机械产品概念设计中的知识表示

研究了计算机辅助概念设计中的关键技术——知识表示．将行为域引入公理化设计作为功能域和载体域之间的转换桥梁，针对该域结构定义扩展“之”字映射，建立了不同层次功能、行为和载体之间的映射关系；构造了概念设计方案的知识表示模型——域结构模板；针对一类产品的知识表示模型——概念空间，给出了基于概念空间的概念设计方案生成过程．最后以工业平缝机为例对所述模型和方法加以说明。     

基于知识模板的产品建模方法

为实现产品概念设计的创新,提出功能模板、行为模板和结构模板的概念,并利用三者之间的映射关系实现行为流产品建模.通过行为流建模的具体算法,在实现产品概念模型设计的同时生成该模型的行为路径.发动机模型的创建实例表明,该算法能实现产品概念模型设计的再创新.     

CAD中的功构映射研究

在回顾从线框模型、实体模型到特征模型发展历程的基础上指出，ＣＡＤ系统要高效率地发挥作用，必须能支持产品的概念设计、初步设计。据此，本文首先应用面向对象的思想，以模具设计为例，建立了产品面向结构的功能模型，然后以此功能模型为基础，讨论了功能与特征、功能与零件间的映射关系，并提出了描述结构设计初期产品各组成成份之间相互关系的形式化方法－－位置ＣＳＧ树。以上述分析为基础，利用现有的特征造型技术和实体造型     

产品原理结构设计模型

原理方案设计是产品功能和结构之间一个重要环节，通过对产品工作原理和零部件功能面的作用分析，提出了原理构件和功能联结的概念、定义和特点，并在此基础上用面向对象的方法建立了产品原理结构模型，为功能到结构的映射提供了一种方法。     

公理化设计与DFA集成的产品信息模型

在公理化设计理论的基础上，讨论了产品设计过程中功能要求的分解以及从功能域向结构域的曲折映射过程；建立了产品功能一结构层次模型和面向装配设计的产品信息模型．该产品信息模型不仅能够描述产品的层次结构信息，而且能够描述零部件之间的装配关系信息，能够实现CAD／CAE／CAPP/CAM的信息集成，为产品并行开发过程的信息集成打下良好的基础．以发动机减速器为例，对产品的功能树、结构树以及装配模型和装配关联矩阵的形成进行了说明。     

基于空间探索的创造性设计方法的研究

文中运用功能－行为－结构框架描述了基于状态空间扩展的探索性创造设计模型。在结构、行为或功能的相似基础上进行基于类比的设计，可以使新的结构、行为或功能从源设计引入至目标设计。     

产品功能结构建模研究

通过对产品功能定义的分析，说明功能、行为、状态、环境是完整描述产品功能的必要信息，阐明了功能-行为-状态-环境的关系，提出了功能结构的两级建模过程：功能任务分解和行为过程生成，在此基础上建立了产品功能结构的FBSE模型，并给出了功能结构设计的流程。     

体数据中物体之间结构关系的描述方法

体数据中不同的物体之间存在不同的结构关系,如隐藏关系、拓扑关系等,不同结构关系是影响用户理解体数据的重要特征之一。从分析体数据中物体之间关系入手,提出一个有效的体绘制应当定义一个从数据到绘制的映射,以便将数据中物体之间的每种关系映射到绘制空间。然后提出描述体数据中物体之间典型结构关系的体绘制方法。在绘制空间对物体之间的不同关系的描述增强了用户对体数据的理解,并拓展了体绘制在体数据分析中的功能。     

面向结构化布局设计的产品布局多色模型

为实现产品原理方案向结构化布局映射,提出产品布局元层次关系网及产品布局多色模型.首先在布局模块及其属性关系网的基础上构建布局元层次关系网;然后应用多色集合的个人颜色、统一颜色,给出产品布局元层次关系网的数学描述,形成产品布局多色模型,并提出推理算法;再利用个人颜色、统一颜色之间的围道布尔矩阵推理原理,进行产品结构布局设计过程中功能—运动分配—布局模块之间映射过程的形式化描述.应用该模型进行结构化布局设计,可有效地支持产品结构布局设计的公理化;最后在Pro/E平台中利用VC.net 2003工具进行二次开发,并以数控机床结构布局设计为例验证了文中模型和设计方法的可行性和有效性.     

特征信息建模及特征映射技术的研究

本文介绍特征技术产生的背景和特征建模的有关概念与方法。根据集成化制造系统的要求，提出了一套新的特征信息描述方法和系统结构；建立了折弯零件的设计特征库，实现了析弯零件的基于特征设计．分析、归纳了特征表达空间之间的关系和基于特征空间的特征映射原理与方法，完成了折弯零件设计特征模型到制造特征模型的转换，从而实现了产品信息的集成与共享．     

Neural network methods for one-to-many multi-valued mapping problems

An investigation of the applicability of neural network-based methods in predicting the values of multiple parameters, given the value of a single parameter within a particular problem domain is presented. In this context, the input parameter may be an important source of variation that is related with a complex mapping function to the remaining sources of variation within a multivariate distribution. The definition of the relationship between the variables of a multivariate distribution and a single source of variation allows the estimation of the values of multiple variables given the value of the single variable, addressing in that way an ill-conditioned one-to-many mapping problem. As part of our investigation, two problem domains are considered: predicting the values of individual stock shares, given the value of the general index, and predicting the grades received by high school pupils, given the grade for a single course or the average grade. With our work, the performance of standard neural network-based methods and in particular multilayer perceptrons (MLPs), radial basis functions (RBFs), mixture density networks (MDNs) and a latent variable method, the general topographic mapping (GTM), is compared. According to the results, MLPs and RBFs outperform MDNs and the GTM for these one-to-many mapping problems.     

A Virtual Head Driven by Music Expressivity

In this paper, we present a system that visualizes the expressive quality of a music performance using a virtual head. We provide a mapping through several parameter spaces: on the input side, we have elaborated a mapping between values of acoustic cues and emotion as well as expressivity parameters; on the output side, we propose a mapping between these parameters and the behaviors of the virtual head. This mapping ensures a coherency between the acoustic source and the animation of the virtual head. After presenting some background information on behavior expressivity of humans, we introduce our model of expressivity. We explain how we have elaborated the mapping between the acoustic and the behavior cues. Then, we describe the implementation of a working system that controls the behavior of a human-like head that varies depending on the emotional and acoustic characteristics of the musical execution. Finally, we present the tests we conducted to validate our mapping between the emotive content of the music performance and the expressivity parameters.     

Bernoulli Embedding Model and Its Application in Texture Mapping

A novel texture mapping technique is proposed based on nonlinear dimension reduction, called Bernoulli logistic embedding （BLE）. Our probabilistic embedding model builds texture mapping with minimal shearing effects. A log-likelihood function, related to the Bregman distance, is used to measure the similarity between two related matrices defined over the spaces before and after embedding. Low-dimensional embeddings can then be obtained through minimizing this function by a fast block relaxation algorithm. To achieve better quality of texture mapping, the embedded results are adopted as initial values for mapping enhancement by stretch-minimizing. Our method can be applied to both complex mesh surfaces and dense point clouds.     

Canonical locality preserving Latent Variable Model for discriminative pose inference

Discriminative approaches for human pose estimation model the functional mapping, or conditional distribution, between image features and 3D poses. Learning such multi-modal models in high dimensional spaces, however, is challenging with limited training data; often resulting in over-fitting and poor generalization. To address these issues Latent Variable Models (LVMs) have been introduced. Shared LVMs learn a low dimensional representation of common causes that give rise to both the image features and the 3D pose. Discovering the shared manifold structure can, in itself, however, be challenging. In addition, shared LVM models are often non-parametric, requiring the model representation to be a function of the training set size. We present a parametric framework that addresses these shortcomings. In particular, we jointly learn latent spaces for both image features and 3D poses by maximizing the non-linear dependencies in the projected latent space, while preserving local structure in the original space; we then learn a multi-modal conditional density between these two low-dimensional spaces in the form of Gaussian Mixture Regression. With this model we can address the issue of over-fitting and generalization, since the data is denser in the learned latent space, as well as avoid the need for learning a shared manifold for the data. We quantitatively compare the performance of the proposed method to several state-of-the-art alternatives, and show that our method gives a competitive performance.     

Visualization maps based on SOM to analyze MIMO systems

Knowledge extraction from large amounts of data is an effective approach for analysis and monitoring of industrial processes. The self-organizing map (SOM) is a useful method for this purpose, because it is able to discover low-dimensional structures on high-dimensional spaces and produce a mapping on an ordered low-dimensional space that can be visualized and preserves the most important relationships. With the aim to extract knowledge about the dynamics of industrial processes, we define 2D SOM maps that represent dynamic features which are useful for usual tasks in control engineering such as the analysis of the time response, the coupling among variables, or the difficulties in control of MIMO (multiple-input and multiple-output) systems. Those new maps make it possible to discover, increase or confirm knowledge about the system, spanned through the entire operation range. A well-known quadruple-tank MIMO system was used to test the usefulness of these maps. First, we perform an analysis of the theoretical dynamic behaviors obtained from the physical equations of the system. After that, we carry out an analysis of experimental data from an industrial pilot plant.     

Automated clustering to support the reflexion method

A significant aspect in applying the Reflexion Method is the mapping of components found in the source code onto the conceptual components defined in the hypothesized architecture. To date, this mapping is established manually, which requires a lot of work for large software systems. In this paper, we present a new approach, in which clustering techniques are applied to support the user in the mapping activity. The result is a semi-automated mapping technique that accommodates the automatic clustering of the source model with the user’s hypothesized knowledge about the system’s architecture.This paper describes three case studies in which the semi-automated mapping technique, called HuGMe, has been applied successfully to extend a partial map of real-world software applications. In addition, the results of another case study from an earlier publication are summarized, which lead to comparable results. We evaluated the extended versions of two automatic software clustering techniques, namely, MQAttract and CountAttract, with oracle mappings. We closely study the influence of the degree of completeness of the existing mapping and other controlling variables of the technique to make reliable suggestions.Both clustering techniques were able to achieve a mapping quality where more than 90% of the automatic mapping decisions turned out to be correct. Moreover, the experiments indicate that the attraction function (CountAttract′) based on local coupling and cohesion is more suitable for semi-automated mapping than the approach MQAttract′ based on a global assessment of coupling and cohesion.     

A chaotic serpentine mixer efficient in the creeping flow regime: from design concept to optimization

Motivated by the three-dimensional serpentine channel (Liu et al. in J Microelectromech Syst 9:190–197, 2000), we introduce a chaotic serpentine mixer (CSM) and demonstrate a systematic way of utilizing a mapping method to find out an optimal set of design variables for the CSM. One periodic unit of the mixer has been designed to create two streamlines portraits crossing each other. As a preliminary study, flow characteristics and mixing in the original serpentine channel has been reinvestigated. The working principle of the CSM is demonstrated via a particle-tracking method. From the design principle and the flow characteristics of the CSM, we choose three key design variables with an influence on mixing. Then, simulations for all possible combinations of the variables are carried out. At proper combinations of the variables, almost global chaotic mixing is observed in the Stokes flow regime. The design windows obtained can be used to determine an optimal set of the variables to fit with a specific application.     

Qualitative analysis of goat and sheep production data using self-organizing maps

Abstract: The aim of this study was to analyse the relationship between different small ruminant livestock production systems with different levels of specialization. The analysis is carried out by using the self-organizing map. This tool allows high-dimensional input spaces to be mapped into much lower-dimensional spaces, thus making it much more straightforward to understand any set of data. These representations enable the visual extraction of qualitative relationships among variables (visual data mining), converting the data to maps. The data used in this study were obtained from surveys completed by farmers who are principally dedicated to goat and sheep production. With the self-organizing map we found a relationship between qualitative and quantitative variables showing that more specialized farms have greater milk incomes per goat, highlighting farms that have a greater number of animals, better facilities (including milking machines) or animals fed with elaborated diets. The use of self-organizing maps for the analysis of this kind of data has proven to be highly valuable in extracting qualitative conclusions and in guiding improvements in farm performance.     

Simple mappings,expressive movement: a qualitative investigation into the end-user mapping design of experienced mid-air musicians

In a New Interface for Musical Expression (NIME), the design of the relationship between a musician’s actions and the instrument’s sound response is critical in creating instruments that facilitate expressive music performance. A growing body of NIMEs expose this design task to the end performer themselves, leading to the possibility of new insights into NIME mapping design: what can be learned from the mapping design strategies of practicing musicians? This research contributes a qualitative study of four highly experienced users of an end-user mapping instrument to examine their mapping practice. The study reveals that the musicians focus on designing simple, robust mappings that minimize errors, embellishing these control gestures with theatrical ancillary gestures that express metaphors. However, musical expression is hindered by the unintentional triggering of musical events. From these findings, a series of heuristics are presented that can be applied in the future development of NIMEs.     

二端面弹性转轴非线性动力学方程拓扑反变解

在非相对性原理下建立的二端面弹性转轴的非线性动力学方程,在工程应用中具有时变、强耦合和高度非线性等动力学特征.基于拓扑反变算法,建立二端面弹性转轴的原始系统空间、目标系统空间以及两空间的拓扑反变映射,得到其简单的解耦的目标系统空间及相应方程的拓扑反变解.