The use of the Taguchi method with grey relational analysis to optimize the thin-film sputtering process with multiple quality characteristic in color filter manufacturing

The paper proposes an approach to improve the yield of Chrome (Cr) thin-film sputtering process of the black matrix (BM) in color filter manufacturing and to find the robust parameters of the process with multiple quality characteristics by using the Taguchi method combined with the grey relational analysis. The research considers the correlation between quality characteristics and applies the principal component analysis to eliminate the multiple co-linearity. The weights of the quality characteristics are determined by employing the entropy measurement method. Two quality characteristics and four control factors with three levels were selected. Based on the Taguchi quality design concept, a L9 orthogonal array table was chosen for the experiments. The confirmation experiment verifies the proposed grey-based Taguchi method has the ability to find out the optimal process parameters with multiple quality characteristics. Besides, manufacturing with the attained optimal process parameters can reduce the opportunities of repair and rework of Cr thin-film and raise the yield.     

Form design of product image using grey relational analysis and neural network models

This paper presents a new approach to determining the best design combination of product form elements for matching a given product image represented by a word pair. A grey relational analysis (GRA) model is used to examine the relationship between product form elements and product image, thus identifying the most influential elements of product form for a given product image. A grey prediction (GP) model and a neural network (NN) model are used individually and in conjunction with the GRA model, in order to predict and suggest the best form design combination. An experimental study on the form design of mobile phones is conducted to evaluate the performance of these models. Based on expert surveys, the concept of Kansei Engineering is used to extract and evaluate the experimental samples, and a morphological analysis is used to extract form elements from these sample mobile phones. The evaluation result shows that all the NN-based models outperform the GP-based models, suggesting that the NN model should be used to help product designers determine the best combination of form elements for achieving a desirable product image. The GRA model can be incorporated into the NN model to help designers focus on the most influential elements in form design of mobile phones.     

Optimal displacement amplification ratio of bridge-type compliant mechanism flexure hinge using the Taguchi method with grey relational analysis

Microsystem Technologies - The compliant mechanism flexure hinge has been frequently utilized in precision enginering in recent years, such as the bridge-type and rhombus-type compliant mechanisms....     

采用灰色关联分析法确定水网络瓶颈的研究

工业的发展使得水资源的需求量越来越大,节水已经成为人们普遍关心的问题之一.在工业生产中节水具有很强的针对性,只有针对系统的关键因素、即瓶颈进行节水才能真正提高整个系统的节水性能.水网络的瓶颈受到很多因素的影响,而灰色关联分析法是一种研究贫信息、少数据的方法,可以综合考虑多个影响因素对目标的影响.本文将灰色关联分析法应用于单杂质用水网络中,根据灰色关联度值给出了瓶颈单元的定义.并对用水单元中影响新鲜水用量的浓度、杂质负荷等多个因素进行了综合分析,通过计算扰动工况与正常工况下各因素差值的灰色关联度值,来确定用水网络中的瓶颈单元,提供了确定瓶颈的详细步骤和方法.本文将提出的瓶颈辨识方法应用于一个含有10个用水单元的单杂质用水网络,对其进行了分析研究,确定了该网络的瓶颈单元.并通过比较当瓶颈单元和非瓶颈单元的杂质负荷分别增加时,新鲜水量的改变值,验证了所确定瓶颈单元的准确性.分析结果表明,改变非瓶颈单元的扰动因素对网络新鲜水用量无影响,改变瓶颈单元的扰动因素网络新鲜水用量则会由166.2t/h增至167.1t/h.灰色关联分析法综合考虑多种影响因素,可以合理有效的确定用水系统的瓶颈单元.     

一种应用PSO优化RBF神经网络的方法

RBF神经网络算法是一种常用的数据训练方法,在该训练过程中,如何选取更合理的个体作为RBF神经网络的神经元,直接关系到该数据训练方法的性能.利用传统的RBF神经网络模型进行数据训练,由于不同的神经元之间的差异性较小,造成建立的RBF神经网络集成模型的精确度过低.为此,提出应用PSO优化RBF神经网络的方法.动态构造PSO优化RBF神经网络结构,针对不同的动态构造方法进行分类,得到网格删除法、网络构造法和综合法等不同的动态构造方法,在动态构造的基础上,建立引用PSO优化RBF神经网络模型,计算RBF神经网络中的粒子变量,获取对应的适应性值,得到RBF神经网络的输出结果,实现应用PSO优化的RBF神经网络建模.实验结果表明,利用改进算法进行RBF神经网络构建,能够降低RBF神经网络的数据训练误差,满足实际需求.     

The use of grey relational analysis in solving multiple attribute decision-making problems

There are many cases in daily life and in the workplace which pose a decision problem. Some of them involve picking the best from among multiple available alternatives. However, no single alternative works best for all performance attributes. This research proposes a multiple attribute decision making (MADM) method, grey relational analysis (GRA), for solving this kind of problem. Two cases, facility layout and dispatching rules selection problem, which have been analyzed by data envelopment analysis (DEA), were also analyzed using the GRA procedure, in order to illustrate the use of GRA. In the case of the facility layout problem, 18 alternative layouts and 6 performance attributes were considered. In the case of the problem of selecting dispatching rules, 9 alternatives dispatching rules and 7 performance attributes were considered. For the two cases examined, the results of comparisons show that GRA is efficient for solving MADM problem.     

基于灰色关联熵的煤与瓦斯突出概率神经网络预测模型*

煤与瓦斯突出是严重威胁矿井安全生产的重大自然灾害之一。为解决煤与瓦斯突出影响因素、突出危险性关联预测问题,在综合分析煤与瓦斯突出影响因素的基础上,利用灰色关联熵理论分析影响因素与突出危险性的关联度,得到各影响因素的权重及关联度排序,并结合概率神经网络（PNN）原理,构建基于灰色关联熵的煤与瓦斯突出PNN预测模型。用煤与瓦斯突出样本数据,对影响因素加权的PNN模型进行训练和测试。结果表明：用灰色关联熵分析可获得影响因素与突出危险性的关系,量化输入变量的重要性;瓦斯放散初速度、开采深度对于煤与瓦斯突出危险性的影响程度最大,可重点对瓦斯放散初速度、开采深度进行预处理以产生更为理想的预测效果;该预测模型能更好地考虑影响因素对突出危险性的综合影响,改善煤与瓦斯突出危险性预测的准确性。     

A new method to optimize the satellite broadcasting schedules usingthe mean field annealing of a Hopfield neural network

Reports a new method for optimizing satellite broadcasting schedules based on the Hopfield neural model in combination with the mean field annealing theory. A clamping technique is used with an associative matrix, thus reducing the dimensions of the solution space. A formula for estimating the critical temperature for the mean field annealing procedure is derived, hence enabling the updating of the mean field theory equations to be more economical. Several factors on the numerical implementation of the mean field equations using a straightforward iteration method that may cause divergence are discussed; methods to avoid this kind of divergence are also proposed. Excellent results are consistently found for problems of various sizes.     

灰色关联分析和支持向量机相融合的网络安全态势评估

为提高网络安全态势评估的准确性, 提出一种灰色关联分析和支持向量机相融合的网络安全态势评估模型。根据网络安全态势评估原则进行评估指标体系选择, 并根据灰色关联分析确定指标权重, 将训练样本输入到支持向量机进行训练, 采用改进粒子群算法优化支持向量机参数, 建立网络安全态势评估模型, 最后采用数据集KDD Cup99对模型性能进行仿真测试。仿真结果表明, 该模型可以准确、客观地对网络安全态势进行评估, 评估结果可以为网络管理员提供一定价值的参考建议。     

Optimizing the design of automotive S-rail using grey relational analysis coupled with grey entropy measurement to improve crashworthiness

The optimization of the crashworthiness and lightweight design of S-rail extracted from the frontal body in white was studied in this paper. A physical test was conducted to verify the validity of S-rail model and then an implicit parameterization model was built based on the S-rail model using the software SFE-CONCEPT. Based on the implicit parameterization modeling, a steel-aluminum S-rail was designed to reduce the peak collision force (PCF) and increase the specific energy absorption (SEA) under the condition that the total weight (M) of S-rail does not increase. L₁₆ (4⁵) Taguchi array was used to collect sample points which will be prepared for the optimization design. The experimental results were analyzed through grey relational analysis (GRA) coupled with grey entropy measurement method. The multi-objective optimization was then converted into a single objective optimization problem based on the grey relational grade. The optimal combination of design parameters for S-rail was obtained using the proposed method. Meanwhile, a comparison was presented between the proposed method and other extensively used methods (i.e. NSGA-II, MOPSO, and ASA), and the proposed method reduces the PCF and M to 26.81% and 46.01% respectively, and increases the corresponding SEA by 176.06%. Moreover the computational cost can be reduced by 143.5% at least when compared with other extensively used methods. Therefore, the hybrid method can efficiently improve the crashworthiness and reduce the computational cost during the design process of S-rail.     

Research on the controller of an arc welding process based on a PID neural network

A controller based on a PID neural network （PIDNN） is proposed for an arc welding power source whose output characteristic in responding to a given value is quickly and intelligently controlled in the welding process. The new method syncretizes the PID control strategy and neural network to control the welding process intelligently, so it has the merit of PID control rules and the trait of better information disposal ability of the neural network. The results of simulation show that the controller has the properties of quick response, low overshoot, quick convergence and good stable accuracy, which meet the requirements for control of the welding process.     

Application of a neural network to a design optimization process

This article discusses the implementation of a class of design problems in a neural network. the problems, identified at an abstract (i.e., connectivity) level of building design, are characterized as optimization types of problems. Architects often have to generate floor-plan layouts of a building optimized with respect to several points of view. Examples of these points of view, often stated in terms of normative requirements, are the social norms: community, privacy, and circulation-cost. an architectural design problem in the presence of even a single norm from among the above norms is computationally hard and intractable. During the last four decades there have been attempts to automate floor-plan design considering a single norm or a limited number norms. Most of these attempts have traditionally been based on combinatorial enumeration methods. Recent progress in AI has paved the way for intelligent handling of the architectural design processes using knowledge-based system technology and heuristics programming. This article examines the possibility of the neural networks approach in generating connectivity patterns of building with respect to specific social norms. Our first attempt has been focused on linear-tree type designs with respect to single norms. the idea in this work was borrowed from the Hopfield model of the neural network for implementation of the Traveling Salesman Problem, because of the similarity of our design problems with this problem. Hopfield and other researchers used constant parameters for different-sized problems. Close examination of the network and experiments revealed that this approach does not guarantee a convergence for every case, and chaotic behavior is expected in cases for which the chosen parameters are not appropriate. to overcome this problem this article suggests problem-dependent and problem-size-dependent parameters that vary for each case. Test results from the implementation convey that the approach yields satisfactory results and is worth it to explore its application to other classes of optimization problems.     

Design of a robust fuzzy controller for the arc stability ofCO2 welding process using the Taguchi method

CO(2) welding is a complex process. Weld quality is dependent on arc stability and minimizing the effects of disturbances or changes in the operating condition commonly occurring during the welding process. In order to minimize these effects, a controller can be used. In this study, a fuzzy controller was used in order to stabilize the arc during CO(2) welding. The input variable of the controller was the Mita index. This index estimates quantitatively the arc stability that is influenced by many welding process parameters. Because the welding process is complex, a mathematical model of the Mita index was difficult to derive. Therefore, the parameter settings of the fuzzy controller were determined by performing actual control experiments without using a mathematical model of the controlled process. The solution, the Taguchi method was used to determine the optimal control parameter settings of the fuzzy controller to make the control performance robust and insensitive to the changes in the operating conditions.     

Using linguistic preferences and grey relational analysis to evaluate the environmental knowledge management capacity

This study proposes a combined fuzzy grey relational analysis method based to deal with study objective. This study objective is aimed to present a perception approach to deal with supplier evaluation of environmental knowledge management capacities (EKMC) with uncertainty and lack of information. The ranking of best supplier might be a key strategic direction of other suppliers prior to EKMC. The solving procedure is as follows, (i) the weights of criteria and alternatives are described both in qualitative and quantitative information using fuzzy set theory; (ii) using a grey relational analysis to result the ranking order for all alternatives; (iii) an empirical study of supplier ranking problem in EKMC are used to resolve with this proposed approach and the result indicates that optimal supplier is with higher protection of the environmental knowledge from inappropriate or illegal use or theft (C7) and from the best alternative supplier to study its criteria ranking.     

实时控制网络中信息的分析及优化

本文介绍了ABB MasterPiece200／1 PLC控制系统在轧机润滑系统中的应用，着重描述了系统配置及控制功能。     

配电网主馈线分段开关优化配置的新方法

配网分段开关的优化配置是一个要同时考虑技术性、可靠性和经济性的非线性整数规划问题.针对这个问题,提出一种在原有网架与开关的基础上,确定分段开关最佳新增数量和安装位置的规划方法.该方法的目标函数综合考虑了年分段开关投资费用、年运行维护费用和年停电损失费用;考虑投资约束对辖区指数区间加以修正,计算需新增的分段开关数量;再以期望供缺电量作为评价函数,用枚举法来对已知数量分段开关配置进行优化.通过对临湘电力局临城Ⅱ回配电系统的实例计算,证明了该方法的工程实用性.     

A sensitivity analysis of a back-propagation neural network for manufacturing process parameters

Back-propagation neural networks that represent specific process parameters in a composite board manufacturing process were analyzed to determine their sensitivity to network design and to the values of the learning parameters used in the back-propagation algorithm. The effects of the number of hidden layers, the number of nodes in a hidden layer, and the values of the learning rate and momentum factor were studied. Three network modification strategies were applied to evaluate their effect on the predictive capability of the network. The convergence criteria were tightened, the number of hidden nodes and hidden layers was increased. These modifications did not improve the predictive capability of the composite board networks.     

A novel analysis and design of a neural network assisted nonlinear controller for a bioreactor

A novel approach is presented for the analysis and the design of a controller for a bioreactor. It is based on the model reference control theory, assisted by a neural network identifier. The control objectives specified in the paper require the controller to be a nonlinear one, however, it is shown that it is stable in the sense of bounded input bounded output and locally stabilizing in the sense of Lyapunov. The feasibility and the efficacy of the proposed approach are tested on the benchmark problem. Copyright © 1999 John Wiley & Sons, Ltd.     

用蛙跳算法优化RBF神经网络参数的研究

针对径向基函数（Radial Basis Functions,RBF）神经网络结构参数确定问题,提出了一种基于蛙跳算法优化RBF神经网络参数的新方法。将RBF神经网络参数组成一个多维向量,作为蛙跳算法中的参数进行优化。以适应度函数为标准,在可行解空间中搜索最优解,并对蛙跳算法进行了改进。非线性函数逼近实验结果表明,该优化算法相对标准遗传优化算法、粒子群优化算法有较小的均方误差,具有更好的逼近能力。     

Structural optimization of the cross-beam of a gantry machine tool based on grey relational analysis

Crossbeam structural design of gantry machine tool is a multi-level, multi-index and multi-scheme decision-making problem. In order to solve the above problem, the optimum seeking model of crossbeam structure was built through using the grey relational analysis and Analytic Hierarchy Process. The finite element analysis of the static and dynamic performance parameters for four kinds of crossbeam structural schemes designed had been done, and the optimal design scheme was selected by using the optimum seeking model. After conducting sensitivity analysis for the optimal crossbeam selected, the reasonable design variables were obtained, and the dynamic optimization design model of crossbeam was established. Six groups of non-inferior solutions were obtained after solving the optimization design model. The optimal solution was selected from the non-inferior solution set through using the crossbeam structural optimization method based on grey relational analysis again, which makes the crossbeam’s dynamic performance improving greatly. The dynamic experiments on the crossbeams before and after optimization design were conducted, then the experimental results show that the first four order natural frequencies of the crossbeam increase 17.56 %, 19.36 %, 17.04 % and 19.58 % respectively, which proves that the structural optimization design method based on grey relational analysis proposed in this paper is reasonable and practicable.