Research Of The Machine Vision Inspection System On The Engine Roller Rocker Arm

In the contemporary society of industrial intelligence and high-tech, car as a means of transport is more and more obvious importance. At present in our country, automobile manufacturers pay more and more attention to production quality, the production cost, the production efficiency, the relevant departments also introduced all kinds of favorable measures, to promote the transformation and upgrading of the automotive industry, to the high efficiency and high quality of development goals. Under the background of these policies, the development of technology innovation, on the engine assembly quality requirements are also getting higher and higher, the research and development of engine rocker roller machine visual detection system is to improve the quality and efficiency of production, arises at the historic moment.Engine roller rocker arm is an important part of the automobile engine, which belongs to the engine valve mechanism, the role is the engine in the force and the displacement of the cam shaft transmission to transfer to the valve, the complete open or closed valve task. Because it is very important, the assembly quality must be tested. The traditional method of testing the quality of the roller rocker arm assembly is that the human eye recognition is qualified, prone to miscarriage of justice. Therefore, it is necessary to develop a set of machine vision system to automatically identify and detect the position of the roller rocker arm, whether the deviation, leakage and not fully in place.The main research contents of this paper are as follows:1, according to the design objectives and tasks of hardware selection, including light source, industrial cameras, with the use of industrial cameras lens, image acquisition card, industrial PC, etc.;2, choose suitable for machine vision software, to establish a proper algorithm, used to process the collected images and data, make the compatible device software and hardware supporting, and good detection of motivation and roller rocker in the assembly process is partial, whether the lack of and is not yet fully in place, and judge whether it meets the requirements of the quality of the assembly;3, on line testing, the machine vision detection system in the algorithm and the optimization of the parameters;4, implementation of the overall visual inspection system and PLC as well as PC communication. To ensure that the system camera can get the trigger signal to the PLC and the detection of the completion of the OK/NG signal can be transmitted to the PLC, to ensure that the engine roller rocker detection data can be transmitted to the PC without error.The novelty of this paper is using the machine vision technology, in engine roller rocker image collection and analysis, the roller arm put partial, missing or not yet fully in place of recognition, testing data and the detection of image processing, research and innovation. Now the system has been successfully applied in the FAW Volkswagen ea211 engine assembly line, system detection accuracy can reach more than 98%, processing time is 2 seconds / platform and improve the efficiency of engine roller rocker arm assembly quality detection, ensures that the engine and the assembly quality of the product.