An artificial neural network approach to multiple-response optimization

In many manufacturing cases, engineers are required to optimize a number of responses simultaneously. A common approach for the optimization of multiple-response problems begins with using polynomial regression models to estimate the relationships between responses and control factors. Then, a technique for combining different response functions into a single scalar, such as a desirability function, is employed and, finally, an optimization method is used to find the best settings for the control factors. However, in certain cases, relationships between responses and control factors are far too complex to be efficiently estimated by polynomial regression models. In addition, in many manufacturing cases, engineers encounter qualitative responses, which cannot be easily stated in the form of numbers. An alternative approach proposed in this paper is to use an artificial neural network (ANN) to estimate the quantitative and qualitative response functions. In the optimization phase, a genetic algorithm (GA) is considered in conjunction with an unconstrained desirability function to determine the optimal settings for the control factors. Two manufacturing examples in which engineers were asked to optimize multiple responses from the semiconductor and textile industries are included in this article. The results indicate the strength of the proposed approach in the optimization of multiple-response problems.     

Multiple response surface optimization with correlated data

Setting of process variables to meet the required specification of quality characteristics is an important problem in the process quality control. There are often several conflicts in quality characteristics, which should be simultaneously satisfied. These types of problems are called “Multiple Response Optimization” (MRO). When quality characteristics are correlated, MRO problems may become increasingly difficult. In design of experiments, identifying covariates effects could reduce error and uncovered variances as well as give more insight about the process. This study aims to identify process variables to consider correlated covariates and correlated quality characteristics. It also accommodates dispersion effects and specification limits as well as location effects in a unified framework based on desirability functions. The features of the proposed method are investigated and the results are compared with some existing techniques by applying two numerical examples.     

Multiple-response optimization for micro-endmilling process using response surface methodology

In the present trend, new fabrication methods for producing miniaturized components are gaining popularity due to the recent advancements in micro-electro mechanical systems. Micro-machining differs from the traditional machining with the small size tool, resolution of x?Cy and z stages. This paper focuses RSM for the multiple response optimization in micro-endmilling operation to achieve maximum metal removal rate (MRR) and minimum surface roughness. In this work, second-order quadratic models were developed for MRR and surface roughness, considering the spindle speed, feed rate and depth of cut as the cutting parameters, using central composite design. The developed models were used for multiple-response optimization by desirability function approach to determine the optimum machining parameters. These optimized machining parameters are validated experimentally, and it is observed that the response values are in good agreement with the predicted values.     

Multiple-response optimisation of electrochemical grinding characteristics through response surface methodology

This paper comprehensively evaluates the influence of voltage and cutting speed on electrochemical grinding of composite carbide inserts. A complete 4² factorial experiment was planned and carried out where each treatment combination has been replicated twice to achieve an acceptable degree of precision. Mathematical models have been developed for the major performance indices like current density, material removal rate (MRR) and surface finish of the job by multiple linear regression analysis. Then multiple-response optimisation has been carried out to find the optimum parameter settings to get a desired yield both by desirability functions as well as by contour overlapping method. It has been found that passivation occurs increasingly at higher voltages and it is removed almost instantaneously at higher speed leading to higher current density and MRR. The surface finish is not very much dependent on grinding speed. The optimum surface finish is obtainable at 12 to 13 V (app.) in electrochemical grinding of composite carbides.     

Development of a multidisciplinary optimization process for designing optimal pharmaceutical formulations with constrained experimental regions

Pharmaceutical formulation design with the development of analytical technologies and methodologies is an integral part of a drug’s research and development. Conventional pharmaceutical formulation design problems are typically studied under the assumption of an unconstrained experimental region with a single quality characteristic. In real world pharmaceutical formulation design problems, however, we are often faced with multiple quality characteristics with a constrained experimental region due to physical limitations. In order to address these needs, we propose a multidisciplinary optimization process for the optimal design of pharmaceutical formulations by developing a response surface simplex design (RSSD) and a response surface D-optimal design (REDD) for experimental modeling and by developing a partitioned composite desirability function (PCDF) for multidisciplinary optimization. Two actual case studies are also presented here for model verification purposes.     

Selection of optimal precision levels and specifications considering measurement error

Measurement errors are unavoidable during any inspection process due to the inherent design of measurement systems and the fluctuation of environmental elements and human errors. The economic and statistical effects of two types of inspection errors should be considered when optimizing the inspection policy. This paper investigates two issues that are important in designing the inspection policy: the determination of specifications and the selection of precision levels for the inspection machine or sensor. For both the “target-the-best” and the “smaller-the- better” quality characteristics, penalty costs and probabilities of two types of inspection errors are incorporated in the optimization to reflect the economic and statistical effects. The total cost to the whole system including both the producer and the consumer is minimized in the decision-making process. The optimal specification limits and the precision level of the inspection machine are determined by solving the optimization model. Numerical examples and sensitivity analysis are given to illustrate the proposed optimization model for both types of quality characteristics.     

Design optimization of slender elastic beam with initial twist using genetic algorithms and finite element analysis

This paper aims to obtain the optimal composite slender beam design for a rotor blade. The cross-sectional dimensions are considered as design variables. The objective is to optimize the box beam to attain a target matrix of stiffness values and maximum elastic coupling. The optimization process is performed using a genetic algorithm (GA), associated with a variational asymptotic method for the structural analysis. The stiffness control of arbitrary, complex section under some design condition is performed in composite beam where the geometrically nonlinear characteristic of the structure is considered. The objective function is defined as the weight, strength and fatigue life. The laminate thicknesses are to be determined optimally by defining a design index comprising a weighted average of the objective functions and determining the minimum.     

Timoshenko梁的变形场重构及传感器位置优化

针对KO位移理论仅适用于重构单方向位移场问题,提出一种适用于六自由度位移场重构的新方法,称之为“多维积分法”。依据Timoshenko梁的静力学平衡方程,建立了位移、转角与外载荷之间的数学模型。并针对不同的外载荷环境,推导出相应的应变场函数和位移场函数,建立了表面应变与截面应变之间的转换关系。为了提升该方法的容差性,以重构位移场的精确性和稳定性为优化目标,建立了关于应变传感器位置的多目标粒子群优化模型。以机翼框架为试验平台,对其进行有限元分析,建立优化目标模型,给出优化后的应变传感器的布置方案。并以此方案为依据,分别利用有限元分析结果和实测梁表面应变值来重构位移场。试验结果表明,提出的“多维积分法”在两种不同形式的外载荷作用下均呈现出较高的重构精度。     

基于多目标非耦合优化策略的可靠性稳健优化设计

以实现多目标问题的可靠性稳健优化设计为目标,通过对目标函数和约束条件进行灵敏度分析,生成目标函数和约束函数的灵敏度附加项,建立了基于灵敏度附加目标函数的可靠性稳健优化设计模型;基于独立公理,使用正交试验和方差分析技术确定设计变量对各个设计目标的影响程度,将设计参数按无耦合或准耦合设计形式分组,把多目标优化问题单目标化,避免多个设计目标之间的反复权衡;结合增广乘子法,应用MATLAB的优化和符号工具箱来实现钳形盘式制动器的可靠性稳健优化设计。算例表明,提出的稳健设计方法具有较高的精度和可靠度。     

Finite element simulation of crack propagation based on phase field theory

The finite element approach is applied to predict crack patterns in a single or composite material under loadings. Crack patterns are represented as variations of a field variable. These variations are determined from the solution of a coupled system of equations consisting of an Allen-Cahn or Ginzburg-Landau type field equation and elasticity equations based on phase field theory. This representation does not require tracking crack tips as in the conventional finite element approach for the modeling of crack propagation problems. For a numerical solution for the system, a finite element algorithm is proposed and implemented into the finite element program “FEAP”. Several numerical simulations are performed and analyzed to predict the crack patterns in 2D single or composite materials under the loadings.     

行星式链-连杆组合机构的优化综合

本文首次提出了设计具有最佳传动角的行星式链-连杆组合机构应满足的条件,并以“链长运动循环”为基础,给出了“输入周期角”、“输出周期角（步进角）”、“步进次数”、“停歇角”和“停歇时间比”等运动参数的定义。本文对该机构进行了优化综合,该优化综合方法不仅适用于上述的情况,而且也适用于输出杆的其他运动或设计要求,只需相应地修改或增减分目标函数和约束条件即可。     

A method of genetic algorithm based multiobjective optimization via cooperative coevolution

The paper deals with the identification of Pareto optimal solutions using GA based coevolution in the context of multiobjective optimization. Coevolution is a genetic process by which several species work with different types of individuals in parallel. The concept of cooperative coevolution is adopted to compensate for each of single objective optimal solutions during genetic evolution. The present study explores the GA based coevolution, and develops prescribed and adaptive scheduling schemes to reflect design characteristics among single objective optimization. In the paper, non-dominated Pareto optimal solutions are obtained by controlling scheduling schemes and comparing each of single objective optimal solutions. The proposed strategies are subsequently applied to a three-bar planar truss design and an energy preserving flywheel design to support proposed strategies.     

基于改进粒子群算法的供应商参与可靠性设计优化

研究供应商参与下的汽车产品子系统可靠性设计的优化问题,考虑供应商参与产品设计的可信度因素,建立以最大化系统的可靠度和供应商的可信度为优化目标的多目标数学规划模型。通过加权的方法把多目标优化模型转化为单目标非线性整数规划模型。采用粒子群(Particle swarm optimization,PSO)算法进行求解,提出适用于“零部件—供应商”关系的离散粒子编码方法。设计带有自适应动态惩罚项的适应度函数,把优化问题转化为无约束优化问题,并将粒子的搜索范围扩展到近可行解空间,进而较好地改进了算法的搜索速度和收敛性能。以某中级轿车传动系统零部件可靠性设计的优化问题为实例,进行仿真研究,应用质量功能展开和模糊评判的方法生成了零部件的权重和供应商可信度初始数据值,仿真结果验证了所提出PSO算法的实用性和有效性。     

On the optimal design of multi-speed gear trainsConception optimale des trains d'engrenages aux vitesses multiples

An efficient method for analyzing multi-speed gear trains is presented and used in this paper. This new technique obtains equations for the diameters of all the gears used in the transmission based on information contained in the speed diagram. The form of the equations is such that they can be generated by the computer automatically. Further, the equations are applicable to the general arrangement, the single composite arrangements, and the double composite arrangements. As a result, all of the promising kinematic arrangements possible for a given number of speeds can be easily studied. By promoting different constraints and objective functions, trace-offs between different parameters can easily be examined. The technique is illustrated using a case study of a 9 speed gear train. A multi-parameter optimization technique is used to solve 19 different arrangements for a weighted objective function minimizing volume and maximizing stiffness.     

Multi-objective reliability-based topology optimization of structures using a fuzzy set model

This research proposes a multi-objective reliability-based topology optimization (MORBTO) for structural design, which considers uncertain structural parameters based on a fuzzy set model. The new technique is established in the form of multi-objective optimization where the equivalent possibilistic safety index (EPSI) is included as one of the objective functions along with mass, and compliance. This technique can reduce complexity due to a double-loop nest problem used previously due to performing single objective optimization. The present technique can accomplish within one optimization run using a multi-objective approach. Two design examples are used to demonstrate the present technique, which have the objectives as structural mass and compliance with the constraint of structural strength. The results show the proposed technique is effective and simple compared to previous techniques.

     

云平台数据驱动的产品与供应商资源主从协同优化

针对传统产品和供应商资源协同研究以历史数据或经验为输入,基于“归一化”或“层次优化”方法造成需求响应滞后和失真等问题,在分析两者正反向交互特征、解析数据传播途径的基础上,提出了基于云平台数据驱动反向设计（DID）的产品与供应商资源主从协同优化系统决策方法。构建了基于云平台DID的产品和供应商资源协同配置框架;建立了基于云平台DID的主从协同优化决策模型,其中,基于正向客户需求的产品配置充当上层主者,下层供应商联合配置充当从者,反向影响产品配置的决策;利用双层嵌套遗传算法进行模型权衡解优化;最后给出了汽车座椅的产品与供应商资源协同优化案例。结果表明,所提方法在真实表征产品和供应商资源联合配置过程的数据传递、协同关系方面具有显著优势。     

面向云制造系统复杂任务请求的服务组合优化框架

为破解因多任务、强QoS约束及任务过量等因素导致云制造系统组合效果不佳的问题,研究并提出了面向复杂任务请求的全局优化策略框架。该框架以多任务全局优化、组合服务捆绑与共享为基本原则,提出了“单组合执行每任务”、“多组合执行每任务”及“多组合执行多任务”三种组合模式并建立了问题模型,最后利用基于混合算子的矩阵编码遗传算法予以实现。实验结果表明,该框架能高效、高质量响应云制造系统的复杂任务请求。     

Infrared and Visible Images Fusion with Multi-visual Cues

Image fusion is an important technique which combines the original information from multiple input images into a single composite image. The fused images will be more beneficial to human visual perception or further computer processing tasks than any individual input. Most of the traditional infrared and visible fusion approaches perform the fusion on the assumption that the original information is measured by local saliency features such as contrast or gradient. There is little consideration of the “interesting” or “useful” information in global. In this paper, an infrared and visible image fusion method is proposed by considering the final aim of image fusion, the human visual perception and further image processing tasks. The fusion is implemented under the non-subsampled contourlet transform based image fusion framework. The low frequency sub-band coefficients which represent the intensity of the scene are fused with the weight map which is constructed by considering both visual saliency uniqueness and task-oriented objectness, and refined by spatial consistency with guide filter. The new fusion strategy ensures that the objects being “interesting” or “useful” are preserved in the fused image. Sixteen pairs of infrared and visual images are used to test the validation of the proposed method. The experimental results show obvious improvement of the proposed method in terms of both objective and subjective quality measurements corresponding to other methods.     

Optimisation of machining parameters of glass-fibre-reinforced plastic (GFRP) pipes by desirability function analysis using Taguchi technique

This paper presents a new approach for optimizing the machining parameters on turning glass-fibre-reinforced plastic (GFRP) pipes. Optimisation of machining parameters was done by an analysis called desirability function analysis, which is a useful tool for optimizing multi-response problems. In this work, based on Taguchi’s L₁₈ orthogonal array, turning experiments were conducted for filament wound and hand layup GFRP pipes using K₂₀ grade cemented carbide cutting tool. The machining parameters such as cutting velocity, feed rate and depth of cut are optimized by multi-response considerations namely surface roughness, flank wear, crater wear and machining force. A composite desirability value is obtained for the multi-responses using individual desirability values from the desirability function analysis. Based on composite desirability value, the optimum levels of parameters have been identified, and significant contribution of parameters is determined by analysis of variance. Confirmation test is also conducted to validate the test result. It is clearly shown that the multi-responses in the machining process are improved through this approach. Thus, the application of desirability function analysis in Taguchi technique proves to be an effective tool for optimizing the machining parameters of GFRP pipes.     

超精密机床中柔性铰链的结构多目标模糊优化设计

应用模糊优化原理和有限元分析技术,对超精密机床中柔性铰链的结构进行了模糊多目标优化设计。在设计过程中,利用权重最大最小法和隶属函数形式的选择,综合考虑了各个目标函数与约束条件的重要程度,实现了该结构的三个垂直方向刚度和扭转刚度的合理分配。设计结果表明,本文所采用的方法对于工程实际问题是完全可行的。