带钢表面缺陷视觉检测系统关键技术研究

为满足钢板表面缺陷在线检测系统宽幅面、高速、高分辨率的检测要求，讨论了基于线阵CCD的钢板表面缺陷视觉检测系统实现的关键技术；优化设计了视觉检测系统的光学照明部分，以检测不同类型的缺陷。通过软件系统的特殊设计，以保证实时在线检测。针对缺陷图像低对比度、高噪声的特点，提出了基于灰度统计特性的图像边缘检测方法，并实现了对缺陷图像的自适应阈值分割。依据图像的缺陷统计特性，定义了缺陷的灰度、几何等特征量，用于缺陷分类。本系统样机已在实验室环境下运行。     

成批抛光钢球表面缺陷自动检测的图像预处理方法

针对目前企业缺乏钢球表面缺陷检测装备,特别是小直径钢球检测装备的现状,笔者研制开发出一套基于机器视觉、可批量检测小直径钢球表面缺陷的新装备.由特制机电装置将钢球成批自动行列相间地上料并使球面翻滚,相机以一定帧频拍摄多幅钢球表面图像.在保证检测精度和效率的前提下,设计出应用于该装置的图像预处理方法,包括模板标定、图像分割和图像增强.实际检测试验证明,该预处理方法一方面使表面缺陷突显,有利于后期检测,另一方面抑制外接环境引入的各种干扰,消除不必要的影响因素.     

Assessment of surface bonding strength in Al clad steel strip using electrical resistivity and peeling tests

Abstract

Aluminium clad steel strip successfully combines the surface properties of an aluminium alloy coating with the satisfactory mechanical properties of the steel substrate. The production of Al clad steel strip by rolling, however, is a more efficient and economical approach compared with other processes. In this investigation, trilayer strips of aluminium/steel/aluminium were produced using the cold roll bonding technique. The bonding strength between the layers and the electrical resistivity of the samples were measured using a peeling test and four point probe test, respectively. The effects of reduction of thickness, the friction condition between the outer layer and rolls, and the rolling velocity on the bonding strength and electrical resistivity of the samples were assessed. Finally, it is shown that the resistivity test can be used as a non-destructive test for the evaluation of the quality of bonding between the layers of aluminium clad steel strip.     

镍-钴合金镀钢带及其制成的锂电池钢壳的耐腐蚀性能

用于锂电池壳体制备的预镀镍工艺会使钢壳表面有大量裂纹,耐蚀性下降。通过3种方式在钢带上镀覆镍-钴合金层,并冲压制成电池钢壳,采用扫描电镜(SEM)、能谱、X射线衍射(XRD)、电化学测试、中性盐雾试验、硫酸铜点滴试验研究了镀覆钢带和电池钢壳的组织结构和耐腐蚀性能。结果表明:电镀2μm镍-钴合金层经700℃热处理3 h后再加镀2μm镍-钴合金层制备的钢带表层致密均匀,加镀的薄膜封闭了热处理后晶粒间存在的位错等缺陷产生的针孔,具有良好的耐腐蚀性能;电镀4μm镍-钴合金层经700℃热处理3 h制备的钢带冲压而成的电池钢壳表面裂纹少且细小,裂纹处为镍/钴/铁固溶体,耐腐蚀性能明显优于其他2种方式制备的钢壳。     

Fast and High-Resolution Optical Inspection System for In-Line Detection and Labeling of Surface Defects

Automated optical inspection systems installed in production lines help ensure high throughput by speeding up inspection of defects that are otherwise difficult to detect using the naked eye. However, depending on the size and surface properties of the products such as micro-cracks on touchscreen panels glass cover, the detection speed and accuracy are limited by the imaging module and lighting technique. Therefore the current inspection methods are still delegated to a few qualified personnel whose limited capacity has been a huge tradeoff for high volume production. In this study, an automated optical technology for in-line surface defect inspection is developed offering high performance in spatial resolution and detection speed for any surface. The inspection system consisting of an LED array which illuminates a wide inspection area on the test object captures scattered light from surface defects using a 12288-pixel line CCD at 12 kHz acquisition rate. A 3.5 \(\mu\) m per pixel resolution of the line CCD provides a detection width capability of at most 43 mm which is equivalent to 147 megapixels image data acquired per second. To handle the large volume of data per acquisition cycle, the data are transmitted from a host CPU to multiple GPU devices where CUDA-based image processing kernels are adopted to perform detection and labeling of surface defects in parallel. The processed data is sent back to the CPU to display user-defined defect maps. 2-D inspection of back-coated flat mirrors, 43 mm x 70 mm\(^{2}\) in size, using a single CCD module and multiple GPU reveals that surface flaws such as bubbles, cracks, and edge defects are detected accurately. The acquisition time to capture and load the data to a CPU is 1.7 s while the processing time to transmit the same data for surface defect detection in a GPU is 248 ms. The latter time scale is considerably faster compared to minute-long computations in solely CPU-based processing algorithm of the same test object. The minimum width of detected surface defects is about 10 \(\mu \)m with true detection rates above 94%. Moreover, the inspection system is easily configurable by tasking multiple CCD imaging modules to different GPU devices to allow inspection of larger test objects. This flexibility can improve both acquisition and detection speeds to boost in-line circuit chips, packaging, and touchscreen panel inspection systems.     

水利工程中金属焊缝缺陷的超声检测方法比较研究

为研究基于超声的无损探伤方法在水利工程金属结构焊缝缺陷识别中的应用,利用常规超声检测技术、超声相控阵技术、衍射时差法(Time of Flight Diffraction, TOFD)超声检测技术对水利工程金属结构焊接试块缺陷进行识别,分析了各种缺陷在超声无损探伤技术中的特征显示。研究结果表明:常规超声检测技术、TOFD检测技术均能对各种缺陷实现信号显示,超声相控阵检测技术对气孔和横向裂纹的显示不够明显,但对其它缺陷的检出效果较为明显;常规超声检测技术对操作人员的要求较高,对缺陷的定性困难,精度不高;TOFD检测结果中气孔和横向裂纹的显示呈现出一种特殊的弧形,有一定高度的内部裂纹和未熔合的信号由上下尖端衍射波组成,根部未焊透上下尖端信号不够明显;相控阵检测结果直观,可以较精确地测量缺陷的埋藏深度、自身高度、长度等,但在扫查点状缺陷或者与超声声束平行的裂纹缺陷时,检出率极低。     

Development of a surface defect inspection system using radiantlight from steel products in a hot rolling line

The steel industry is constantly trying to reduce production cost and improve quality by making the steel manufacturing processes continuous and faster. Currently, the rolling process of steel production is largely automated, while the finishing process is not yet appreciably automated. The finishing processes involve many tasks difficult to automate, such as defect inspection and repairing the detected defects. In recent years, however, many automated and labor-saving systems have been developed for use in the finishing processes. The surface defect inspection of steel products is the largest bottleneck in the finishing process. This paper describes an inspection system of steel surface defects for large sections, such as wide flange beams and I-beams. This system is based on applied radiant light and it senses the temperature deviation caused by defects. The wavelength of the detector is optimized to improve the signal-to-noise ratio. An optical attenuator was developed to compensate for the known temperature distribution across the product immediately after rolling. The image processor takes only 50 ms per image frame. Each time frame has the necessary image information to detect defects     

Discrimination Between Cracks and Recrystallization in Steel Using Nonlinear Techniques

One major problem in ultrasonic NDT for steel products and welding inspection is that standard linear methods are often unable to distinguish the nature of signals. Partially recrystallized grains, voids, small cracks or inclusions in the piece under investigation could produce indications very similar in terms of acoustic energy reflected and ultrasonic peaks envelope. Here, we analyze the nonlinear response to ultrasonic excitations of steel bars with both kind of imperfections purposefully generated. Using the Scaling Subtraction Method as a tool for the analysis, we show differences in the nonlinear signature, which can be used to distinguish nondestructively a crack/delamination from a region with imperfect grains formation, with possible applications of this technique in the production cycle.     

Influence of surface roughness on the bonding quality in twin-roll cast clad strip

Manufacturing of aluminum-steel clad strips by means of twin-roll casting is of great interest due to a shorter production chain in comparison with convenient technologies. Experiments on twin-roll casting of clad strips of pure aluminum and an austenitic stainless steel of 2.5?mm total thickness with the inline abrasive cleaning of a steel substrate’s surface performed. The corresponding device for the inline surface preparation is designed. The influence of surface roughness of the substrate’s surface on the bonding strength between strip layers is analyzed. The mechanism of intermetallic bonding on the roughened surfaces is discussed. It is shown that surface roughness of steel substrate up to 4.2?µm provides bonding strength above 100?MPa.     

PECVD mit der Bandhohlkathode

Plasma‐enhanced High Rate Surface Treatment of Steel Sheet by Means of the Strip Hollow Cathode Method Low pressure glow discharges are of strong interest for the development of new technologies for steel strip surface treatment for different applications. The challenge of the integration of plasma processes into existing continuous steel sheet coating processes consists in the requirements of a very high deposition rate combined with minimal loss due to parasitic oatings. Regarding these requirements a novel strip hollow cathode process (SHC) for surface treatment based on the hollow cathode effect between two flat and parallel cathodes has been developed. This method has significant potential for surface treatment of large area metal strips in continuous technological processes.     

Computer vision and robotic manipulation for automated feeding of cork drillers

The production of cork stoppers is currently a process in rapid evolution. Where in the past hand labor was common, today we observe increasing attempts to introduce technologies that increase the productivity of production lines. One these example are automated cork drillers that produce thousands of cork stoppers per hour. In order to harness this increase in productivity to its full potential, it must the followed by other processes in the production line upstream of the driller, as for example the feeding of the latter.This article presents the application of computer vision techniques that extract information of cork strips that move on a conveyor belt, to obtain automated feeding of a cork driller using a robotic manipulator. Image Processing extracts information regarding the strip position and orientation, and also which side of the strip is visible. Thus the strip is consistently placed in the driller in order to extract stoppers from the best quality cork. The segmentation of the cork strips is obtained by background subtraction. To estimate the strip visible surface, we apply Machine Learning techniques that enable a robust classification given a set of features extracted from the cork texture.In the experiences carried out, we were able to obtain 100% classification rate with a test dataset of more than a hundred cork strips.     

Dielectric constant measurement technique for a dielectric strip using a rectangular waveguide

A dielectric constant measurement technique for dielectric strips is presented in this paper. In the measurement, the strip is placed parallel to the broad walls of a rectangular waveguide, and it is found that the measured reflection coefficient is insensitive to the position of the strip when it is placed around the middle of the waveguide cross section. The new sample placement scheme becomes very convenient, especially when a large number of strips have to be measured. To develop the forward scattering formulation, the reflection coefficient of the strip placed in the waveguide is evaluated using the method of moments. With this forward model, a genetic algorithm is developed to retrieve the dielectric constant of the strip from the measured reflection coefficient. The validity of the calculated reflection coefficient is verified by measuring a Teflon strip in a WR187 waveguide, and the dielectric constant of the Teflon is successfully retrieved from the measurement. For the measurement of many strips, a special sample holder is made to ensure insensitivity of the measured reflection coefficient to the position of the strip.     

A Discussion on Evolution of Microstructures and Influence Factors during Continuous Rolling of Compact Strip Production

The evolution of microstructures and influence factors of ultrathin hot strips of low carbon steel produced by compactstrip production (CSP) techniques were investigated. The steel blocking samples of CSP six-passes were obtained,and microstructures at the different positions of workpiece for each pass were studied. At the same time, anexplicit finite element technique was used to reveal the continuous rolling process. By combining experiment resultswith simulation analysis, the effects of plastic strain, temperature, precipitation and interval time on evolution andrefinement of crystal grains have been investigated. The results are useful for the development of high strength hotstrips.     

超声Δ法与TOFD法在薄壁电子束焊缝上的检测结果对比分析

采用超声Δ衍射法和TOFD法对模拟缺陷进行检测,得出了缺陷检测波形图和B/C扫描图。对比分析了超声Δ衍射法与TOFD法的检测结果。实验表明利用超声Δ衍射法能够识别近表面垂直表面的面状缺陷,有效地对缺陷进行定位,能够很好地克服表面盲区,对缺陷的检测识别能力不受缺陷方向限制。从检测结果数据可以看出超声Δ衍射法的检测灵敏度和缺陷识别能力优于TOFD法。     

A Design Study of an Automatic System for On-line Detection and Classification of Surface Defects on Cold-rolled Steel Strip

In collaboration with the British Steel Corporation, Sira Institute Ltd is currently pursuing the development of sophisticated equipment for the automatic inspection of cold rolled steel strip. This paper will examine the special difficulties associated with the application of optical inspection techniques to a dull metallic surface at line speeds of up to 50 ms m 1. The development work has centred on the use of high-speed scanned laser systems, unique in their ability to provide diffraction-limited resolution with intrinsically high optical throughput. In order to detect and delineate all significant defects down to the most subtle imperfections, the polar scattering characteristics of the surface have been exploited and optimum signal processing techniques are being identified. The paper will discuss some of the optical, mechanical and electronic constraints imposed by the inspection requirement.     

Industrial on-line texture determination in rolled steel strips

In the manufacture of sheet metal, it is ofgreat importance that the quality should be homogeneous over the entire length of a strip. This can only be monitored using a continuous, on-line measuring method. A new X-ray transmission measuring technique has been developed for the nondestructive determination of texture-dependent technological data of rolled strip. It can be applied on-line for both hot and cold rolled metal strip, especially steel strip. The paper describes the measuring principle and gives information on the application for measuring ther _m -values and the earing-values of cold rolled steel strip. With the data from these values, the user can control the deep drawing characteristics of his strips. The extension of the basic measuring principle to a condition-free determination system will be demonstrated with its application for measuring texture dependent technological data of hot and cold rolled steel strip. Long term experience at Hoesch Stahl AG, typical results measured over the total strip length, integration in quality assurance systems, questions of maintenance, etc. will be discussed.     

Surface Wave Acoustic Microscopy for Rapid Non-destructive Evaluation of Silicon Nitride Balls

A rapid nondestructive evaluation (NDE) technique based on surface wave acoustic microscopy (SWAM) has been developed for inspection of surface/subsurface defects in silicon nitride (Si₃N₄) balls. This technique exploits advantages of convergent surface acoustic waves generated by a sharply focused ultrasonic transducer by placing the transducer’s focal point below the ball surface, or defocusing. The use of surface waves to interrogate an area, rather than a point as in conventional acoustic microscopy, at a time on the ball surface can greatly improve the efficiency of defect detection. To ensure the inspection of surface defects with SWAM, we first simulated surface wave generation and propagation on a Si₃N₄ ball using finite element method (FEM). Then we tested different ultrasonic transducers for effective surface wave generation and detection. Transducers with varying frequency, f-number, defocusing distance, etc. for surface defect detection were experimentally compared. C-shape surface cracks on Si₃N₄ balls of different sizes were successfully detected. As a result, a SWAM prototype for rapid detection of surface defects in silicon nitride balls was successfully demonstrated.     

Optical technique for the automatic detection and measurement of surface defects on thin metallic wires

In industrial applications of thin metallic wires it is important to characterize the surface defects of the wires. We present an optical technique for the automatic detection of surface defects on thin metallic wires (diameters, 50-2000 mum) that can be used in on-line systems for surface quality control. This technique is based on the intensity variations on the scattered cone generated when the wire is illuminated with a beam at oblique incidence. Our results are compared with those obtained by atomic-force microscopy and scanning-electron microscopy.     

An investigation on the impact fatigue characteristics of valve leaves for small hermetic reciprocating compressors in a new automated test system

This paper presents an investigation on the impact fatigue characteristics of valve leaves that are prevalently used in hermetic reciprocating compressors especially for the household type refrigerators. A unique automated impact fatigue test system has been designed and produced, which enables to carry out impact fatigue tests of the compressor valve leaves under the desired impact velocities. The test system incorporates a noncontact actuation, a data acquisition system and an acoustic‐based damage detection technique, which continuously monitors the health of the structure. The damage detection system allows parametrical investigation on the impact fatigue life by detecting any possible damage and automatically terminating the test. The investigation relates the impact fatigue lifetime of the valve leaves with the impact velocity, asymmetrical impact, operation temperature, material type (carbon strip steel, stainless strip steel and new stainless strip steel grade) and tumbling operation duration. The observations show that the cracks have initiated from the edges of the valve leaf where is in contact with the valve plate. Subsequently, the cracks initially have propagated in the radial direction inwards the center of the impact area. Various failure cases have been resulted in by either a single crack or inter‐related multiple cracks. Microscopic and metallographic observations have been performed on the specimens to enhance the understanding of the damage mechanisms. The investigation and introduced test system guide the design optimization of the valve leaves in terms of compressor performance due to the energy consumption and lifetime of the valve leaf.     

In-situ cracking for nondestructive inspection process validation

ork at the Idaho National Engineering and Environmental Laboratory (INEEL) includes efforts to weld steel storage containers. Producing repeatable high quality welds and assuring weld quality is paramount to this work. The nondestructive examination (NDE) method used to measure weld quality is the magnetic particle testing (MT) process since the containers are fabricated from mild steel. A paradox occurred during the initial mockup welding and inspection system demonstrations. Since the welding was flawless, the semi-automatic remote MT system had no defects to find. The MT system was calibrated, as per the welding code, with calibration blocks, but it never demonstrated that it worked during the demonstration because of the flawless welds. To validate the MT system during the mockup welding demonstration, in-situ defects were artificially generated. Two-in. (5.1-cm) long copper wire segments were placed into the weld groove joint and consumed into the weld metal during welding. This in-situ crack generation technique provided physical defects for the inspector/regulator to observe, which then validated the inspection system.