粒子群及遗传算法在相对论返波管中的应用

在全三维粒子模拟软件CHIPIC平台上,分别开发了粒子群及基因算法模块.以相对论返波管为例,采用三种不同类型的参数(连续参数、离散参数、混合参数),对粒子群及基因算法进行比较.优化结果表明:粒子群算法的收敛速度更快,在有限的迭代步数内得到的目标结果也更优良,总体表现优于基因算法.     

基于多算法优化SVM的粗糙面参数反演

支持向量机（SVM）是粗糙面参数反演中常用的一种反演算法,SVM反演中的惩罚参数C和核函数参数G对反演结果精度的影响较大,若参数取值不当,会使模型产生"过学习"或者"欠学习"的现象,从而降低预测精度.给出几种SVM参数C和参数G的优化算法,如K折交叉验证（K-CV）、遗传算法（GA）和粒子群算法（PSO）,并在此基础上提出一种基于K-CV和GA改进的PSO算法（GA-CV-PSO）.利用矩量法（MoM）获得的粗糙面后向散射系数构造训练集和测试集,通过不同参数反演的仿真结果对比不同优化算法的反演精度和计算时间,表明GA-CV-PSO算法克服了单一优化算法的缺陷,具有更精确的反演精度和更强的泛化能力.     

Particle swarm optimization and its application to the design of diffraction grating filters

Particle swarm optimization (PSO) is an evolutionary, easy-to-implement technique to design optical diffraction gratings. Design of reflection and transmission guided-mode resonance (GMR) grating filters using PSO is reported. The spectra of the designed filters are in good agreement with the design targets in a reasonable computation time. Also, filters are designed with a genetic algorithm (GA) and the results obtained by the GA and PSO are compared.     

Shaping the Wavefront of Incident Light with a Strong Robustness Particle Swarm Optimization Algorithm

We demonstrate a modified particle swarm optimization(PSO) algorithm to effectively shape the incident light with strong robustness and short optimization time. The performance of the modified PSO algorithm and genetic algorithm(GA) is numerically simulated. Then, using a high speed digital micromirror device, we carry out light focusing experiments with the modified PSO algorithm and GA. The experimental results show that the modified PSO algorithm has greater robustness and faster convergence speed than GA. This modified PSO algorithm has great application prospects in optical focusing and imaging inside in vivo biological tissue, which possesses a complicated background.     

Application of particle swarm+ant colony optimization to calculate the interaction parameters on phase equilibria

In this study, a hybrid particle swarm+ant colony optimization (PSO+ACO) was applied to solve the vapor–liquid phase equilibrium. The NRTL activity coefficient model was optimized with this new algorithm and the binary interaction parameters of twenty mixtures were obtained. The results were compared with the Levenberg–Marquardt algorithm, and show that the PSO+ACO algorithm is a good method to describe the vapor–liquid equilibrium of any binary system.     

基于太赫兹时域光谱和模式识别技术软玉和仿品鉴别

玉石是一种稀有的矿物质,自古以来备受国人喜爱,其真伪鉴别一直是珠宝鉴别行业的棘手难题,传统的鉴别方法已经难以实现对真假玉石的准确鉴别。太赫兹检测技术可以实现快速无损检测,在混合物的分类鉴别方面,有广泛的应用。基于太赫兹时域光谱技术和模式识别技术,对来自我国新疆、青海,以及巴基斯坦、阿富汗四个地区的软玉样品及玻璃、大理石、石包玉三种仿品,使用透射模式测得样品在0.1~1.5 THz频率范围内的太赫兹谱,通过参数提取得到其折射率谱线。由于其化学成分的复杂和多样性,仅靠其特征谱线图,并不能正确的区分软玉和仿品,为了更好的对其进行鉴别,需要建立分类模型。采用主成分分析（PCA）对实验得到的原始折射率数据进行降维和特征提取,作出样品在第一、二主成分上的二维得分图,在图中可以看出软玉和仿品能够很明显的区分开来。在经过降维处理之后的数据中,随机选取其中的四分之三作为训练集,剩下的作为测试集,输入到支持向量机（SVM）建立的分类模型中,并引入网格搜索（GridSearch）、遗传算法（GA）和粒子群算法（PSO）对支持向量机参数进行优化。结果显示,基于网格搜索的支持向量机最优参数c=2.828 4,g=2,识别率为97.7%,运行时间为1.39 s,用时最短;基于遗传算法的支持向量机最优参数c=1.740 1,g=4.544 6,识别率为98.3%,运行时间为3.6 s;基于粒子群算法的支持向量机最优参数c=11.287 2,g=1.833 1,识别率为98.6%,运行时间为6.13 s,用时最长。虽然三种优化算法得到的最优参数不同,但均可实现正确的分类。研究结果表明,使用太赫兹时域光谱技术结合模式识别方法可以快速、准确的鉴别软玉和仿品,这为玉石的鉴别提供了一种新手段。     

群体智能优化中的虚拟碰撞：雨林算法

启发式优化算法中寻优代理过早收敛易陷入局部最优. 本文对此进行机理分析并发现, 虚拟碰撞作为一种隐性过早收敛现象将直接影响群体智能优化算法的准确性与快速性, 而采样过程的无约束性和样本分布信息的缺失是导致虚拟碰撞的根本原因. 为解决上述问题, 本文提出雨林优化算法. 该算法仿照植物生长模式, 利用规模可变种群代替规模限定种群进行分区分级寻优采样, 并结合均匀与非均匀采样原则来权衡优化算法的探索与挖掘, 可以有效减少虚拟碰撞的发生, 在提高寻优效率的同时, 获取精准性和稳定性较高的全局最优解. 与遗传算法、粒子群算法对标称函数的寻优对比实验表明, 雨林算法在快速性、准确性以及泛化能力等方面均具有优势. 关键词： 优化算法 群体智能 进化计算 计算智能     

High-efficiency design of combustion kinetic experiments based on genetic algorithm

Experimental data is essential for the improvement of combustion kinetic models. Experimental design based on model analysis results can screen optimal experimental conditions with maximum information content. However, the computational cost of designing experiments by enumeration becomes unaffordable when an enormity of conditions with different temperatures/pressures/mixtures are to be investigated. An approach to facilitate the efficient discovery of optimal experimental conditions based on the genetic algorithm (GA) is proposed in this work. This approach regards the task of experimental design as an optimization problem to minimize an objective function that measures the information content provided by an experiment. The sensitivity entropy and surrogate model similarity are combined to form the objective function of optimization. Three designs of dimethyl ether experiments are provided to demonstrate the approach. The first case utilizes a benchmark for optimal experiments to validate the effectiveness of GA. The results show that GA can achieve better design results than the traditional enumeration strategy with less than 10% computational cost. The second case illustrates how GA is applied in the design of multiple experiments. The last one is an application in designing multiple experiments of various types, including ignition, species measurements in a jet-stirred reactor (JSR) and a plug flow reactor (PFR). The model parameters are calibrated with the designed experimental data using a Bayesian-based optimization approach. The uncertainties of model parameters are significantly reduced after the optimization.     

Particle swarm modeling of vapor-liquid equilibrium data of binary systems containing CO<Subscript>2</Subscript> + imidazolium ionic liquids based on bis[(trifluoromethyl)sulfonyl]imide anion

Based on particle swarm optimization (PSO), a thermodynamic modeling for the vapor-liquid equilibrium of binary mixtures of carbon dioxide with ionic liquids is presented. The Peng-Robinson equation of state with the Wong-Sandler mixing rules is used to evaluate the fugacity coefficient of the systems. Simulations are carried out in five systems containing 1-alkyl-3-methylimidazolium ionic liquids based on bis[(trifluoromethyl)sulfonyl]imide anion. Then, PSO algorithm was used to minimize the difference between calculated and experimental bubble pressure, and calculate the interaction parameters and the excess Gibbs free energy for all systems used. The results show that the bubble pressures were correlated with low deviations between experimental and calculated values. These deviations show that the PSO algorithm is the preferable method to optimize the interaction parameters of the phase equilibria of binary systems of supercritical carbon dioxide with ionic liquids, and can be used for other similar systems.     

Intelligent reversible watermarking technique in medical images using GA and PSO

The advances in recording, editing, and broadcasting multimedia contents in digital form motivate to protect these digital contents from illegal use, such as duplication, manipulation, and redistribution. However, watermarking algorithms are designed to satisfy requirements of applications, as different applications have different concerns. We intend to design a watermarking algorithm for applications which require high embedding capacity and imperceptibility, to maintain the integrity of the host signal as well as embedded information. Reversible watermarking is a promising technique which satisfies our requirements. In this paper, we concentrate on improving the watermark capacity and reducing the perceptual degradation of an image. We investigated the Luo's [1] additive interpolation-error expansion algorithm and enhanced it by incorporating with two intelligent techniques: genetic algorithm (GA), and particle swarm optimization (PSO). Genetic algorithm is applied to exploit the correlation of image pixel values to obtain better estimation of neighboring pixel values, which results in optimal balance between information storage capacity and imperceptibility. Particle swarm optimization (intelligent technique) is also applied for the same purpose. Experimental results show that PSO and GA nearly give the same results, but GA outperforms the PSO. Experimental results also reveal that the proposed strategy outperforms the state of art works in terms of perceptual quality and watermarking payload.     

嵌入式粒子群-遗传算法的水质COD检测特征波长优化算法

基于紫外-可见光谱法的水质测量中,光谱信号易受到系统噪声干扰、悬浮物散射干扰,且存在信息冗余、多重共线性等特征,导致水质COD测量中特征波长的选取产生较大偏差。因此,提出了基于嵌入式粒子群-遗传（EPSO_GA）算法的水质COD检测特征波长优化算法,以提高波长选择精度。为验证检测特征波长优化算法的可行性,采集了某高校池塘水样、生活污水和排水沟水样的光谱数据,利用EPSO_GA算法对预处理后的光谱数据选取特征波长。EPSO_GA算法采用实数编码方法实现了粒子群（PSO）优化算法和遗传（GA）优化算法的统一编码,在PSO算法中更新粒子时嵌入GA算法的选择、交叉、变异等操作,改善了这两种算法各自在光谱波长特征选取问题上的局限性。将EPSO_GA算法选取的特征波长结合偏最小二乘法（PLS）构建了EPSO_GA_PLS的水质COD预测模型,并且与传统的PSO算法、GA算法选取特征波长建立的PSO_PLS、GA_PLS和全光谱构建的PLS水质COD预测模型做了对比。结果表明：与PSO_PLS,GA_PLS和全光谱构建的PLS水质COD预测模型相比,EPSO_GA改善了PSO算法和GA算法在光谱特征波长选择中早熟和收敛速度慢的问题,降低了全光谱构建PLS水质COD预测模型的复杂度,提高了模型的预测精度。基于EPSO_GA算法建立的EPSO_GA_PLS水质COD预测模型,均方根误差降到了0.212 3,预测精度增加到0.999 3,可以快速定量检测水质COD,为紫外-可见光谱法测COD提供了更好的预测模型。     

激光诱导击穿光谱精确测定燃煤工业分析指标的研究

燃煤工业指标的在线精确分析对于指导燃煤工业优化生产、降低燃煤煤耗至关重要。利用激光诱导击穿光谱(LIBS)分析燃煤煤质时,因受我国复杂多样煤种所导致的“基体效应”,测量精度有待提高。实验中对激光诱导燃煤等离子体光谱至燃煤工业分析指标转化过程中的光谱预处理和定标建模方法进行了优化选择。实验结果表明,利用单/多峰Lorentzian光谱拟合计算谱线强度相比于传统计算方法,谱线强度RSD均值可由12.1%降至9.7%;对于核函数参数寻优,相比于网格参数(Grid)和遗传算法(GA),粒子群算法(PSO)的平均绝对误差(MAE)最小;采用PSO参数寻优式支持向量机(SVM)回归建模的预测均方根误差(RMSEP)小于偏最小二乘回归分析法(PLS);采用单/多峰Lorentzian光谱拟合方法和PSO参数寻优式SVM回归建模,对燃煤工业分析指标预测的平均绝对误差(AAE)为：灰分为16%～30%时AAE为1.37%,灰分大于30%时AAE为1.77%,发热量为9～24 MJ·kg-1时AAE为0.65 MJ·kg-1,挥发分低于20%时AAE为1.09%,挥发分大于20%时AAE为1.02%。     

基于粒子群优化算法的自适应图像分割方法

提出了一种基于粒子群优化算法的图像分割新方法。粒子群优化(PSO)算法是一类随机全局优化技术,它通过粒子间的相互作用发现复杂搜索空间中的最优区域缩短了寻找阈值的时间。将PSO用于基于改进的最佳加权熵阈值法的图像分割中,试验结果表明,该方法不仅能够避免陷入局部极值,而且其速度得到了明显的改善,是一种有效的图像分割新方法。     

A comparative study between LMS and PSO algorithms on the optical channel estimation for radio over fiber systems

In this paper the particle swarm optimization (PSO) and least mean square (LMS) algorithms are comparatively studied to estimate the optical communication channel parameters for radio over fiber systems. It is observed that especially in low noise one tap optical channels, the convergence of LMS algorithm is approximately same with PSO algorithm. On the other hand, as a communication medium, selecting high noisy fiber optical channels or free space optical channels; PSO reaches better mean square error values. The computational complexity which is one of the most important features for optimization algorithms has also been taken into account.     

基于粒子群优化算法的船舶动力定位云模型控制器设计

云模型控制理论是智能控制学科的新兴领域,因此如何扩展云模型的应用范围并使其走向工程化实用化成为其研究重点。针对船舶运动模型具有不确定性和外部扰动随机性等特点,尝试将云模型应用于船舶动力定位的控制过程中。由于云模型控制器存在参数难以整定的问题,提出了基于粒子群算法的优化设计方法。针对标准粒子群优化算法容易出现早熟收敛的问题,引入自适应粒子群优化算法。仿真研究表明云模型控制及粒子群优化的可行性和有效性。     

弱光探测器中复合抛物面聚光镜

基于非成像光学的原理,设计并研制了一种用于弱光探测器的复合抛物面聚光镜.提出基于遗传算法的优化设计聚光镜电子光学系统的新方法,将探测器阳极对光电子的接收率作为遗传进化的目标函数,把复合抛物面聚光镜系统的几何结构参量和电参量作为优化设计的搜索变量,通过控制遗传算法的遗传进化方向,进行全局优化调整.仿真实验结果表明,优化后的聚光镜在会聚光线的同时可以对光电子实现有效的会聚.当入射角为30°时,光透过率高于75%,电子收集率为100%,可以满足具有反射式光电阴极和大探测面积的弱光探测器要求.     

用改进的遗传算法和高斯牛顿法联合反演三维地下水流模型参数

遗传算法在处理非线性优化问题时具有较好的全局搜索性能,但在局部搜索时搜索效率不高,解的精度亦不高,高斯牛顿法在处理非线性优化问题时的性质正好和遗传算法相反,利用遗传算法和高斯牛顿法的优点,用改进的遗传算法和高斯牛顿法联合反演地下水数值模型参数.首先用遗传算法求出地下水模型参数的初值,然后利用这组初值用高斯牛顿法进行数值模型参数的反演,并以一非均质各向同性三维承压非稳定流理想模型为例,结合有限元法讨论了用遗传算法和高斯牛顿法联合反演地下水数值模型参数的过程.计算结果表明,联合参数反演方法,具有收敛速度快、解的精度高的特点,在地下水渗流和水资源评价等领域可广泛应用.     

Simplified Approach for Optimizing Optical Asymmetric Curved Waveguides of Broadband Directional Couplers

ABSTRACT

Optical broadband directional couplers (BDCs) are indispensable components for providing wavelength-insensitive and flexible optical splitting in the construction of functional photonic integrated circuits (PICs). The existing BDC device structures are usually required to determine specific design parameters for different waveguide structures and operating wavelength bands. To circumvent this dilemma, here we present a novel optimization procedure to realize a compact BDC by using the asymmetric curved waveguide structure. The versatile particle swarm optimization (PSO) technique is adopted to determine the optimal device parameters of the compact and broadband asymmetric curved directional couplers (ACDCs) for different coupling ratios. In order to reduce the computational complexity in the optimization, the 3D ACDC is first converted to an equivalent 2D structure by using the modified effective index method (MEIM). The device parameters of the equivalent 2D ACDC are optimized by the PSO with the objective function of a wavelength flattened coupling ratio. Afterward, the optimized 2D structure is converted to the 3D one by including the waveguide thickness. To cope with the approximation error by the MEIM, the 3D ACDC is further fine-tuned by sweeping one of the device parameters with the full 3D simulation but keeping all of the other optimal parameters obtained from the PSO intact. As a result, a DC with broad bandwidth of 100 nm is obtained over the wavelength range from 1.50 µm to 1.60 µm with a very small coupling length of 6 µm. The semi-optimized ACDC is used to construct an unbalanced Mach-Zehnder interferometer (MZI) and a Sagnac loop mirror (SLM), both of which show high extinction ratios of >25 dB over a broad wavelength range with low excess loss.     

ISG-FHEV等效燃油消耗最小控制策略

针对室内复杂环境下火灾识别准确率会降低的问题,提出了一种改进的粒子群算法优化支持向量机参数进行火灾火焰识别的方法。首先在 颜色空间进行火焰图像分割,对获得的火焰图像进行预处理并提取相关特征量;其次采用PSO算法搜索SVM的最优核参数和惩罚因子,并在PSO算法中加入变异操作和非线性动态调整惯性权值的方法,加快了搜索SVM最优参数的精度和速度;然后将提取的火焰各个特征量作为训练样本输入SVM模型进行训练,并建立参数优化后的SVM分类器模型;最后将待测试样本输入SVM模型进行分类识别。算法的火灾识别准确率达到94.09%,分类效果明显优于其他分类算法。仿真结果表明,改进的PSO优化SVM算法提高了火焰识别的准确率和实时性,算法的自适应性更强,误判率更低。     

基于相对论返波管的全三维PIC/PSO数值优化研究

本文研究了粒子群优化算法(PSO),在全三维粒子模拟(PIC)软件CHIPIC平台上, 设计了PSO优化模块, 成功研制了PIC/PSO代码.接着,研究了多频(包括单频)微波的输出功率特性, 并根据该特性设计了一类目标函数.采用该类型优化目标函数, 分别对单频与双频相对论返波管(RBWO)进行了模拟优化.模拟优化结果显示:随着优化过程的进行, 单频RBWO的频率特性向单频靠近,双频RBWO的频率特性则向等幅双频靠近,它们的输出功率都逐渐增大. 这表明通过控制该类型目标函数参数,该PIC/PSO代码可分别对单频与双频RBWO进行优化. 关键词： 粒子群优化算法 粒子模拟 相对论返波管 双频