YU Shengrui 1,2 , CAO Ligang 1 , FENG Hao 1
(1. School of Mechanical and Electronic Engineering, Jingdezhen Ceramic Institute, Jingdezhen Jiangxi 333403 China; 2. State Key Lab of Material Processing and Die & Mold Technology, Huazhong University of Science and Technology, Wuhan Hubei 430074, China)
Abstract: The model for the deposition rate of glaze lays an important basis for automatic glazing of ceramics. For the automatic track-planning of the offline programming robot working on ceramic production line to ensure the accuracy and uniformity of glaze thickness, the software module fitting the glaze deposition model is developed with MFC. The fitting scheme is proposed based on neural network. Model fitting and analysis are realized by hybrid programming with VC and MATLAB based on COM. DLL file is developed to produce word report of the model. The software module is tested by experiments. The result shows that the error of the thickness model is within 5μm. The model built by the software module is correct and effectual. The software's UI is friendly and satisfactory for project practice. The method increases the control over the accuracy of glazing thickness. The paper provides a specific theoretical and methodological support for the realization of process planning and simulation system in ceramic glazing. It will make the future developed system meet the actual processing requirement.
Keywords: robot, glaze deposition model, glazing, neural network, hybrid programming