| Global competition has dramatically increased the need for manufacturing enterprises to produce high-quality competitively priced products both quickly and efficiently. With the development of computer networks and information technology, the approach to product development is evolving rapidly. It is urgent for small and medium manufacturing enterprises in China how to use effectively the distributed data, knowledge resources, and how to create a distributed design and manufacturing environment that enables integrated design, analysis, decision and manufacturing process. One of the chiefly technologies is to realize product information reconstruction from old drawings and modeling, and find the effective methods for managing and sharing of information, especially those related to design and manufacturing, throughout the E-commerce and Internet/Intranet.The distributed product data management (PDM) can provide support for product information management and information integration. The product design information can be reconstructed through engineering drawings interpretation and reconstruction. Though commercial PDM software has successfully been applied to many manufacturing enterprises to manage whole product design processes and all data, it is not available for small and medium manufacturing enterprises because high requirements of application technique, high price and high updating cost. Therefore, based on analyzing the features and requirements of small and medium manufacturing enterprises, the principle of distributed PDM and implementing key technologies are researched in this paper, including the integrated method of enterprise's information, some new algorithms of reconstructing product information mode and a new architecture based CORBA standards for engineering drawing management system developed by author. The objective of this paper is to create a engineering drawings management and remote design environment to ensure the right information to the right person at the right time and in the right format anywhere in small and medium manufacturing enterprise. The main contents and conclusions are as follows:1) A knowledgeldirected interpretation of mechanical engineering drawings is presented. Furthermore, this paper discusses some implementing methods of knowledge acquisition, knowledge base management and inference machine in the computer networks environment.2) A new class adaptive median filter algorithm is presented. The most important feature is that the shape of the filter window may be changed, according to the mean square error of the child window. A standard testing image and an experimental result are presented. The result shows that this algorithm is superior to standard median filter in preserving the details and removing the noise.The dissertation is supported by key project of science and technology of Sichuan province3) Image enhancement using fuzzy sets is superior in preserving the details and removing the noise. But it is the key to select an optimum threshold Xc. This paper presents a new adaptive image enhancement using fuzzy sets, which can select an optimum threshold automatically. At last an experimental result is presented.4) An experimental system of image analysis is established based on Microsoft Foundation Class (MFC). The algorithms in the dissertation are tested and verified, also an edit system for raster graph and vector graph is developed by author.5) A simple modified algorithm for thinning scanned engineering drawings is presented, which is needed for removing the burr based on mathematical morphology. The result shows that this strategy is correct and effective.6) According to analyzing the features of learning algorithm of backpropagation neural networks, a method of classifier is proposed, which is applied for character recognition based on self-adaptive training rate neural networks. And also, the experimental result shows that this classifier is superior in self-study, self-adaptive recognition with noise and can be used to the recogn...
|
PDF Full Text Request