Evolutionary fuzzy SVR modeling of weld residual stress

Residual stresses are an integral part of the total stress acting on any component in service. It is important to determine and/or predict the magnitude, nature and direction of the residual stress to estimate the life of important engineering parts, particularly welded components. Researchers have developed many direct measuring techniques for welding residual stress. Intelligent techniques have been developed to predict residual stresses to meet the demands of advanced manufacturing planning. This research paper explores the development of Finite Element model and evolutionary fuzzy support vector regression model for the prediction of residual stress in welding. Residual stress model is developed using Finite Element Simulation. Results from Finite Element Method (FEM) model are used to train and test the developed Fuzzy Support Vector Regression model tuned with Genetic Algorithm (FSVRGA) using K-fold cross validation method. The performance of the developed model is compared with Support Vector Regression model and Fuzzy Support Vector Regression model. The proposed and developed model is superior in terms of computational speed and accuracy. Developed models are validated and reported. The developed model finds scope in setting the initial weld process parameters.     

On the machining induced residual stresses in IN718 nickel-based alloy: Experiments and predictions with finite element simulation

Residual stresses after machining processes on nickel-based super alloys is of great interest to industry in controlling surface integrity of the manufactured critical structural components. Therefore, this work is concerned with machining induced residual stresses and predictions with 3-D Finite Element (FE) based simulations for nickel-based alloy IN718. The main methods of measuring residual stresses including diffraction techniques have been reviewed. The prediction of machining induced stresses using 3-D FE simulations and comparison of experimentally measured residual stresses for machining of IN718 have been investigated. The influence of material flow stress and friction parameters employed in FE simulations on the machining induced stress predictions have been also explored. The results indicate that the stress predictions have significant variations with respect to the FE simulation model and these variations can be captured and the resultant surface integrity can be better represented in an interval. Therefore, predicted residual stresses at each depth location are given in an interval with an average and standard deviation.     

On the equivalences and differences of evolutionary algorithms

Evolutionary algorithms (EAs) are fast and robust computation methods for global optimization, and have been widely used in many real-world applications. We first conceptually discuss the equivalences of various popular EAs including genetic algorithm (GA), biogeography-based optimization (BBO), differential evolution (DE), evolution strategy (ES) and particle swarm optimization (PSO). We find that the basic versions of BBO, DE, ES and PSO are equal to the GA with global uniform recombination (GA/GUR) under certain conditions. Then we discuss their differences based on biological motivations and implementation details, and point out that their distinctions enhance the diversity of EA research and applications. To further study the characteristics of various EAs, we compare the basic versions and advanced versions of GA, BBO, DE, ES and PSO to explore their optimization ability on a set of real-world continuous optimization problems. Empirical results show that among the basic versions of the algorithms, BBO performs best on the benchmarks that we studied. Among the advanced versions of the algorithms, DE and ES perform best on the benchmarks that we studied. However, our main conclusion is that the conceptual equivalence of the algorithms is supported by the fact that algorithmic modifications result in very different performance levels.     

Multiobjective optimization for crash safety design of vehicles using stepwise regression model

In automotive industry, structural optimization for crashworthiness criteria is of special importance. Due to the high nonlinearities, however, there exists substantial difficulty to obtain accurate continuum or discrete sensitivities. For this reason, metamodel or surrogate model methods have been extensively employed in vehicle design with industry interest. This paper presents a multiobjective optimization procedure for the vehicle design, where the weight, acceleration characteristics and toe-board intrusion are considered as the design objectives. The response surface method with linear and quadratic basis functions is employed to formulate these objectives, in which optimal Latin hypercube sampling and stepwise regression techniques are implemented. In this study, a nondominated sorting genetic algorithm is employed to search for Pareto solution to a full-scale vehicle design problem that undergoes both the full frontal and 40% offset-frontal crashes. The results demonstrate the capability and potential of this procedure in solving the crashworthiness design of vehicles.     

石油含水率测量动态补偿模型研究

针对传统微波透射法测量石油含水率存在的测量误差大等问题,提出了一种基于神经网络的动态补偿方法,确定衰减和相移两个参量作为动态补偿模型的输入;针对传统BP算法具有收敛速度慢、容易陷入局部极小值等缺点,采用微粒群训练算法对神经网络动态补偿模型进行训练,可使微波透射石油含水率测量结果的补偿过程具有寻优全局性和精确性。实验结果表明,利用该技术对石油含水率测量结果进行校正是一种有效的方法,具有一定的应用价值。     

基于现场总线的L-天冬酰胺酶II发酵过程优化控制

简要介绍L-天冬酰胺酶Ⅱ和PROFIBUS—DP现场总线,针对传统发酵过程控制系统布线复杂、自动化程度低等缺陷,设计了一个基于PROHBUS—DP现场总线的L-天冬酰胺酶Ⅱ发酵过程的优化控制系统,通过PROFIBUS—DP总线,主站、从站和各智能仪表间可直接通信,形成开放性网络。为了提高最大产酶量,将智能、优化控制技术融入过程控制系统,着重采用神经网络非线性预测控制和粒子群优化算法解决溶解氧的优化问题,达到较好的控制效果,产酶量比原工艺提高1倍以上。实验证明：该系统经济、可靠。     

一种基于多神经树集成的分类模型

针对数据挖掘领域分类问题的特点.提出了基于多神经树集成的分类模型(CMBNTE).该模型利用改进遗传规划算法和粒子群算法,实现单个神经树模型的优化;借鉴集成学习思想,将多个神经树模型组合成最终的分类模型.在6个UCI数据集上的实验结果表明,该模型能较好地解决分类问题,尤其适用于多分类属性的复杂分类问题.     

一种新的基于进化计算的聚类算法

聚类是数据挖掘领域的重要研究内容之一。针对遗传聚类算法较好的稳定性与粒子群优化算法较强的局部搜索能力,在交叉、变异算子后叠加粒子群优化算子的方法实现了二者的结合,提出了GAPSO聚类算法,既保持了遗传算法的稳定性与泛化性的优势,又发挥了PSO算法收敛效率高的特点。通过对10组二维空间上的聚类样本进行实验研究显示,GAPSO聚类算法在收敛效率上显著优于GA聚类算法,在稳定性上优于PSO聚类算法。     

Optimization of type-2 fuzzy weights in backpropagation learning for neural networks using GAs and PSO

In this paper the optimization of type-2 fuzzy inference systems using genetic algorithms (GAs) and particle swarm optimization (PSO) is presented. The optimized type-2 fuzzy inference systems are used to estimate the type-2 fuzzy weights of backpropagation neural networks. Simulation results and a comparative study among neural networks with type-2 fuzzy weights without optimization of the type-2 fuzzy inference systems, neural networks with optimized type-2 fuzzy weights using genetic algorithms, and neural networks with optimized type-2 fuzzy weights using particle swarm optimization are presented to illustrate the advantages of the bio-inspired methods. The comparative study is based on a benchmark case of prediction, which is the Mackey-Glass time series (for τ = 17) problem.     

面向环境监测的WSN节点定位技术研究

WSN节点定位在无线传感器网络研究中意义非凡,设计出一种精确的定位算法是当今的重大挑战。传感器节点采集的数据只有在获取到节点的位置信息后才有意义,结合环境监测特点和应用需求,DV-Hop(Distance Vector-Hop)算法因其受环境影响相对较小,无需大量硬件开销,适用于环境监测场景。针对传统DV-Hop算法定位精度不高的问题,提出基于加权因子的混合DV-Hop算法——HDV-Hopw,其采用两种策略对传统DV-Hop算法进行改进。首先,通过对信标节点的平均每跳距离进行加权处理,减小平均每跳距离带来的误差;然后,将未知节点位置估计转换成目标优化,采用混合GA-PSO算法对未知节点的坐标进行优化,通过限制初始种群的可行域以及改进初始种群的质量来提高算法的定位精度。仿真实验结果表明,在没有增加额外硬件设备的情况下, 相比于DV-Hop算法 ,HDV-Hopw算法的 定位误差平均降低了11%左右。     

求解多目标优化问题的一种多子群体进化算法

提出一种新的多目标粒子群优化（MOPSO）算法,根据多目标优化问题（MOP）的特点,将一个进化群体分成若干个子群体,利用非劣支配的概念构造全局最优区域,用以指导整个粒子群的进化．通过子群体间的信息交换．使整个群体分布更均匀,并且避免了局部最优,保证了解的多样性,通过很少的迭代次数便可得到分布均匀的Pareto有效解集．数值实验表明了该算法的有效性.     

A novel classification method: A hybrid approach based on extension of the UTADIS with polynomial and PSO-GA algorithm

In this study, a new multi-criteria classification technique for nominal and ordinal groups is developed by expanding the UTilites Additives DIScriminantes (UTADIS) method with a polynomial of degree T which is used as the utility function rather than using a piecewise linear function as an approximation of the utility function of each attribute. We called this method as PUTADIS. The objective is calculating the coefficients of the polynomial and the threshold limit of classes and weight of attributes such that it minimizes the number of misclassification error. Estimation of unknown parameters of the problem is calculated by using a hybrid algorithm which is a combination of particle swarm optimization algorithm (PSO) and Genetic Algorithm (GA). The results obtained by implementing the model on different datasets and comparing its performance with other previous methods show the high efficiency of the proposed method.     

一种粒子群—神经模糊系统的研究与应用

针对神经网络与模糊逻辑协同系统NFCS(Neuron-Fuzzy Cooperation System)的学习算法存在收敛速度慢和易陷入局部极小点等问题,提出将粒子群优化算法PSO(Particle Swarm Optimization)与NFCS结合的新型系统PSO-NFCS.在PSO-NFCS中,PSO代替原先的学习算法,由其进化预置网络的连接权值、阈值和补偿参数,以实现网络的学习和精确推理.将其应用于某石油化工装置的故障诊断,结果表明PSO-NFCS是有效的,其全局收敛能力、收敛速度和泛化精度等性能均优于原先的学习算法.     

粒子群算法在投影寻踪模型优化求解中的应用

粒子群优化(Particle Swarm Optimization,PSO)算法是一种新兴的优化技术,其思想来源于人工生命和进化计算理论.PSO算法通过粒子追随自己找到的最好解和整个群体的最好解完成问题的优化.针对投影寻踪模型中的最佳投影方向优化问题.运用PSO算法和惩罚函数法相结合对该优化问题进行了计算.仿真实验结果表明:PSO算法对于求解有复杂约束的非线性目标函数优化问题是可行的,且算法的收敛速度快,编程结构简单,易于实现,从而为各领域运用投影寻踪模型评价方法提供了强有力的寻优方法,具有较广的应用前景.     

基于种群小生境微粒群算法的前向神经网络设计

根据自然界中鱼鸟等所具有的种群运动特征,借鉴递阶编码的思想,构造出一种种群小生境微粒群算法,具有小生境内个体微粒自由运动特征分量和小生境种群运动特征分量分层递阶进化的特征,克服了标准微粒群算法或其改进算法在多蜂函数寻优时出现的微粒“早熟”现象,应用该算法进行三层前向神经网络连接权值和网络结构联合并行自适应设计,在混沌时间序列预测中显示了良好的性能。     

A support vector regression based prediction model of affective responses for product form design

In this paper, a state-of-the-art machine learning approach known as support vector regression (SVR) is introduced to develop a model that predicts consumers’ affective responses (CARs) for product form design. First, pairwise adjectives were used to describe the CARs toward product samples. Second, the product form features (PFFs) were examined systematically and then stored them either as continuous or discrete attributes. The adjective evaluation data of consumers were gathered from questionnaires. Finally, prediction models based on different adjectives were constructed using SVR, which trained a series of PFFs and the average CAR rating of all the respondents. The real-coded genetic algorithm (RCGA) was used to determine the optimal training parameters of SVR. The predictive performance of the SVR with RCGA (SVR–RCGA) is compared to that of SVR with 5-fold cross-validation (SVR–5FCV) and a back-propagation neural network (BPNN) with 5-fold cross-validation (BPNN–5FCV). The experimental results using the data sets on mobile phones and electronic scooters show that SVR performs better than BPNN. Moreover, the RCGA for optimizing training parameters for SVR is more convenient for practical usage in product form design than the timeconsuming CV.     

WSN低能耗数据收集遗传粒子群算法研究

针对设施农业无线传感器网络节点分布不均匀、能量约束严格的特点,为降低网络总能耗,提出一种改进的遗传粒子群算法,构建一棵树高受限且网络总能耗最小的数据收集树。首先,随机生成连通图网络,采用父节点表示法将生成树编码成粒子;然后,设计一种随机生成数据收集树算法,随机产生满足树高限制的生成树;最后,考虑节点能耗均衡,设计一种粒子单点突变算法,实现对节点能耗最优值的比较。通过粒子单点变异、交叉以及优化新粒子,提高了种群多样性,避免了算法过早陷入局部最优解,在满足时延要求的同时,降低了网络总能耗。实验表明,与有树高约束的DL-DCT算法相比,所提算法降低了7.34%的网络总能耗,延长了网络平均生存期。     

基于BP神经网络的医学图像分割新方法

对于医学图像而言,其分割结果的准确性对医生诊断病情并给出正确的治疗方案至关重要。应用传统的BP神经网络对医学图像进行分割,存在对初始权重值敏感、学习速率固定、收敛速度慢和易陷入局部极小值等问题。因此,提出了一种基于改进的粒子群优化算法的BP 神经网络的医学图像分割方法。首先,应用粒子群优化算法与BP神经网络的映射关系,通过粒子群强大的搜索功能找到最佳适应函数,使对应的BP神经网络的均方误差达到最小值,克服了BP 神经网络产生多个局部最小值的可能;其次,确定粒子的最佳位置后,在BP神经网络学习中获得最合理的权值和偏置值,以提高网络的收敛速度;最后,BP神经网络经反复训练后,获得最佳输出值,并计算阈值,通过阈值来分割图像区域。实验结果表明,利用改进的算法能够得到更清晰的图像分割效果,提高了图像的分割精度,对临床的诊断也具有重要参考意义。     

基于多种群进化与粒子群优化混合的频谱分配算法

为了解决认知无线网络中的频谱分配问题,提出一种基于多种群进化与粒子群优化混合的频谱分配算法。它采用图论着色模型,首先使用遗传算法将多个种群进行独立进化,以提高种群的全局搜索能力;然后选出每个种群中的最优的个体作为粒子群优化的粒子,并通过控制每个粒子的初始速度方向来加快算法的收敛速度。最后以系统总收益最大化和用户间的公平性为优化目标与遗传算法和粒子群算法进行了对比实验,仿真结果表明,该算法在收敛速度、认知用户接入公平性和系统总收益3个方面的性能均优于遗传算法和粒子群算法。     

Simulation-based heuristic method for container supply chain network optimization

With the emerging of free trade zones (FTZs) in the world, the service level of container supply chain plays an important role in the efficiency, quality and cost of the world trade. The performance of container supply chain network directly impacts its service level. Therefore, it is imperative to seek an appropriate method to optimize the container supply chain network architecture. This paper deals with the modeling and optimization problem of multi-echelon container supply chain network (MCSCN). The problem is formulated as a mixed integer programming model (MIP), where the objective is subject to the minimization of the total supply chain service cost. Since the problem is well known to be NP-hard, a novel simulation-based heuristic method is proposed to solving it, where the heuristic is used for searching near-optimal solutions, and the simulation is used for evaluating solutions and repairing unfeasible solutions. The heuristic algorithm integrates genetic algorithm (GA) and particle swarm optimization (PSO) algorithm, where the GA is used for global search and the PSO is used for local search. Finally, computational experiments are conducted to validate the performance of the proposed method and give some managerial implications.