Suppliers of traditional old industrial X-ray technology have not been able to solve the real problems of their customers: productivity is improved by preventing defects in production instead of finding defects. The idea of ​​relying on inspection to ensure quality is not suitable, and the result is only low quality and high cost. Dr. Deming's most famous work in Japan in 1950 led to a revolution in quality and production economy. We use his method to prevent defects and ensure quality by identifying and adjusting the wheel casting process and focusing on the cause of the defect. The main quality goal of any wheel factory is to use the feedback information of the X-ray equipment to adjust the wheel defects made by the casting machine to less (as shown in Figure 1).
Productivity and X-ray chain real-time detection
1. Productivity definition
Improving product quality is a more economical approach than using methods such as automation to increase productivity. Because automation is used, although the automation of the output is increased, it also has to pay for the automation of the input. Even if the final productivity increases, the growth rate is limited.
2. Advantages of X-ray real-time detection and automatic defect recognition Compared with non-destructive detection such as ultrasonic wave and eddy current method, X-ray real-time detection has its advantages, and it can obtain complete information of wheel casting defects, not just the surface directly visible. defect. Based on the information available, the information on the wheel castings (space position, product type, geometry, size and orientation, etc.) obtained by X-ray real-time detection is intact. In addition, it is also important for other aspects of factory basic management, such as automation (Automatic Defect Recognition: ADR, no manual inspection), improved real-time defect display and evaluation, easy operation, synchronization with manufacturing process, and high reliability. Simple implementation.
3. Two reasons for X-ray image using ADR (1) Automatically find defects in wheel castings, classify good and bad wheels (reduce cost and improve quality). The purpose of ADR in general open-loop mode is to save human resources costs in a continuous production process (7 days a week, 24 hours a day).
(2) Using the feedback information of the X-ray machine, the casting machine can be adjusted to have almost no defects in production (adjusting the manufacturing process during die casting). In this closed-loop mode, the purpose of ADR is to increase the yield (eg, reduce the scrap rate from 40% to 5%). The X-ray image of the unqualified wheel is transmitted by the fiber to the end of the casting machine for analysis by the casting operator. Based on the observed defects and their trends, the wheel casting parameters can be adjusted in time.
Problem definition
Using X-ray real-time imaging, we will find some irregular structures - such as pores, pinholes, loose and foreign objects - that are not allowed in the measured wheel. The defects in the image are small and bright and are enclosed in a larger, darker image area. As shown in FIG. 2, the image area containing the defect has some fine structure, and the area containing no defect has a white rough texture.
The microstructure is identified at the time of detection while the roughness is ignored. But there are also some uncertainties between the measured and actual dynamic dimensions, such as the following: Difficulties associated with finding defects in wheel X-ray images:
(1) Defect location At a certain part of the wheel, multiple defects appear in the same place, and these defects are superimposed on each other.
(2) Defect Contrast The method of defect detection is based on finding local density differences. The image input system (the image enhancement chain when the camera is photographed) must submit enough contrast, that is, an image with contrast is required.
(3) Defect type A large number of different defects in the image and their typical variations make it impossible to perform direct detection according to a preset reference picture.
(4) Image structure The X-ray image of the wheel includes some image structures (such as irregular surface and shape of the wheel, repeated reinforcement), which appears to be defective, but not. These (pseudo-defects) can cause false positives in the image analysis program.
Spiral Wound Ptfe Gasket,Expanded Ptfe Filler,Tape For Sprial Wound Gasket,Waterproof Eptfe Tape For Swg
Cixi Congfeng Fluorine Plastic Co.,Ltd , https://www.cfptfeseal.com