Plastic thermal riveted assembly monitoring method and device

A monitoring device and hot riveting technology, which is applied in the direction of measuring devices, optical device exploration, and material analysis, can solve problems such as high surface quality requirements, product welding defect assembly quality, unclear edges between targets and backgrounds, etc., to improve production efficiency and quality effects

Active Publication Date: 2018-02-27

JIANGSU RUNMO AUTOMOBILE TESTING EQUIP

8 Cites 1 Cited by

AI-Extracted Technical Summary

Problems solved by technology

However, for interior and exterior trim parts such as automobile door panels and instrument panels, the surface quality requirements are high and the structure is complex. During the assembly process, defects are easily formed due to welding process parameter control and positioning errors. Especially in China, manual welding is still mainly used. Assembly-based, resulting in common welding defects or poor assembly quality in final products

In view of the problems referred to above, patent CN106078153A (electric car door structure and its assembly device and assembly and detection method) proposes a car door design and a detection method corresponding to the design form of the car door, but does not involve optical detection technology; patent CN103245263A (a car door Door guard ...

Abstract

The invention relates to the technical field of automobile part detection, in particular to a plastic thermal riveted assembly monitoring method and device. The method comprises the steps that a to-be-detected workpiece is conveyed to a uniform optical detection view field position by using an auxiliary conveying and positioning module; a powerful alignment light source module shines obliquely togenerate a shadow region, and image collection is conducted by using a camera module; it is judged whether the to-be-detected workpiece is qualified or not through image analysis, and quality evaluation is conducted on the welding conditions through an SVM classifier. By means of the plastic thermal riveted assembly monitoring method, it can be achieved that missing installation, defect identification and quality evaluation of finished products and semi-finished products of automobile doors and other parts can be quickly monitored at a high precision on line, and the production efficiency andquality of an automatic assembly line are improved.

Application Domain

Optical detectionOptically investigating flaws/contamination

Technology Topic

Image analysisWork in process +6

Image

Examples

Experimental program(1)

Example Embodiment

[0018] The present invention will be further described below in conjunction with the accompanying drawings and implementation examples.

[0019] The plastic hot riveting assembly device includes a strong collimated light source module, a camera module, a system control module, and an auxiliary transportation and positioning module. It is used to form a specific shadow area for oblique riveting posts. In specific applications, multiple sets of strong collimated light source modules can be set according to needs. Different sets of strong collimated light source modules are arranged symmetrically in a ring; Appropriate object distance to photograph the welding column, where the object distance is determined by the selected lens and shooting resolution requirements; the system control module is used to control the light intensity and switch of the strong collimated light source, image acquisition cycle, auxiliary transport cycle, image analysis and Riveting quality judgment.

[0020] Schematic diagram of the optical path layout of the plastic thermal riveting assembly device figure 1 As shown in the figure, 1 in the figure is a camera module, 2 is a strong collimator lens, and 3 is a welding post. This embodiment includes two sets of strong collimated light source modules. The welding post is composed of riveting piles and riveting posts. The strong collimated light source obliquely shoots to the welding post of the component to be tested, and the camera takes pictures from the top.

[0021] figure 2 It is a flow chart of plastic hot riveting assembly inspection, including the following steps:

[0022] 1) Initialize the grayscale threshold and relative threshold, where the grayscale threshold and relative threshold can be appropriately adjusted according to the brightness of the collimated light source under working conditions. In this embodiment, the grayscale threshold is 250 and the relative threshold is 64. Go to step 2) ;

[0023] 2) Train the SVM (Support Vector Machine) classifier and classify it, that is, score the classification quality level of the training samples according to the brightness statistics, mean square error, and contrast characteristics of the welding local image under the corresponding illumination, and train the SVM classifier respectively , obtain a plurality of SVM classifiers corresponding to the feature; wherein, training the support vector machine SVM includes the following steps:

[0024] 2.1) Initialize the parameters of the particle swarm. In this embodiment, the inertia weight parameter is set to w=1; the learning factors are all c1=c2=2;

[0025] 2.2) Determine the fitness function, establish the learning model of the support vector machine through the initial penalty coefficient and kernel function parameters corresponding to the individual particles, and introduce training samples to calculate the fitness value corresponding to the individual particles;

[0026] 2.3) Update the extremum, speed and position, that is, when the fitness value of the individual is smaller than the individual extremum or the global extremum, the individual fitness value will replace the individual extremum or the global extremum, and the newly generated extremum will be speed update;

[0027] 2.4) Evolutionary search, each particle re-searches according to the updated speed and position, and iteratively repeats step 2.3) until the exit iteration condition is met (the total number of iterations or the iteration update amount is sufficiently small as the iteration exit condition);

[0028] 2.5) The iteration is completed, and the optimal parameters c and g are assigned to the support vector machine classifier;

[0029] 3) Locate the piece to be tested, and the auxiliary transportation and positioning module will transport the piece to be tested to a unified optical detection field of view position, then go to step 4);

[0030] 4) Image acquisition, under the control of the system control module, collect images under different strong collimated light sources respectively through the camera module, and turn to step 5); in this embodiment, turn on the left light source, turn off the right light source, and shoot figure 1; Turn on the right light source, turn off the left light source, and shoot figure 2;

[0031] 5) Judging whether the metal screw is missing, that is, judging the pixel value in the area corresponding to the metal screw in the collected image. If the gray value of the pixel in the area corresponding to the metal screw is less than the gray threshold, it is determined that the metal screw is missing, that is, it is unqualified , go to step 7), otherwise go to step 6);

[0032] 6) Subtract the corresponding pixel values of the two images collected and take the absolute value. If the absolute value exceeds the relative threshold, it is judged as missing welding, that is, unqualified, and go to step 7), otherwise go to step 8);

[0033] 7) Use SVM classifier to evaluate the quality of welding conditions;

[0034] 8) The parts to be tested are qualified.

PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

Classification and recommendation of technical efficacy words

Improve productivity and quality

Plastic thermal riveted assembly monitoring method and device

AI-Extracted Technical Summary

Problems solved by technology

Abstract

Image

Examples

Example Embodiment

PUM

Description & Claims & Application Information

Similar technology patents

High-efficiency production line for tempered glass pot lids

Defect detection method for product of mirror surface bowl-shaped structure

Fastening assistive device for screws in semi-closed space of small-sized table body

Method and device for producing agglomerated stone with controllable patterns

Pouring system and injection mold with same

Classification and recommendation of technical efficacy words

Severe plastic deformation technology of magnesium alloy board

Double beam coupling device

An automatic operation system of inner leading bonding device of semiconductor device

Method for measuring thickness of rigid transparent material with laser

Method for optimally adjusting console parameters in semiconductor technology