What Is a Weldment?

There are four types of welded joints: butt joints, T-joints, corner joints and lap joints. The size of the seam of the joint of the welding workpiece is determined by the joint form, thickness and bevel form of the weldment. The welds operated by electricians are usually angle steel and flat steel. Generally, the groove is not opened, and the size of the joint is 0 ~ 2mm.

With the acceleration of the modern industrial process, the conventional inspection methods of weldment tables have been unable to meet the requirements of industrial production. Therefore, the application of automation technology to the inspection of weldment tables has become an important development direction in the field of welding inspection. The design scheme of welding seam size and defect assessment through automatic acquisition of the surface image of the weldment by the CCD camera to the calculation of the weldment table is systematically explained, and the corresponding control system is designed and developed. At the same time, the system can be used to process the image of the weldment table collected. The experimental results show that the proposed design scheme of the weldment appearance inspection system is feasible. The system can realize automatic measurement of weldment table and weld seam size and evaluation of table and defect assessment, which has high practical application value.

Welding has developed into an important processing method in the manufacturing industry and is widely used in aviation, aerospace, metallurgy, petroleum, automobile manufacturing, and national defense. In welding products, the quality of the weld seam directly affects the life of the product. Therefore, in the production process, the weld size must be controlled strictly in accordance with the design requirements, and the occurrence of various defects must be strictly controlled. When measuring the weld surface size and assessing the surface weld defect, the visual inspection method is one of the most commonly used methods in industrial production inspection due to its flexibility and easy operation. At present, when measuring the size of a weld by visual inspection, tools such as a magnifying glass, ruler, undercut measurer are usually used for measurement; defect assessment requires strong professional knowledge and rich work experience of the rating staff. At the same time, during the visual inspection process, workers are susceptible to factors such as heavy workload, poor working environment, and differences in knowledge and cognition, resulting in a decrease in the accuracy of the results. Therefore, it is difficult to ensure that the results of the visual inspection method are standardized, objective, and scientific. Therefore, how to reduce the influence of the above factors on the final result has become one of the hot topics for welding workers in recent years. In recent years, with the advancement of technologies such as computers, white animation, and pattern recognition, it has also brought new development momentum to the appearance inspection of welds ^[1]

There are four types of welded joints: butt joints, T-joints, corner joints and lap joints. The size of the seam of the joint of the welding workpiece is determined by the joint form, thickness and bevel form of the weldment. The welds operated by electricians are usually angle steel and flat steel. Generally, the groove is not opened, and the size of the joint is 0 ~ 2mm.

There are four welding methods: flat welding, vertical welding, horizontal welding and overhead welding. Different welding methods should be selected according to the structure, shape, volume and location of the welding workpiece.

During flat welding, the welding seam is in a horizontal position, and the operation technology is easy to master. The diameter of the welding rod can be larger and the production efficiency is high, but the phenomenon of slag and iron moisture is easy to appear. Welding methods used for welding are all straight. When welding two sides, welding the front welding seam, the moving speed of the welding rod is slower to obtain a greater depth and width. When welding the welding on the opposite side, the moving speed is Go faster and make the weld width smaller.

During vertical welding and horizontal welding, due to the weight of the molten metal flowing down, defects such as incomplete penetration and welding knobs are generated. Therefore, smaller diameter electrodes and shorter arc welding are required. Welding current is 12% smaller than in flat welding.

The overhead welding operation is difficult, and a smaller diameter electrode should be used for welding. The shortest arc welding is used. ^[2]

According to the characteristics of weldment appearance size measurement and defect distribution, the weldment appearance inspection system mainly implements the following functions:

1) Appearance collection realizes multi-angle white animation collection. In order to realize the calculation of the appearance of the weldment, the full-scale image acquisition of the entire weldment is required, and then the corresponding parameter calculation and defect rating are performed according to the characteristics of different images;

2) Feature processing of collected images. Image processing is one of the important steps to achieve white motion detection. Only by processing the collected images accordingly can we obtain the required parameters, and then obtain the relevant component appearance parameters and defect categories and levels;

3) Management and inquiry of weldment information. In order to realize the user's query of related component information, the corresponding data real-time storage query function needs to be developed in the system. And should make the system support fuzzy query function to improve the query efficiency of the system. At the same time, the collected images should be classified and saved with the corresponding information;

4) System management and maintenance. In order to ensure the security and reliability of the stored information, it is necessary to set different permissions for different users, and provide corresponding data backup and restore functions, so as to prevent user's misoperation leading to information loss ^[3]

The weld appearance inspection system needs to obtain a digital image of the weldment surface, so the image acquisition of the system needs to be based on hardware. At the same time, in order to meet the needs of production mobilization, the system's mechanism should also be able to achieve mobilization and intelligent acquisition of images.

In order to enable the mechanism to perform an all-round detection of the size and surface condition of the weld on the surface of the weldment, a camera needs to collect images of the weldment from at least two perspectives. Therefore, the mechanism needs to achieve 90 ° rotation of the front and back of the white movement in order to collect the front and side images of the weldment.

The mechanism's white movement forward and reverse function will be controlled by PLC. The user can set the rotation cycle in the interactive interface of the system according to the needs, then the system will pass the parameters set by the user to the control address in the PLC control module, and adjust the PLC power cycle to achieve the mechanism's 90 ° forward and reverse rotation, and then the required image acquisition of the component by CCD camera. At the same time, the system can also control the mechanism in real time through the control buttons of the user interaction interface to meet the needs of users' real-time monitoring.

Weldment appearance inspection system usually includes the following modules: image acquisition control module, image processing module, parameter calculation and defect assessment module, information management module, system maintenance module ^[1]

Module divided image acquisition control module

In order to realize the calculation of the weld size on the surface of the weldment and the rating of the defects, the digital acquisition of the surface condition of the weld is needed first. Therefore, a function to control the image acquisition method must be designed in this system in order to achieve the system's white-motion detection function. Module In this module, two image acquisition methods are designed: single-frame acquisition and continuous acquisition.

The purpose of single sheet collection is to realize the calculation of the relevant dimensions of the weldment after welding and verify whether the relevant parameters meet the design requirements. The purpose of continuous collection is to monitor the surface quality of the weldment after welding. Once a defect is found on the surface, the system will alarm. And switch to single-sheet acquisition mode, collect the surface condition of the weldment in this segment, and evaluate the surface defects at the same time.

This module can also realize the remote control of the camera's rotation function. The user can control the positive and negative 90 of the acquisition camera according to the needs of the body. Rotation to meet all-round monitoring of weldments.

The image processing module is the basis for the calculation of surface dimensions and defect identification of weldments.

The original image collected by the CCD camera has characteristics such as low contrast, uneven illumination, and large background fluctuations, which make the image after segmentation has multiple target areas. Therefore, each target area needs to be separated to achieve separate Calculation. At present, the commonly used extraction methods are divided into two categories: one is based on the analysis of the column gray waveforms of different weld defect images, and the positioning of the edge of the target area is performed according to the image column gray curve characteristics; the other method is the image The silhouette method is a method of reconstructing a background region and performing a difference between the reconstructed image and the segmented image to obtain a related target region. Each of these two types of methods has its own advantages and disadvantages, and should be flexibly selected according to the image quality and characteristics. The method of image silhouette is used to extract the target area.

Parameter calculation and defect assessment module

The target area processed by the image processing module has been basically segmented from the background area in the image, and then the target area feature value can be calculated according to the set calculation method. The feature value will also provide important information for the classification of the target area. in accordance with. According to GB6417-86 "Classification and Description of Welding Defects in Metal Fusion Welding", the six defects in the weld are defined and described, and the characteristic quantities of surface defects are set.

The classification of surface defects on surface welds is also a core part of weldment appearance inspection systems. The defect classification module uses an inference engine based on the expert system for defect classification. First, the feature quantity in the target area feature is used as the inference selection condition, and it is input into the expert system for classification reference value. The expert system will judge the selected condition according to the established inference criteria, and assign the inference result to the confidence level. For user reference. If there is any objection to the judgment result, the user can modify the reasoning result according to the actual situation, fill in the relevant content according to the prompts, and supplement the relevant reasoning guidelines in order to improve the classification accuracy ^[1]

Information Management and System Maintenance Module

This module has the function of recording the calculation results in real time. The module classifies and saves the data calculated by the system. At the same time, it provides users with information query, modification, deletion, update, and printing functions, and establishes an information cache in the module for people. Machine information exchange. The man-machine exchange interface not only facilitates users to check information in real time, but also can quickly check information records, providing relevant reference for production. At the same time, the module can also generate inspection reports in standard prescribed formats for archiving by inspectors as needed.

The system maintenance function in the module can assign rights to users. The assignment of permissions can ensure the security of the system and prevent users from losing important information due to misuse. At the same time, different types of users are allowed to access information from different information channels to ensure that information can still be transmitted smoothly when multiple users use the system at the same time. Information redundancy will also be generated during system operation. Users can log in to the data exchange area in the administrator mode and organize the data information to achieve the purpose of "thinning" the system. ^[1]

The above has systematically explained the design and development process of the weldment appearance inspection system. Through the introduction of each module of the system, the functions of the system have been explained. Through a detailed introduction to the image processing process of the system, it can be seen that the system can achieve the purpose of calculating and evaluating the target area, and then realize the construction of an intelligent detection system. The information of the system design is searched, modified, deleted, stored, and printed in order to facilitate the user's use and reflect the friendliness of the system ^[1] .