Recognizing manufacturing features from a design-by-feature model

This paper presents a methodology for recognizing manufacturing features from a design feature model. The common feature-feature interacting relationships are categorized. A feature recognition processor first translates the design feature model of a part into an intermediate manufacturing feature tree by handling design features according to their properties and the interacting relationships between features. Through combination, decomposition, and (tool approach direction) TAD-led operations, alternative interpretations of manufacturing feature model for the part are then generated, and the manufacturing feature tree is updated and extended with AND/OR operators to store these interpretations. Finally, a single interpretation with the lowest machining cost will be selected in the manufacturing feature tree. The proposed processor can support a dynamic and effective recognition process of manufacturing features during the design stage of a part. By defining the interactions between volumetric features elaborately, and utilizing design features and auxiliary information, the processor can recognize manufacturing features from complex parts. The processor recognizes not only some essential manufacturing features but also replicate, compound and transition features defined in STEP. The alternative interpretations can be used for a generic manufacturing application environment.     

并行环境下注塑件智能设计支撑技术研究

注塑件设计是注塑模设计制造的第一步,注塑件设计的优劣以及注塑件产品模型信息的完备与否,对注塑模设计制造的难易以及成形工艺参数的选择都具有十分重要的作用。本文从构造新一代注塑模设计制造系统的出发,探讨注塑件智能设计的支撑技术,提出采用并行工程设计思想,以基于特征的产品建模方法构造注塑件产品模型;结合神经网络和专家系统的特长,构造神经网络与专家系统混合系统;利用特征和知识紧密的关系,实现注塑件以及注塑     

Manufacturing features recognition using backpropagation neural networks

A backpropagation neural network (BPN) is applied to the problem of feature recognition from a boundary representation (B-rep) solid model to facilitate process planning of manufactured products. It is based on the use of the face complexity code to represent the features and a neural network for the analysis of the recognition. The face complexity code is a measure of the face complexity of a feature based on the convexity or concavity of the surrounding geometry. The codes for various features are fed to the network for analysis. A backpropagation network is implemented for recognition of features and tested on published results to measure its performance. Any two or more features having significant differences in face complexity codes were used as exemplars for training the network. A new feature presented to the network is associated with one of the existing clusters, if they are similar, or the network creates a new cluster, if otherwise. Experimental results show that the network was consistent in recognizing features, hence is appropriate for application to the problem of feature recognition in automated manufacturing environment.     

Integrated manufacturing features and Design-for-manufacture guidelines for reducing product cost under CAD/CAM environment

The main contribution of the work is to develop an intelligent system for manufacturing features in the area of CAD/CAM. It brings the design and manufacturing phase together in design stage and provides an intelligent interface between design and manufacturing data by developing a library of features. The library is called manufacturing feature library which is linked with commercial CAD/CAM software package named Creo Elements/Pro by toolkit. Inside the library, manufacturing features are organised hierarchically. A systematic database system also have been developed and analysed for each feature consists of parameterised geometry, manufacturing information (including machine tool, cutting tools, cutting conditions, cutting fluids and recommended tolerances and surface finishing values, etc.), design limitations, functionality guidelines, and Design-for-manufacture guidelines. The approach has been applied in two case studies in which a rotational part (shaft) and a non-rotational part are designed through manufacturing features. Therefore, from manufacturing feature library a design can compose entirely in a bottom-up manner using manufacturable entities in the same way as they would be produced during the manufacturing phase. Upon insertion of a feature, the system ensures that no functionality or manufacturing guidelines are violated. The designers are warned if they attempt to include features that violate Design-for-manufacture and Design functionality guidelines. If a feature is modified, the system validates the feature by making sure that it remains consistent with its original functionality and Design-for-manufacture guidelines are re-applied. The system will be helped the process planner/manufacturing engineer by automatically creating work-piece data structure.     

A new approach for automated parts recognition using time series analysisand neural networks

This paper presents a new approach for automated parts recognition. It is based on the use of the signature and autocorrelation functions for feature extraction and a neural network for the analysis of recognition. The signature represents the shapes of boundaries detected in digitized binary images of the parts. The autocorrelation coefficients computed from the signature are invariant to transformations such as scaling, translation and rotation of the parts. These unique extracted features are fed to the neural network. A multilayer perceptron with two hidden layers, along with a backpropagation learning algorithm, is used as a pattern classifier. In addition, the position information of the part for a robot with a vision system is described to permit grasping and pick-up. Experimental results indicate that the proposed approach is appropriate for the accurate and fast recognition and inspection of parts in automated manufacturing systems.     

An intelligent grouping algorithm for cellular manufacturing

The methodology presented in this paper will provide a means of identifying part families/machine cells using design and manufacturing characteristics simultaneously. The technique used is a self-organizing neural network called interative-activation and competition (IAC) which acts as a content-addressable memory. This neural network is used to define a similarity index of the pairwise comparisons of parts based on a variety of design and manufacturing characteristics. A bond energy algorithm partitions the matrix of part similarity indices to create part families and inferred from the part families are machine cells. A brief example will be examined as well as discussion of the results.     

Engineering applications of neural networks

This paper describes several prototypical applications of neural network technology to engineering problems. The applications were developed by the authors as part of a graduate-level course taught at the University of Illinois at Urbana-Champaign by the first author (now at Carnegie Mellon University). The applications are: an adaptive controller for building thermal mass storage; an adaptive controller for a combine harvester; an interpretation system for non-destructive evaluation of masonry walls; a machining feature recognition system for use in process planning; an image classification system for classifying land coverage from satellite or high-altitude images; and a system for designing the pumping strategy for contaminated groundwater remediation. These applications are representative of many of the engineering problems for which neural networks are applicable: adaptive control, feature recognition, and design.     

Integrated optimal topology design and shape optimization using neural networks

In this paper, neural network- and feature-based approaches are introduced to overcome current shortcomings in the automated integration of topology design and shape optimization. The topology optimization results are reconstructed in terms of features, which consist of attributes required for automation and integration in subsequent applications. Features are defined as cost-efficient simple shapes for manufacturing. A neural network-based image-processing technique is presented to match the arbitrarily shaped holes inside the structure with predefined features. The effectiveness of the proposed approach in integrating topology design and shape optimization is demonstrated with several experimental examples.     

Exact Recognition of Compound Features by Feature Adjacency Matrix Elimination Algorithm

Aiming at the axiom of design for manufacture (DFM), this paper describes a recognition method for abstracting compound features from a part model and discloses the basic mechanism of compounding, also builds the corresponding 2D-simulation model. The inner association between feature neighboring and feature compounding is deeply discussed and, based on the essential transforming rule of two neighboring features, the corresponding feature adjacency matrix (FAM) of multi - feature entities are generated. For the manufacturing feature converted from the pure design feature; an innovative concept-homogenous compounding is presented to clarify the architecture of machining domain. Then, the FAM recurrence elimination algorithm is developed to determine all the compound features, and according to machining sequence, outputs a group of machining domains.     

A NEURAL NETWORK APPROACH FOR EARLY COST ESTIMATION OF PACKAGING PRODUCTS

Product costs need to be identified early, i.e., during the design stage, where they can be controlled best. This implies the need to estimate the product's cost without full knowledge of the manufacturing process plans. In this paper, a feature-based cost estimation using a back-propagation neural network is proposed and a prototype system has been developed for estimating the costs of packaging products based on design information only. All the cost-related features of a product design were extracted and quantified according to their cost effects. The correlation between these cost-related features and the final cost of the product was established by training a back-propagation neural network using historical cost data. The extraction of cost-related features and the construction, training and validation of the neural network are described. The performance of the trained neural network based on a set of testing samples is also given.     

Object-oriented modeling and simulation of neural networks

The field of neural networks is being investigated by many researchers in order to provide solutions to difficult problems in the area of manufacturing systems. Computer simulation of neural networks is an important part of this investigation. This paper applies concepts from an important trend in software engineering research, namely object-oriented programming, to model neural networks.The design and implementation of a software object library is crucial to obtaining the full benefits of object-oriented programming. In this paper we discuss the design and implementation of a foundation library of software objects for the purpose of simulating and validating different network architectures and learning rules. The library contains objects that implement various types of nodes and learning rules. We discuss the results of our experiments to illustrate the benefits of using an object-oriented approach to modeling neural networks.     

不同类型特征对神经网络分类结果的影响

目前一般的乳腺X光片计算机辅助诊断系统经常采用神经网络来进行分类。分类过程中特征的提取极为重要,因为这关系到神经网络的分类能力,需要采用最具代表性的特征作为分类系统的输入部分。文章采用神经网络对14个基本的灰度特征、4个BI-RADS特征、病人年龄特征进行训练和测试,研究这些特征对分类结果的影响。试验结果显示,联合使用14个灰度特征和4个BI-RADS特征可以改进肿块异常的分类率。     

Design of a feature-based order acceptance and scheduling module in an ERP system

Customer-oriented manufacturing competes on timely responses to customer requirements, and precise scheduling control for delivery. This challenge demands engineering design and production planning to be fully integrated via advanced enterprise resource planning (ERP) systems. This paper proposes a generic feature association method and a detailed framework that can unify product and process models in order to satisfy customer orders with small batch sizes and high variations. A conceptual solution is introduced by integrating two traditionally separate feature domains: design configuration features and manufacturing process features. To achieve the proposed method, a customer feature class is suggested for the characterization of customers’ profiles related to the manufacturer. An instantiated customer feature object functionally tracks each customer’s selection of product configurations related to its requirements, specific orders, and production schedules with dynamic associations to the live manufacturing capacity. With the new associative integration method, a preliminary order acceptance system (OAS) prototype system has been implemented within an ERP order management system and its conceptual structure model is demonstrated within a multi-facet feature framework.     

基于深度学习的主轴承盖分类识别算法

机械零件自动分类识别算法,在智能工业、自动化加工等领域具有广阔地应用前景。针对汽车 发动机主轴承盖零件自动分类时,存在特征多表面分布和光照敏感等难点问题,提出多分支特征融合卷积神经 网络(MFF-CNN)。MFF-CNN 具有 2 个子网络分支,分别提取主轴承盖 2 个表面的特征,经过特征融合,形成 最终的零件分类特征。在网络结构设计上,MFF-CNN 基于密集连接型卷积神经网络设计,通过增强网络层级 间的特征重用,有效降低模型的参数量,缓解较小样本量条件下,深层网络的过拟合和梯度消失问题。实验结 果表明,在实际采集的主轴承盖图像数据集上,MFF-CNN 的识别率为 91.6%,并对实际生产中的零件图像光 照不均匀问题,具有良好的鲁棒性。     

Introduction to assembly features: an illustrated synthesis methodology

The article is an introduction to a new concept of assembly features able to support intelligent design and manufacturing of complex products. An assembly feature is defined as a generic solution referring to two groups of parts that need to be related by a relationship so as to solve a design problem. The concept of assembly feature encompasses the notions of design intent, technical function, technological solution and manufacturing process as well as it provides a justification for the use of part features. After a general introduction and a justification of the interest of assembly features, some guidelines are provided to show how assembly features can be characterized in any domain of application concerned with engineering design of assemblies. As an illustration of the proposed methodology, the concept is finally applied to the engineering phase of aeroplane design.     

Integrated process planning using tool/process capabilities and heuristic search

CAD–CAM integration has involved either design with standard manufacturing features (feature-based design), or interpretation of a solid model based on a set of predetermined feature patterns (automatic feature recognition). Thus existing approaches are limited in application to predefined features, and also disregard the dynamic nature of the process and tool availability in the manufacturing shop floor. To overcome this problem, we develop a process oriented approach to design interpretation, and model the shape producing capabilities of the tools into tool classes. We then interpret the part by matching regions of it with the tool classes directly. In addition, there could be more than one way in which a part can be interpreted, and to obtain an optimal plan, it is necessary for an integrated computer aided process planning system to examine these alternatives. We develop a systematic search algorithm to generate the different interpretations, and a heuristic approach to sequence operations (set-ups/tools) for the features of the interpretations generated. The heuristic operation sequencing algorithm considers features and their manufacturing constraints (precedences) simultaneously, to optimally allocate set-ups and tools for the various features. The modules within the design interpretation and process planner are linked through an abstracted qualitative model of feature interactions. Such an abstract representation is convenient for geometric reasoning tasks associated with planning and design interpretation.     

Application of a neural network to improve an automated thermoplastic tow-placement process

This study demonstrates the use of an on-line neural network to calculate process set points for PID controllers in a manufacturing process such as the automated thermoplastic tow-placement (ATP) technique. The set points are computed by the neural network so that the throughput is near maximum and a desired minimum quality is maintained. A novel neural network predictive scheme is developed to enable performance over a wide range of processing inputs. Process history can greatly affect the final part quality and, therefore, is an integral part of the method for determining the set points. The system is first trained and tested in simulation and then validated for the highly non-linear ATP process resulting in significantly improved process operation. The developed approach is applicable to many other manufacturing processes where process simulations exist and conventional control techniques are lacking.     

一种卷积神经网络和极限学习机相结合的人脸识别方法

卷积神经网络是一种很好的特征提取器,但却不是最佳的分类器,而极限学习机能够很好地进行分类,却不能学习复杂的特征,根据这两者的优点和缺点,将它们结合起来,提出一种新的人脸识别方法。卷积神经网络提取人脸特征,极限学习机根据这些特征进行识别。本文还提出固定卷积神经网络的部分卷积核以减少训练参 数,从而提高识别精度的方法。在人脸库ORL和XM2VTS上进行测试的结果表明,本文的结合方法能有效提高人脸识别的识别率,而且固定部分卷积核的方式在训练样本少时具有优势。     

卷积神经网络特征重要性分析及增强特征选择模型

卷积神经网络等深度神经网络凭借着其强大的表达能力、突出的分类性能,已在不同领域内得到了广泛应用.当面对高维特征时,深度神经网络通常被认为具有较好的鲁棒性,能够隐含地对特征进行选择,但由于网络参数巨大,如果数据量达不到足够的规模,则会导致学习不充分,因而可能无法达到最优的特征选择.而神经网络的黑箱特性使得无法观测神经网络选择了哪些特征,也无法评估其特征选择的能力.为此,以卷积神经网络为例,首先研究如何显式地表达神经网络中的特征重要性,提出了基于感受野的特征贡献度分析方法;其次,将神经网络特征选择与传统特征评价方法进行对比分析发现,在非海量样本的情况下,传统特征评价方法对高重要性特征和噪声特征的识别能力反而能够超过神经网络.因此,进一步地提出了卷积神经网络增强特征选择模型,将传统特征评价方法对特征重要性的理解结合到神经网络的学习过程中,以辅助深度神经网络进行特征选择.在基于文本的社交媒体用户属性建模任务下进行了对比实验,结果验证了该模型的有效性.     

Design and manufacturing of sheet-metal parts: Using features to aid process planning and resolve manufacturability problems

This paper describes an integrated system for the design and production of sheet-metal parts. We have identified some important features for the sheet-metal bending process. These features are automatically generated as the design progresses. After the designs are complete, our automatic process planning system uses the features and generates new ones to aid the production of plans with near-minimum manufacturing costs. Finally, these plans are used to produce parts on an automatic bending system.Once a plan is generated, it can be used to manufacture the part, and to provide feedback to design and other factory systems. The application of features and the potential feature interaction problems are discussed. Several key manufacturing problems are also considered and the result of planning is used to resolve these problems. By solving these feature interaction problems and often practical manufacturing issues, we are able to plan and manufacture the majority of the parts we have tested less than one hour after the flat patterns are prepared.