Application of a Particle Swarm Optimization Algorithm in Truss Structure Optimal Design

The particle swarm optimization (PSO) algorithm developed in recent years is a stochastic optimization algorithm based on swarm intelligence. It possesses advantages such as being a simple concept, ease of implementation and low resource occupation. The PSO algorithm was adopted to solve the problem of size optimization of truss structures with stress and displacement constraints. We present the basic principle of the original PSO algorithm in detail, and then introduce a constriction coefficient to modify it. Moreover, reasonable values of the coefficients are proposed for the modified PSO algorithm. Several classical problems are solved using the modified PSO algorithm, and the results compared with those using traditional optimization algorithms and genetic algorithms. Numerical results show that the modified PSO algorithm has good convergence and stability, and can be applied to the size optimization of truss structures.     

Intelligent optimization design of novel flash structure for crank-shaft forging dies

The parameters of resistance wall, which is a novel flash structure for forging die, are studied. The effects of the resistance walls parameters are analyzed and the important influence factors are screened by using the fractional factorial design. The Latin hypercube method is used to select sample points of the important design variables which are analyzed by finite elements simulation. The surrogate models are established by taking the simulation result as response and the parameters of the resistance wall structure as variables. The model is converted into single objective function by linear weighting method and is optimized by using particle swarm optimization algorithm for global optimization. Finally,the optimization results are compared and verified with those obtained by genetic algorithm. The results show that the PSO (particle swarm optimization) algorithm has better convergence than the traditional genetic algorithm and can realize optimization of the parameters of the resistance wall structure.     

Airfoil design using improved particle swarm optimization

Based on the theory of standard particle swarm optimization (PSO), an improved particle swarm optimization algorithm is presented, and it has a better optimized performance than standard PSO. A multi-objective wind turbine airfoil shape optimization model is established and 4 kinds of different thick wind turbine airfoils with better performance are designed by using the improved PSO algorithm. The aerodynamic performance of the CQU-A18 and CQU-A21 airfoils are analyzed in detail compared with the commonly used wind turbine airfoil with the same thickness. The results show that the new airfoils show very good aerodynamic characteristics, and they are found to be very insensitive to leading edge roughness. The new airfoils exhibit the higher lift coefficient and larger lift/drag ratio in both smooth condition and rough condition at the main angle of attacks. The performances of the new airfoils show a significant improvement compared with the typical airfoils.     

Image reconstruction of electrical impedance tomography based on particle swarm optimization algorithm

Electrical impedance tomography (EIT) has many advantages in practical application,but image reconstruction of EIT is a highly ill-posed, non-linear inverse problem. Newton-Raphson algorithms are widely used in EIT, which have to calculate the Jacobian matrix and use regularization techniques. So this kind of algorithms is complex and less stable. To address the problem, a new static image reconstruction method for EIT is proposed based on particle swarm optimization (PSO) algorithm. PSO is a population-based, adaptive search optimization technique. It is simple in concept, few in parameters, quick in convergence and easy in implementation. The model of EIT forward problem is given and some appropriate improvements in PSO are made to accommodate the solution of EIT. Compared with Newton-Raphson(MNR) algorithms, PSO only uses an iterative processing to get the best solution instead of using a complicated Jacobian matrix. The experimental results indicate that using PSO-based algorithm to solve image reconstruction of EIT, the position of mutation region is more accurate and graphics space resolution is much higher.     

Truss Structure Shape Optimization withDifferential Evolution Algorithm

Differential Evolution (DE) was introduced to get the global optimum and overcome the difficulties encountered by coupling two types of design variables in the shape optimization of truss structures with stress, geometry, and local stability constraints. The basic principle of DE algorithm was presented in detail first, and then mathematical model for shape optimization of truss structures was presented, in which two types of design variables, such as the node coordinates and section areas, were considered simultaneously. Several classical problems were solved with DE algorithm, and the results were compared with those using the other optimization methods. It was shown that DE algorithm had good convergence and stability and could be applied for shape optimization of truss structures effectively.     

A parallel cluster algorithm for self adaptive particle swarm optimization

Full scale data mining, such as in cluster problems, requires large numbers of computations. A parallel cluster algorithm for self adaptive particle swarm optimization was proposed to deal with this problem. The proposed parallel particle swarm optimization algorithm reduced the impact of the initial conditions via parallel searches of the globally best position amongst a varied population. Task parallelization and partially asynchronous communication of the algorithm were employed to decrease computing time. Furthermore, if combined with the characteristics of self adaptive and dynamical optimization parameters of the parallel particle swarm algorithm, the problems of particle mobility loss and the end of evolution could be dealt with successfully. When modified thusly, the algorithm maintains individual diversity and restrains degeneration. The simulation experiments indicate the algorithm helps increase computing speed and improve cluster quality.     

Optimum Analysis of Damper Braces Based on Genetic Algorithm

The aims at optimization of structure with damper braces is studied in this paper. The sum of damping coefficient of all damper braces is considered as a goal function, and the storey displacements are considered as constraint conditions. The program,which introduces the non-proportional damp matrix, based on genetic algorithm and time history is used to analyze the optimization of a damper braced frame. The results are reasonable and show that the genetic algorithm is an efficient method to optimize the damper braced structure.     

Organically Combinatorial Inference and Optimization in Intelligence System

The inference algorithm is the most important part in intelligence system because the level of intelligence in the system is decided by it. By means of the mutual benefit for inference algorithm of expert system ,fuzzy logic and neural network ,the combinatorial inference technology which is organically composed of these three parts is put forward for inference mechanism in intelligence system. The optimization decision model is also set up. In order to bring all the advantage of every inference algorithm into play and overcome the disadvantage of single inference algorithm the common expert knowledge base is applied to organic combination and concurrent operation of all inference algorithm. In order to realize knowledge share the inference results are optimized by decision technology optimization. The research results show that the organized combinatorial inference and optimization can be applied in engineering practice effectively and is benefit for raising the inference level.     

De nosing method based on multiscale morphological filter optimized by particle swarm optimization algorithm

The traditional morphological filter is difficult to remove the noise of the vibration signal,because the signal has the characteristics of shocking and nonlinear. A new method based on multiscale morphological filter optimized by particle swarm optimization algorithm is proposed. The multiscale morphological filter is constructed according to the character of morphological algorithm. The particle swarm optimization algorithm is used to select the adaptive structure element,which plays an important role in morphological filter,achieving to get the optimal morphological filter. The signal is filtered through different scales of morphological filters and the noise removed signal is gotten through weight algorithm. The simulated signal and the bearing fault signal are analyzed,and the results show that the optimal morphological filter works better in removing noise and can effectively reduce the noise of the mechanical equipment.     

Research on multi objective global optimization model and algorithm of integrated service chain

Based on the network model of integrated service chains and evaluation index of candidate service resources, optimizing integrated service chain can be formally defined as a multi objective global optimization model with multiple constraints. We propose a multi objective global optimization algorithm based on improved multi objective genetic algorithms. The proposed algorithm uses a distance based nonparametric population diversity measurement operator, and diversity control is involved in the process of adaptive value assignment, elitist maintaining and selection operation. The proposed algorithm can optimize multiple objectives at the same time on the premise of meeting the constraints, and finally get a constrained Pareto optimum solution set which satisfy decision makers’ prefers. The simulation experiments indicate that the proposed algorithms can achieve global convergence and has better solution quality and distribution, which efficiently solve the problem of integrated service chain multi objective global optimization.     

改进的粒子群算法及在数值函数优化中应用

为提高粒子群算法的优化能力，提出了一种改进的粒子群优化算法。在该算法中，采用Beta分布初始化种群，采用逆不完全伽马函数更新惯性权重，在速度更新式中，引入了基于差分进化的新算子，对于粒子的越界处理，采用了基于边界对称映射的新方法。以50个不同类型的数值函数作为优化实例，基于威尔柯克斯符号秩检验的测试结果表明，该算法明显优于普通粒子群优化算法、差分进化算法、人工蜂群优化算法和量子行为粒子群算法。     

基于智能算法优化SVM的横向通风过程中温度场预测方法探究

将改进的智能预测技术应用于储粮横向通风过程中的粮堆温度预测,为粮食通风智能预测与决策提供了一种新思路。选取河北清苑国家粮食储备库冬季横向通风的实时监测数据,在分析主要影响因素的基础上,应用三种智能优化算法——网格寻优算法、GA遗传算法寻优、PSO粒子群算法,结合回归支持向量机理论,对粮堆的通风过程进行建模。结果表明,优化过的回归预测模型能较好地拟合粮食温度与其他变量之间的非线性关系,尤其是当样本数量较为有限时,该方法具有更高的拟合精度,更适合对储粮通风这一强非线性过程的预测研究,对于人工干预操作具有一定的现实指导意义。     

基于PSO优化LS-SVM的木材含水率软测量建模

木材含水率是木材干燥过程中重要的技术指标。针对木材干燥过程具有强耦合、大滞后、非线性的特点以及木材含水率检测存在的问题,提出一种软测量方法。利用最小二乘支持向量机(LS-SVM)对非线性系统时间序列数据进行学习,建立被控对象的软测量模型,同时通过粒子群优化(PSO)算法对LS-SVM的惩罚因子和核函数参数进行滚动优化,提高软测量模型的预测精度。将木材干燥窑内的温度、湿度以及木材含水率作为样本数据,通过PSO优化的LS-SVM方法建立木材含水率的软测量模型,进而利用该模型实现对目标检测点木材含水率的软测量。仿真结果表明,PSO-LSSVM软测量模型预测精度高,泛化能力强,满足木材干燥控制系统的实际测量要求。     

Evaluation of government‐supported provincial innovation competitiveness—Evidence from China

Government support plays a crucial role in enhancing regional innovation capabilities and creating distinctive innovation clusters. Using the innovation competitiveness indicators of the 31 provinces in China, this paper examines the provincial government collaboration and innovation competitiveness values through Super‐SBM data envelopment analysis and analytic hierarchy process. Then, the particle swarm optimization (PSO) algorithm is employed to relate government collaboration to innovation competition and cooperation among provinces in hope of exploring the developmental trends of innovation cluster optimization. Research findings show that only 25.8% of the provincial government support proves effective for regional innovation, and the competitiveness values of the provinces are generally low and need to be improved. By incorporating the role of government support into the innovation competition and cooperation among provinces, there has been a tendency of innovation activities clustering in innovative regions and three major clusters have formed. The trend of clustering will continue to evolve outwards so that the overall innovation level of China would be enhanced.     

基于粒子群算法和支持向量机的黄花菜叶部病害识别

使用数字图像处理技术,以黄花菜叶部病害图像为识别对象,基于Lab空间和K-means聚类算法分割病害区域,提取目标区域的颜色特征、方向梯度直方图(histogram of oriented gradient,HOG)特征和形状特征,分别建立单一特征模型和特征融合模型,采用粒子群(particle swarm optimization,PSO)算法通过交叉验证优化支持向量机(support vector machine,SVM)模型的惩罚因子和核参数,建立基于PSO-SVM的多特征融合分类模型识别黄花菜病害。基于SVM的多特征融合分类模型识别率高于单一特征分类模型,识别率可达为81.67%;基于PSO-SVM多特征融合分类模型识别率高达92.39%。基于PSO-SVM的多特征分类模型识别率高,可以及时、便捷、高效地识别黄花菜病害。     

自适应加权最小二乘支持向量机的空调负荷预测

为了提高建筑空调负荷的预测精度,在分析空调负荷主要影响因素的基础上提出了一种基于自适应加权最小二乘支持向量机(AWLS-SVM)的建筑空调负荷预测方法。该方法根据预测误差的统计特性,采用基于改进正态分布加权规则,自适应地赋予每个建模样本不同的权值,以克服异常样本点对模型性能的影响。建模过程中采用粒子群优化(PSO)算法对模型参数进行优化,以进一步提高模型预测精度。基于DeST模拟数据将AWLS-SVM方法应用于南方地区某办公建筑的逐时空调负荷预测中,并与径向基神经网络(RBFNN)模型、LS-SVM模型及WLS-SVM模型作比较,其平均预测绝对误差分别降低了51.84 %、13.95 %和3.24 %,并进一步基于实际空调负荷数据将该方法应用于另一办公建筑的逐日空调负荷预测中。预测结果表明：AWLS-SVM预测的累积负荷误差为4.56 MW,亦优于其他3类模型,证明了AWLS-SVM具有较高的预测精度和较好的泛化能力,是建筑空调负荷预测的一种有效方法。     

罗非鱼池塘养殖溶解氧预测研究

为及时有效地掌握池塘养殖中溶解氧浓度的变化趋势,保障罗非鱼稳定高效养殖,在对罗非鱼池塘养殖实际情况进行研究和分析的基础上,采用粒子群算法对BP神经网络模型进行参数优化,针对无锡市2015 年8 月23—11 月4 日这段时间内南泉实验基地的水产养殖溶解氧进行预测。同时,将粒子群优化BP神经网络模型与BP神经网络模型的训练结果和预测结果进行对比。研究结果表明,PSO-BP优化模型的训练和预测结果远远优于普通BP神经网络模型,除异常点外,误差率基本均低于0.5%。同时,该模型收敛速度快,计算复杂性低,能够较好的体现和预测罗非鱼池塘养殖的溶解氧趋势,也为其他水质指标的预测提供了研究方向。     

电源分配网络阻抗分析及去耦电容优化

高速印制板电源分配网络（power delivery networks,PDN）受电电源端口具有瞬时电流大、目标阻抗小的特点。针对任意形状电源板的阻抗问题,提出一种基于解析式与有限差分法的互补模型,并在此基础上结合粒子群算法实现多个受电电源端口的阻抗优化。通过Matlab仿真,发现互补模型的结果与全波有限元法的基本吻合,计算时间则大幅缩短。测量结果表明,优化后实验板的多个受电电源端口阻抗均满足各自的目标值要求。