Papers by Author: Yan Ling Zhao

Abstract: In the process of designing aeronautic composite mold parts, previous designing knowledge and expertise cant be well used, the cycle of parts modeling is long and its production efficiency is low. According to the classification and characteristics of aeronautic mold parts and the knowledge engineering technology, based on the fusion module in UG6.0, we establish its repository by the knowledge acquisition, knowledge representation and knowledge reasoning of aeronautic mold parts. With the UG secondary development tools, UG/Open Menuscript and UG/Open Uistyler, we develop user menu and parametric design interface of aeronautic mold parts, make good use of its designing knowledge, and fulfill the knowledge-driven parts parametric design. They satisfy the rapid design requires of aeronautic mold parts and shorten its designing cycle.

Abstract: At present, the steel ball surface defects are usually detected by manual work, but it has low efficiency and low reliability. For the current situation, in this paper, we design the steel ball deployment mechanism based on image processing technology, establish the mathematical model of the shooting point trajectory and determine the amount and location of the shooting points by the steel ball motion analysis. By the simulation based on Mathematica and Java, verify the effectiveness of the steel ball deployment mechanism in steel ball unfolding and defects recognition. Thus, the steel ball surface can be completely detected.

Abstract: For features of large data, complex operation and high transport in the process of measurement surface roughness, In this paper, roughness measurement system based on TMS320DM642 (DM642) as its core was designed, then established architecture of system and function modules is described in details, on this basis using of regression analysis to calibrate the relationship between image measuring roughness and actual values. Finally, this paper develops DSP software development platform.

Abstract: Steel ball, as a rolling body of all kinds of bearings, it direct affects the bearings precision, dynamic performance and service life. This paper introduces the digital image technology Radial Basis Function (RBF)-Neural network, based on extracting the Steel Ball surface defect image features, used the strategy which is combined with static- dynamic clustering to union the two-stage study and design the hidden layer structure. Simulation and experiment show that the RBF-neural network runs stably, has fast convergence and overall accuracy rate of 96%. These can meet the needs of practical application.

Abstract: Aiming at the problem of image recognition in the process of defected chip generation of automatic machining, fuzzy category methods of RBF net and eight neighborhood Euler numbers are researched in this paper. They are based on fuzzy theory and neural net. The gradient steepest descent of optimization theory is used and aberration is minimized by step between required output and actual output. By modifying studying algorithm, recognized capacity is increased. This method is tested in Matlab platform. It can be concluded that fraction of chip image under sophisticated surrounding may be recognized accurately through this net.

Abstract: This paper presents the implementation of an internet-based facility for the drilling process simulation, which complies with TCP/IP and is supported by Browser/Server model. The process simulation is one of pivotal technologies and implemental techniques of the internet-based agility manufacturing technology. A network architecture model is developed with the WebServer toolbox of MatLab. The complicated drilling process is studied. Semi-experimental mathematic models of drilling force and torque are established. Users input different machining parameters in client-side and then they will get the simulation curves of cutting forces and torques and can evaluate the simulation results. At last, optimum machining parameters can be chosen by users via internet.