Research And Development Of On-line Measurement And Monitoring System For Size Parameter Of Common Rail Needle Based On Machine Vision

Nozzle needle is one of the three major precision parts of the diesel fuel system.It is used in industrial diesel engines and light passenger and passenger diesel engines.It is a key component to ensure the stable operation of diesel engine fuel injection system.One of its high precision,yield,the corresponding measurement accuracy,detection efficiency and other requirements also will increase.How to realize the fast,stable,efficient,high precision and all-line measurement of the parameters of the joint size of the common rail needle parts is the urgent problem to be solved in industrial production.This dissertation focuses on the measurement and process control of the tail size parameters of the common rail needle parts of a foreign fuel injection nozzle.Aiming at the precision,efficiency,economy and integrity of the visual measurement system,the whole design of the system platform,the high precision processing of the image,the system calibration,the optoelectronic integrated control and the system development are studied.The main work of this dissertation is as follows:First of all,do the overall design of the visual on-line measurement and monitoring system.Including the optical vision system hardware selection matching and motion mechanism design.According to the importance of the light source in the visual system,the composite light is studied by combining the characteristics of the target part contour,the size of the test to be measured,the status of the enterprise production line,the lighting mode of the optical vision system,and the selection of the camera,lens and light source.And the effect of monochromatic light on the measurement system,it is concluded that monochromatic light is superior in backlighting mode.At the same time,the relationship between light intensity and contour pixel location is studied.It is concluded that the visual system is within a certain light intensity range,the effect is optimal;according to the characteristics of the parts to be measured,determine the feeding,positioning and sorting movement mechanism program.Secondly,the algorithm is designed to process the image acquisition system.In the aspect of noise reduction,the median filter operator is selected by filtering and contrasting.Based on this,the method of median filtering based on small area is introduced,which greatly shortens the filtering time.In the aspect of edge detection,The Sobel edge detection algorithm is used to improve the edge localization efficiency,and the small area edge feature point tracking method is used to locate the target feature points based on the learning and training logic.In the aspect of sub-pixel localization and fitting,Sexual regression linear fitting method,in a specific small area to complete high-precision contour fitting,and complete the diameter,angle,distance multi-parameter simultaneous calculation.Next,the theoretical research and design of visual system calibration are deeply carried out.This dissertation analyzes the parameters of the camera system,the distortion and correction of the camera system.The method is based on the Microscape software system,and the two-step calibration method is used to calibrate and calibrate the camera's internal and external parameters and lens.Grid and standard parts,the camera position and pixel equivalent calibration and calculation.Then,do a specific design and development of the visual measurement system software,hardware control.The image processing process is designed under Microscape visual system,and its secondary development in visual inspection,control,statistical calculation and human-computer interaction is completed based on VB.And the design of the process quality real-time monitoring system for the production line real-time online monitoring,statistics and analysis to provide a complete application;the use of CATIA software to the overall design of the automation body to complete the overall feeding,positioning clamping and cutting sorting agencies The design of the system workflow,automation equipment,communication and control,complete the entire system of logic and operation control.Finally,do the prototype development and stability evaluation,through the test,it runs in good condition.The Q-DAS statistical analysis software was introduced to evaluate the stability,bias capability index and repeatability of the prototype platform system.The statistical analysis showed that the diameter,angle and distance of the joint measurement,in the tolerance range to achieve more than 10 times the accuracy and the sub-pixel resolution of the system reaches 0.0008~0.0003 mm,to meet the practical needs of industrial applications;and Taking the diameter parameter as an example,the real-time monitoring system of the process quality of the platform is tested,and the feasibility and validity of the theoretical research results are verified.This dissertation completes the research and development of the on-line detection and monitoring system of the tail size parameter of the common rail needle parts,which has important reference function to the on-line detection of other similar products.