基于本体案例匹配的扰动作业车间智能调度辅助决策

为了快速响应扰动事件,恢复生产稳定,创新性地提出了多类型扰动事件本体与车间生产调度案例本体智能匹配技术。通过引入本体技术（ontology technology）,将以往生产动态调度的成功案例进行规范化和标准化处理,构建动态调度本体化的成功案例库和案例库自我完善的学习机制。结合案例推理（case-based reasoning,CBR）技术及相似度理论,将新扰动的生产车间调度问题与案例库的成功案例进行匹配,依据相似度所在阈值区间,实现扰动车间智能调度辅助决策,从而达到提高决策过程时效性,缩短决策时间的目的。通过仿真结果表明：与传统CBR及混合驱动调度策略相比,在本体-CBR的方法下比传统CBR的案例检索精度大约提高了6%,比混合驱动调度策略相比,在决策时间上要快25 min,故有效地提高了案例检索的精度与响应时间。     

基于Java的基因本体工具包

针对基因本体的需求,利用Java技术,以模块化设计的方式实现了基因本体工具包GO4J.该系统包括4个模块.解析器模块把GO载入到内存中;图数据结构模块记录了所有的GO条目之间的关系;语义相似性模块评估两个GO条目之间的相似程度;可视化模块直观展示GO条目和根节点之间的路径.GO4J把这4个模块整合到本体浏览器中,可以结合GO注释的生物数据来对具体的生物数据进行分析.     

基于CBR的中心渔场预报

针对海洋渔业遥感信息与资源评估服务系统中中心渔场预报的问题,提出了一种多策略的基于CBR的趋势预测方法,通过不同的相似性度量方法,复用完全相似或者条件相似的历史序列,利用领域专家规则对预报结果进一步修正,取得了较好的预防结果,相关系统正在应用推广之中。     

The tooling system configuration in a new manufacturing system architecture

A new manufacturing system architecture, called Mod Flex Prod (MFP), has been recently proposed to match the balance between productivity and flexibility required by suppliers of components for the automotive industry. Among the relevant issues that cause the new architecture to be more profitable than transfer lines and flexible manufacturing systems, tooling has been shown to have an important role. Starting from this result, this paper deepens the advantages in tooling attainable with the new architecture. In particular, it presents a new approach for configuring the tooling system (i.e. selecting the number of tool copies for each tool type) which takes into account that tool copies can be required by the machines to perform operations, by the tool carrier to transport tools within the system and by the tool room when inserts have to be reconditioned. With reference to a real case study, the new approach allows one to assess that tooling investment can be reduced by about 20% if the new architecture replaces traditional flexible manufacturing systems.     

OPTIMIZING MULTI-RESPONSE PROBLEMS IN THE TAGUCHI METHOD BY FUZZY MULTIPLE ATTRIBUTE DECISION MAKING

One of the conventional approaches used in off-line quality control is the Taguchi method. However, most previous Taguchi method applications have only dealt with a single-response problem and the multi-response problem has received only limited attention. The theoretical analysis in this study reveals that Taguchi's quadratic loss function and the indifference curve in the TOPSIS (Technique for order preference by similarity to ideal solution) method have similar features. The Taguchi method deals with a one-dimensional problem and TOPSIS handles multi-dimensional problems. As a result, the relative closeness computed in TOPSIS can be used as a performance measurement index for optimizing multi-response problems in the Taguchi method. Next, an effective procedure is proposed by applying fuzzy set theory to multiple attribute decision making (MADM). The procedure can reduce the uncertainty for determining a weight of each response and it is a universal approach which can simultaneously deal with continuous and discrete data. Finally, the effectiveness of the proposed procedure is verified with an example of analysing a plasma enhanced chemical vapour deposition (PECVD) process experiment. © 1997 by John Wiley & Sons, Ltd.     

A module generation algorithm for product architecture based on component interactions and strategic drivers

Benefits of modularity are often achieved from module independence that allows for independent development to reduce overall lead time and economies of scale due to sharing similar modules across products in a product family. Current modularity methods tend to describe only one of these views, either the module–module independence or the product–product shared module similarity. This paper proposes a new hybrid module generation algorithm that balances both module independence and product similarity, allowing product similarity strategy to influence the coupling-driven architecture considerations. The proposed method builds on two popular matrix-based methods: the design structure matrix approach and modular function deployment that each has been developed to support these two different aspects of the module generation. This paper presents a novel algorithm that integrates both views and allows a balanced clustering that takes both interactions and company portfolio strategy into account. Usefulness of the algorithm is presented using a cordless handheld vacuum cleaner as a case study and by comparing it to alternative approaches.     

基于案例的产品设计系统中程序型知识检索研究及实现

基于Internet的计算机辅助工业设计系统,基于案例的计算机辅助工业设计系统研究基础上,结合Case-based Reasoning CBR技术,采用语义隶属度分析的方法,提出了基于设计问题的程序型知识获取模型以及基于相似度计算的程序型知识检索算法,为计算机辅助工业设计提供了一个新的思维方式,并在CBID系统构建中得到实践与验证.     

机械结构相似度评价及可靠性预计研究

在用传统的相似产品法进行可靠性预计时,所选取的相似产品与评价对象之间差异较大,会造成预计结果的偏差较大。为解决这一问题,提出了一种基于"功能—运动—动作"结构化分解的可靠性预计新方法。通过明确产品的各大运动功能建立功能层,分析完成各功能的机械部件运动建立运动层,确定实现各机械部件运动的元动作运动单元建立元动作层,建立了可靠性预计的"功能—运动—元动作运动单元"层次模型。引入区间层次分析法,确定了"功能—运动—元动作运动单元"层次模型中各子代运动单元对父代运动单元的可靠性影响权重,并构建了由子代运动单元预计父代运动单元可靠性的数学模型。确定影响评价机械结构相似度的因素,构建了区间数综合评判云模型并选取与元动作运动单元相似的参考结构,以区间数综合评价云模型得出的相似度值修正参考结构的可靠性数据来预计元动作运动单元的可靠性。将元动作运动单元的可靠性预计值代入数学模型,层层预计父代运动单元的可靠性,从而得出整个产品的可靠性水平。最后以数控转台为例进行了分析,得出了可靠性预计值,并将该方法与传统方法进行了比较。结果表明,新的可靠性预计方法具有较好的可行性和较高的准确性。     

船舶铁舾件工时定额估算方法研究

为了解决人工工时估算不准确且费时等问题,准确测算船舶铁舾件舾装作业的工时定额,提出了基于CBR和BP神经网络的铁舾件装配工时定额制定方法.针对铁舾件细分工种安装工艺相似但型号各异的特点,提出基于非数值型和数值型特征参数相结合的相似性规律,在已知部分型号舾装工时情况下,利用相似性检索方法找出与新铁舾件最相似的铁舾件集,根...     

基于本体的机械产品知识仓库研究与应用

为解决产品开发全生命周期中知识管理与知识重用问题,提出了基于本体的机械产品知识仓库系统构架。从语法、属性、结构、实例四个方面进行相似性的计算,实现将源本体的实例映射到目标本体的过程。采用OWL本体表示语言对产品知识进行本体建模。并对基于本体的案例推理进行了深入的研究。最后将研究成果应用于制造业知识管理系统的机械产品知识仓库子系统中。     

A case-based reasoning approach toward developing a belief about the cost of concept

It is generally acknowledged that product development involves a sequence of decision making under uncertainty, including setting target requirements for a new product, selecting product concept, and developing conceptual and detailed design of a chosen concept. To select a product concept, engineers need to assess the uncertainty of a future market share, market size, and a cost of concept (cost of the final product developed from a concept). This paper proposes a case-based reasoning (CBR) approach to model beliefs about the uncertainty of a cost of concept. The proposed CBR approach consists of storing information about various products in a knowledge-base, defining a new product concept, retrieving a cluster of products in the knowledge-base that are highly similar to the concept, and adapting the cost of the retrieved product to construct a distribution of the cost of concept. This paper illustrates the proposed approach using printers as an example.     

Rapid tooling route selection and evaluation for sand and investment casting

Today, a number of direct routes using rapid prototyping (RP) processes (fused deposition modelling (FDM), laminated object manufacturing (LOM), stereolithography apparatus (SLA), selective laser sintering (SLS), etc.), as well as indirect routes (RP coupled with secondary or soft-tooling processes like RTV vacuum casting) are available for rapid fabrication of tooling for sand and investment casting processes. Each route is unique in terms of geometric, material, quality and cost characteristics; no comprehensive database of their capabilities has been reported, especially for metal casting applications. The problem of selecting the optimal rapid tooling (RT) route is a complex multi-criteria decision-making problem. This paper describes a systematic approach for RT route selection and planning. A database of RT process capabilities was generated through benchmarking experiments, covering 20 different widely used RT routes (both direct and indirect: two- and three-step processes) involving an impeller pattern. In this approach, RT process route selection involves translating the tooling requirements specified by the casting engineer into weighted tooling attributes using quality function deployment and analytic network process (QFD-ANP), which along with part specifications is used for RT route selection by calculating the overall process compatibility indices. The routes are ranked as per the value of the overall compatibility index. Once the optimal route is selected, process planning is carried out by retrieving a similar process plan using case-based reasoning (CBR). The methodology has been implemented in a software program using Visual C + + programming language in a Windows environment. The methodology is demonstrated and validated with an industrial example of a separator body casting. It has proved to be a robust evaluation and decision-making tool for selecting appropriate tooling route for a given casting based on customer requirements.     

基于案例推理的产品个性化定制方法

目的 针对产品个性化定制过程中用户需求与产品模块匹配度低的问题,提出一种基于案例推理的产品个性化定制方法。方法 首先,构建某类产品案例库和用户需求信息库,并经过对产品案例的模块化分解,构建模块化案例库。当用户在个性化定制系统输入需求信息时,企业利用质量功能展开,建立用户需求项与产品模块属性之间的映射关系,确定用户需求权重和各产品模块的属性及其权重。然后,运用相似度计算、检索,匹配出符合用户需求的产品模块。将产品模块按相似度值进行排序,相似度值较高的模块作为备选模块,使用户可以通过个性化定制系统进行产品模块的配置。最终,系统生成产品个性化定制方案。结论 以民族风桌布为例,阐述了产品个性化定制的具体方法与流程,并验证了该方法的有效性,为企业提供了一种高效的产品个性化定制方法。     

基于SVM的多生物特征融合识别算法

针对单生物特征识别的局限性,提出融合手背静脉和虹膜两种生物特征实现身份识别.基于尺度不变特征变换(SIFT)提取手背静脉的局部SIFT特征并对特征点进行匹配,利用特征匹配率作为手背静脉图像的相似度测度.通过Haar小波变换实现虹膜特征编码,利用加权汉明距对虹膜进行相似度测试.最后基于支持向量机(SVM)实现两种生物特征在匹配层的融合识别.利用CASIA虹膜数据库和TJU手背静脉数据库对算法性能进行测试,其等错率为0.02%,实验结果表明,该融合算法具有很高的识别性能,为生物特征识别研究提供了新思路.     

An extended multi-objective replacement model for flexible automation investments

A multi-objective replacement model is developed for flexible automation investments, for use in a decision-support-system environment. The model is based on a general mixed integer goal programming (GP) formulation. The objectives of the model are to derive an optimal implementation sequence for new CNC modules and tooling, replacement schedule for current equipment and production plans for transition and later periods in the planning horizon. Several types of flexibility are considered, in addition to tooling costs and standardization issues. The GP model minimizes a weighted function of the deviational variables for direct profitability, as well as other strategic goals of the firm.     

Fatigue and ratcheting assessment of AISI H11 at 500°C using constitutive theory coupled with damage rule

Hot extrusion is one of the most commonly used manufacturing methods for metal plastic deformation, and the consumption of extrusion tooling is considerably high due to its fatigue damage under cyclic serving condition. Hot‐work tool steel AISI H11 is one of these typical materials employed in extrusion tooling. This work is dedicated to calculating the stress/strain state of AISI H11 and predicting its lifetime at high temperature 500°C by building a unified constitutive model coupled with Lemaitre's damage law. Tensile tests and strain/stress reversed cycling tests have been conducted at 500°C to investigate mechanical properties and damage evolution. A unified constitutive model with Armstrong‐Fredrick/Ohno‐Wang kinematic hardening rule and a new proposed isotropic hardening rule is built; Lemaitre's damage law is employed as well. Parameters are determined based on tests and are temperature dependent. Finite element simulation of the deformation behaviour and fatigue lifetime is implemented into commercial software ABAQUS Standard v6.14‐2 with user material subroutine to validate the proposed method. The comparison shows good agreement with experimental results, and this part of work is essential and crucial to subsequent structure analysis.     

A Bayesian network approach for fixture fault diagnosis in launch of the assembly process

Verification and correction of faults related to tooling design and tooling installation are important in the auto body assembly process launch. This paper introduces a Bayesian network (BN) approach for quick detection and localisation of assembly fixture faults based on the complete measurement data set. Optimal sensor placement for effective diagnosis of multiple faults, structure learning of the Bayesian network and the diagnostic procedure are incorporated in the proposed approach. The effective independence sensor placement method is used to reach the desired number of optimal sensor locations, which provide the concise and effective sensor nodes to build the diagnostic Bayesian network. A new algorithm based on conditional mutual information tests is put forward to learn the Bayesian network structure. The body side assembly case was used to illustrate the suggested method and the simulation analysis was performed to evaluate the effectiveness of the diagnostic network. The work demonstrated that the proposed methodology composes a feasible and powerful tool for fixture fault diagnosis in launch of the assembly process.     

基于邻域相似比的图形图像分类方法研究

通过研究图像邻域相似关系,提出了"邻域相似比"的概念。邻域相似比能够表征图像邻域颜色的相似程度。然后以此为核心构建了图形图像分类模型,在Matlab环境中编译算法进行了仿真实验。结果表明,利用邻域相似比进行图形图像分类,算法简单快捷,运算成本低,正确率高。     

Video enhancement through image registration based on structural similarity

Abstract

It is commonly known that the mean square error (MSE) does not accurately reflect the subjective image quality for most video enhancement tasks. Among the various image quality metrics, structural similarity (SSIM) metric provides remarkably good prediction of the subjective scores. In this paper, a new registration method based on contribution of structural similarity measurement to the well known Lucas–Kanade (LK) algorithm has been proposed. The core of the proposed method is contributing the SSIM in the sum of squared difference of images along with the Levenberg–Marquardt optimisation approach in LK algorithm. Mathematical derivation of the proposed method, based on the unified framework of Baker et al., is given. The proposed registration algorithm is applied to a video enhancement successfully. Various objective and subjective comparisons show the superior performance of the proposed method.     

A hybrid case-based reasoning CAD system for injection mould design

Injection mould design generally lies on the critical path of a new product development. The design efficiency will have significant impact on the overall lead time of a new product. This paper presents a prototype injection mould-design system using a hybrid case-based reasoning (HCBR) approach. Case-based reasoning (CBR) is a solving paradigm that uses previous episodes on solving problems similar to the problem at hand (the new problem) as the basis for solving the new problem. In this hybrid system, CBR is incorporated with generalized design knowledge, and provides a flexible and comprehensive model of design. The knowledge base of the system would be accessed by mould designers through interactive programs so that their own intelligence and experience could also be incorporated with the total mould design. The approach provides a workable model of mould design system with CBR and knowledge-based expert system intelligent support, which could suggest good and proven design solutions to new design problems quickly, avoiding the time necessary to create those designs from scratch, for the plastic products manufacturing industry.