反馈控制算法.仿真计算表明,该偏振控制器是无端的,在较优的控制参量下,其输出的偏振态是稳定的,光强波动小于1%.对于四波片偏振控制器,模拟退火算法的控制参量研究表明,为得到较佳的控制效果,可把内外循环次数各限制为3次,以减少处理器时间;其初温T0应取在0.1附近;温度更改常数p宜取在0.2左右;状态产生常数s对输出的稳定性影响最大,仿真说明设在 0.005~0.01间为好.     

粒子群-模拟退火混合算法在群控电梯中的应用

针对高层建筑电梯多、分布散、维修不及时等问题,提出了一个优化调度、报警维修及时的群控电梯系统;采用多目标优化方法建立数学模型,利用粒子群算法的概念简单、收敛速度快、易于实现的优点,同时引入模拟退火思想来克服粒子群算法易陷入局部最优的缺陷,对目标函数进行优化,PLC编程计算出参数,上位机调度计算,根据算法结果,优化电梯,合理调度;经过MATLAB仿真分析表明,该系统节省了平均候梯时间、平均乘梯时间和系统能耗,缩短了故障时间和维修时间,具有较大的应用前景。     

用模拟退火算法设计光学有限脉冲响应滤波器

提出了利用记忆型模拟回火退火算法来进行基于晶体双折射的光学有限脉冲响应滤波器设计的新方法，用以实现任意光谱响应双通道输出。模拟退火算法模拟了固体缓慢退火从而获得最低内能的物理过程，是一种简单而有效的随机性全局优化算法。应用具有记忆功能的回火退火算法，在较少的晶体延迟片数目下，实现了较优的设计。为证明该设计方法的可行性和有效性，给出了宽波带绿色／品红(Green／Magenta，G／M)滤波器的设计实例以及实验结果。初步研究表明，所设计结果完成满足LCOS投影分合色的要求。     

模拟退火算法与遗传算法在光谱椭偏数据处理中的应用比较

分别利用模拟退火算法和遗传算法在MATLAB 6.5中编写了一个数据处理程序,用来处理VASE椭偏仪在光谱范围为500nm～900nm所测得的椭偏数据。根据VASE椭偏仪的光度法原理,对评价函数方程进行了改进,并选择合适的退火参数和遗传参数进行计算,得到的数据处理结果与仪器处理结果基本一致。最后对两种算法在解决该类问题上的优劣进行了比较。     

逆波束形成在圆阵中的应用

本文讨论将逆波束形成算法应用于圆阵，给出了圆阵FIM（Fourier Intergral Method）积分方程的近似解，并通过滑动平均进一步改善了逆波束形成算法。仿真结果表明：本文所采用的逆波束形成算法旁瓣比常规波束形成算法低4dB，空间方位分辨力提高约1／3。算法的性能通过海试数据处理得到了进一步验证。     

薄膜光学常数的模拟退火算法研究

在薄膜的光学参数拟合时，操作者给定的模型初值对拟合结果有很大的关系。采用广泛用于求解复杂系统优化问题的模拟退火算法来求解薄膜光学常数，降低了操作者给定拟合模型初值的要求。通过对常规的模拟退火算法以及非常快速的模拟退火方法(Very Fast Stimulated Annealing，VFSA)分析，针对性地在模型扰动及退火计划上存在的缺陷作了改进，提高了VFSA算法的计算稳健性。通过大量的拟合实践，验证了提出的改进模拟退火方法的有效性。     

基于模拟退火算法的宽角度X射线超反射镜设计研究

应用于硬X射线波段的宽带多层膜光学元件———宽角度X射线超反射镜的设计可以归结为一个连续变量的多维多极值的全局优化问题。缺少一种有效的全局优化方法是阻碍解决这一难题的一个关键。模拟退火算法是一种简单而且通用的全局优化算法。结合光学多层膜的设计原理提出了利用模拟退火算法来进行宽角度X射线超反射镜设计的新方法。结合已有的方法选择了W和C作为膜层的膜对材料,设计出Cu的Kα线处角度范围0.9°~1.1°反射率达到20%的宽角度X射线超反射镜。并在此基础上采用改进的自适应模拟退火算法实现了Cu的Kα线处宽角度X射线超反射镜的理想设计结果。设计结果表明了模拟退火算法在多层膜最优化设计领域的正确性和有效性。     

Phase unwrapping with regularized phase-tracking technique based on simulated annealing algorithm

We propose an improved regularized phase-tracking (RPT) technique for unwrapping two-dimensional principal phase maps. The size of the neighborhood selected for plane approximation at each pixel is adjustable according to local noise severity. In the area of low noise, the size is small to improve processing speed. On the contrary, a neighborhood with large size is selected to enhance the noise-reduced ability in the high-noisy area. Further, simulated annealing (SA) algorithm is applied to RPT technique for finding the global optimum solutions. Numerical simulation proves that this SA-based RPT phase unwrapping technique is highly robust to noise and experimental results also show its feasibility in practical applications.     

用模拟退火法进行纯永磁波荡器磁块组合优化

 纯永磁波荡器由多个磁块组成,磁块的剩磁离散性会引起波荡器磁场误差,从而影响储存环工作状态和自发辐射谱质量。在波荡器磁块安装之前,使用模拟退火法对磁块进行组合排序优化,可以使峰值场强误差降低到10 ^-4量级以下,磁场一次积分降低到10^-6 T·m量级,二次积分降低到10 ^-6 T·m ²量级,优化结果不依赖于初始状态的选择。给出优化的详细过程,提出了根据磁块剩磁快速计算波荡器峰值场强误差和积分场的方法。     

Sequence alignment using simulated annealing

We apply simulated annealing to amino acid sequence alignment, a fundamental problem in bioinformatics, particularly relevant to evolution. Our goal was obtaining results comparable to those reached through dynamic programming algorithms, like the Needleman-Wunsch algorithm, as well as making a connection between physics and bioinformatics through a representative example.     

Noise barrier optimization using a simulated annealing algorithm

This paper presents a novel method to design a desired noise barrier using the global optimization of a simulated annealing (SA) algorithm. This method focuses on minimizing the barrier dimensions, which are related to material and construction costs, as well as satisfying the target sound pressure levels at receiver points on the condition of traffic noise. Various examples are presented to evaluate the optimized barrier dimensions according to given traffic noise sources, ground topography, surface conditions, and the influence of different receiver positions. As a result, a robust method is found that can determine optimized barrier configurations and dimensions that yield cost reductions.     

基于退火遗传算法的少数投影CT图像重建

不完全投影重建算法研究是当今图像重建算法中的一个难点问题，近年来，遗传算法在其中有一定的应用，但还存在一些问题。针对遗传算法在优化中的不足(如早熟收敛及易陷入局部最小等)，将模拟退火机制引入到遗传算子中，提出一种新的算法；并通过选择测试函数，建立将该算法应用于不完全投影图像重建的优化模型，进行计算机模拟。通过对从投影角间距为15°的12个不同投影视角方向获取的投影数据进行分析，结果表明，该算法的重建图像具有较高的精度。     

基于遗传模拟退火算法的室内照明节能控制方法研究北大核心CSCD

目前的节能照明控制算法仍有陷入局部最优的问题。为了寻求全局最优解,提高室内照明的节能效果,设计一种遗传模拟退火算法对照明系统的控制参数进行优化求解。该算法通过在遗传操作后对优秀个体进行模拟退火处理,增强了算法的局部搜索能力。根据迭代的次数和种群的适应度对遗传概率进行自适应调节,使得算法在前期丰富种群多样性,避免算法“早熟”。提出基于人工神经网络的照度模型来计算室内照度分布,对照明舒适度进行评估,为构造优化算法的适应函数提供了依据。通过仿真实验,在本文介绍的照明场景应用遗传模拟退火算法,并与传统粒子群算法和遗传算法进行比较,其照明节能性能分别高出5.30%和13.61%。     

基于模拟退火算法的超声回波参数估计

本文应用模拟退火算法来改进超声回波模拟问题。仿真结果表明，该算法不需计算梯度，与初始选择值无关，并可在较大范围内搜索。从而避免了应用高斯-牛顿迭代法模拟超声回波的缺陷，即解决了选择的初始值只有在真实值附近才能迭代出最优解的问题。     

Structural optimization and segregation behavior of quaternary alloy nanoparticles based on simulated annealing algorithm

Alloy nanoparticles exhibit higher catalytic activity than monometallic nanoparticles, and their stable structures are of importance to their applications. We employ the simulated annealing algorithm to systematically explore the stable structure and segregation behavior of tetrahexahedral Pt–Pd–Cu–Au quaternary alloy nanoparticles. Three alloy nanoparticles consisting of 443 atoms, 1417 atoms, and 3285 atoms are considered and compared. The preferred positions of atoms in the nanoparticles are analyzed. The simulation results reveal that Cu and Au atoms tend to occupy the surface, Pt atoms preferentially occupy the middle layers, and Pd atoms tend to segregate to the inner layers. Furthermore, Au atoms present stronger surface segregation than Cu ones. This study provides a fundamental understanding on the structural features and segregation phenomena of multi-metallic nanoparticles.     

Memoryless cooperative graph search based on the simulated annealing algorithm

We have studied the problem of reaching a globally optimal segment for a graph-like environment with a single or a group of autonomous mobile agents. Firstly, two efficient simulated-annealing-like algorithms are given for a single agent to solve the problem in a partially known environment and an unknown environment, respectively. It shows that under both proposed control strategies, the agent will eventually converge to a globally optimal segment with probability 1. Secondly, we use multi-agent searching to simultaneously reduce the computation complexity and accelerate convergence based on the algorithms we have given for a single agent. By exploiting graph partition, a gossip-consensus method based scheme is presented to update the key parameter--radius of the graph, ensuring that the agents spend much less time finding a globally optimal segment.     

Optimization of multiple-layer microperforated panels by simulated annealing

Sound absorption by microperforated panels (MPP) has received increasing attention the past years as an alternative to conventional porous absorbers in applications with special cleanliness and health requirements. The absorption curve of an MPP depends on four parameters: the holes diameter, the panel thickness, the perforation ratio, and the thickness of the air cavity between the panel and an impervious wall. It is possible to find a proper combination of these parameters that provides an MPP absorbing in one octave band or two, within the frequency range of interest for noise control applications. However, when a wider absorption frequency band is required, it is necessary to design multiple-layer MPP (ML-MPP). The design of an N-layers MPP depends on 4N parameters. Consequently, the tuning of an optimal ML-MPP by exhaustive search within a prescribed frequency band becomes impractical. Therefore, simulated annealing is proposed in this paper as a tool to solve the optimization problem of finding the best combination of the constitutive parameters of an ML-MPP providing the maximum average absorption within a prescribed frequency band.