Development Of New Intelligent Riveting Equipment And Quality Monitoring System For Frieight Wagons Bogies

Along with the continuous development of China's railway rolling stocks industry,enterprises have more and more urgent needs for On-line monitoring process quality and establishment of quality tracing system.Although computer aided assembly(CAA)has been applied in many high-end products such as EMU,Assembly quality of pull rivets depended on the technologist's experience,technology and skill.The method of judging the quality of riveting by visual measurement has some faultinesses like low efficiency,big cost and high labor hardness.Based on the above situation,a solution for on-line monitoring of riveting quality and off-line analysis and diagnosis system is proposed in this paper.Based on image recognition and signal measurement techniques,multisource data was collected,analyzed,and stored.The riveting system can not only determine the quality of riveting and recognize the position of riveting,but also realise information management of process data and real time early-warning.In this paper,the hardware structure and software development of the new intelligent riveting equipment and quality monitoring system are studied,and the effectiveness of the system is verified through field Riveting test.Firstly,according to the structural characteristics and assembly process of the short tail rivet pin,the problems that need to be solved in the system and the application status of riveting equipment at home and abroad are analyzed.Based on the detection of rivet quality at home and abroad,a computer-aided inspection system based on image recognition and sensor technology is designed to achieve the automation,real-time and quality traceability of rivet quality detection.After analyzing the assembly process of the type K6 bogie brake device pull rivet assembly,the system architecture of intelligent riveting system is constructed.Then on the base of these,this paper completed the mechanical structure and electrical control system,including the layout of the system,the design of electrical schematic diagram,the selection of sensors and so on.Based on the LabVIEW programming platform,the overall function level diagram of the system software is designed.The system software was developed with the idea of modularization,including automatic control system,data acquisition and riveting quality determine system,image acquisition system and other functions of the system established,and provided design interface of the system software.As the core part of the system software quality traceability system,4 kinds of image recognition method based on riveting position determination were proposed including contour extraction based on Hough transform,image fusion and image recognition method based on Harris/Sift/Suft feature extraction,image recognition method based on Histogram features and BP neural network and Identification algorithm of bar code / two-dimensional code based on open source tool Zbar.Finally,after verifying the effectiveness of each module of the riveting quality identification including riveting quality determination module,image recognition module,report generation module,etc,and the trial operation of the equipment and system in the factory,The feasibility of the scheme was proved.