On-line Inspection Research Articles

Evaluation of the acoustic emission monitoring method for stress corrosion cracking on aboveground storage tank floor steel

Stress corrosion cracking (SCC), which is one of the main threats to the structural integrity of aboveground storage tanks (ASTs), has recently attracted much attention. In this paper, the SCC of the storage tank floor was studied by the acoustic emission (AE) and electrochemical methods. Throughout the whole SCC process, the AE signal characteristic parameters, cumulating features, and correlations were analyzed, and the signal time-frequency local features were extracted via wavelet packet analysis based on the MATLAB program. A good correlation of the instantaneous variations between the AE activity and the corrosion behavior was found. Microcrack creeping and macrocrack propagation were the most important AE sources in SCC; these sources can be recognized by the AE parameters and the AE frequency spectrum. The results in this paper will be beneficial for the diagnosis of corrosion severity and the recognition of the corrosion sources by AE online inspection and will further provide an assessment of the AST integrity by means of a risk-based inspection technique.

Laser Induced Phased Arrays (LIPA) to detect nested features in additively manufactured components

Additive manufacturing (AM) has the capability to build complex parts with internal features, which have many advantages over conventionally manufactured parts. This makes AM an alternative for advanced manufacturing sectors. AM components suffer from defects due to the lack of understanding in the build process. This makes the adaptation of AM in safety-critical industries, such as aerospace, problematic. The current AM work flow calls for costly off-line inspections to qualify components as defect-free. The layer by layer nature of the AM provides an opportunity for an on-line inspection to take place. This can provide early detection of defects as well as information for optimization and repair of the build. Laser Induced Phased Arrays (LIPA) present themselves as a viable remote, non-destructive, ultrasonic technique capable of being implemented as part of an on-line inspection of AM. Lasers are used to generate and detect ultrasound and a phased array is synthesized in post-processing. This paper demonstrates the capability of LIPA to successfully detect and locate features within AM components off-line. Cylindrical features as small as 0.2 mm in diameter and 26 mm above the inspection surface were detected using LIPA and verified using X-ray computed tomography (XCT).

A straightforward and low-cost pre-inspection measurement method for small gears

On-line or rapid inspection of parts is essential in streamlining manufacturing processes. A novel pre-inspection method for measuring small gears is proposed. This method requires little specialist equipment outside that normally found in a laboratory or on a shop floor. Gears are measured using a standard optical microscope. These are then processed using an image processing algorithm to give a percentage error in tooth profile. This allows only gears of sufficiently high quality to be “passed on” to more sophisticated inspection techniques. The technique is used to analyse sub-centimetre diameter gears produced using WEDM. It is found that the technique provides a simple but effective method of determining how close a gear is to the required geometry. While the result is basic, it is extremely quick and inexpensive to perform alongside the manufacturing process. This method provides a capability to small companies and production lines for pre-inspection of gears before detailed analysis without purchasing specialist equipment.

AKM2D: An adaptive framework for online sensing and anomaly quantification

In point-based sensing systems such as coordinate measuring machines and laser ultrasonics where complete sensing is impractical due to the high sensing time and cost, adaptive sensing through a systematic exploration is vital for online inspection and anomaly quantification. Most of the existing sequential sampling methodologies focus on reducing the overall fitting error for the entire sampling space. However, in many anomaly quantification applications, the main goal is to estimate sparse anomalous regions at pixel-level accurately. In this article, we develop a novel framework named Adaptive Kernelized Maximum-Minimum Distance to speed up the inspection and anomaly detection process through an intelligent sequential sampling scheme integrated with fast estimation and detection. The proposed method balances the sampling efforts between the space-filling sampling (exploration) and focused sampling near the anomalous region (exploitation). The proposed methodology is validated by conducting simulations and a case study of anomaly detection in composite sheets using a guided wave test.

A mechanical transmission based image de-blurring method used for on-line surface quality inspection

A mechanical transmission based approach was employed to achieve motion blur removing of the image by eliminating relative movement. To achieve linear motion of the camera especially during exposure time, “velocity curve splicing” was proposed to plot cam outline which contained a linear phase. A lab-scale equipment was established where velocity data were collected to test performance of object tracking. It was shown that the camera velocity would properly match the object velocity, the linear relationship between object and cam motions was achieved as well. The images were captured on-line, and it was indicated that the blurring length of object image was less than 1 pixel, which proved the reliability of proposed method for motion blur removal. Further, this method was compared with conventional Lucy-Richardson algorithm. It was found that our method not only saved the time for image restoration, by also surpassed the latter in guaranteeing image quality.

Research and Application of Cable Controlled Layered Water Injection Technology in Daqing Oilfield

Daqing Oilfield has entered the development stage of high water cut later stage, the contradiction between inter layers and interior layers is more prominent, the rate of water injection pass rate decreases rapidly, the manual test workload is huge, the existing test team can not meet the requirements; at the same time, the existing high-efficiency measurement and adjustment technology cannot provide Continuous monitoring data and real-time dynamic adjustment of downhole injection volume, for this purpose, Daqing Oilfield carried out research and research on cable-controlled intelligent layered water injection technology. The technology is mainly composed of two parts: underground cable-controlled intelligent water injection technology pipe column and ground wireless remote control system. It can realize real-time monitoring and control of underground stratified flow and pressure in the office, realize online inspection, automatic measurement and adjustment, and report output. With the function, 30 wells were tested in the field and achieved good results. This technology is a continuous, efficient and real-time new generation water injection technology.

Research on Controlled Jet Ultrasonic Inspection Method for LZ50 Hot Rolled Large Steel Bars

Ultrasonic inspection is the main method of on-line inspection for large size steel bars. For maximum Φ 600 mm in diameter, single maximum length up to 10 m of LZ50 hot rolled steel bar, to meet the requirement of the on-line inspection efficiency and precision at the same time, adopting controlled jet method that the probe and the workpiece is non-contact, coupled with one certain length water body, and its comprehensive inspection ability is stronger. In the application process of controlled jet method, the self-developed jet coupling module and the zonal inspection technology are used to effectively suppress the turbulence noise and structure echo. The test results show that under the premise of signal-to-noise ratio (SNR) not less than 8 dB, the controlled jet method can effectively identify Φ 3.0 mm artificial flat bottom hole of 5 mm and 300 mm in depth, and verify the effectiveness of the controlled jet method.

Prediction of flashover voltage using electric field measurement on clean and polluted insulators

The economy of transporting large amount of power over long distance makes high voltage transmission line an important component of the power system. Outdoor insulators provide electrical insulation as well as mechanical support between energized conductor and grounded parts of the tower. Its performance is closely related to the reliability of power delivery. Various parameters affect the insulator flashover performance, but the measurement and analysis of all those parameters would be a challenging and complex task during online inspection. The performance of the insulators is directly dependent on the electric field distribution inside and surrounding them. The aim of this work is to obtain the relationship between flashover voltage and electric field distribution. Accordingly, a novel electric field measurement technology, which was recently commercially available, was utilized in this paper to measure the electric field distribution around six different varieties of composite and ceramic insulators. The relationship between conductivity of the contamination layer and the environmental relative humidity was initially evaluated. Further, the patterns of the electric field distribution for clean, uniform and non-uniform contamination cases were studied and its relationship with the flashover voltages was explored. Electric field calculation with commercially available package was performed to validate the experimental results and extend the conclusion to higher voltage levels.

A Machine Vision System for Online Metal Can-End Rivet Inspection

Can end rivet fracture is considered a serious defect that may arise during the manufacturing of metal cans used in food industry. Its occurrence may result in inefficiency of the easy-opening system, leakage or contamination of the food. Thus, an inspection procedure must be performed in order to remove the defective can ends from the production line. However, the location of the rivet and the characteristics of the fracture defect make it difficult to use simple devices for online inspection of 100% of the cans. In this paper, a new method for detecting fracture defects in can end rivets based on machine vision is described. In the proposed technique the rivets are detected employing Otsu segmentation, morphological filtering and connected component analysis. The classification of each rivet as being defective or non defective is performed using a minimum distance classifier based on circularity and area features. The software and hardware architecture of a real time inspection system and the technical aspects for its integration into a production line are presented. The described system can achieve an accuracy as high as 98.13% for rivet defects detection according to experiments and tests performed in real conditions of use.

Dynamic inspection of a rail profile under affine distortion based on the reweighted-scaling iterative closest point method

The precision and efficiency of rail inspection are of vital importance to the operation and safety of railway systems. However, because of vibrations that occur in the rail inspection vehicle and installation error of the line laser sensor, the collected rail profile is distorted. As a result, the accuracy and robustness of conventional registration and inspection methods cannot be guaranteed. In this paper, an accurate and robust calibration method for measuring rail profile distortion is proposed. In this study, the impact of vibration on the rail profile, which was obtained with a laser sensor, was analyzed. Then, the collected rail profile was divided into two segments according to the actual condition where only the upper part of the rail was subjected to wear. An accurate and robust reweighted-scaling iterative closest point algorithm was proposed to calibrate the distorted rail profile, and the upper and lower boundaries of the scale ratio were used as constraints to solve the registration problem effectively and efficiently. Finally, the Hausdorff distance method was introduced to clearly visualize the rectified rail profile and wear. The experimental results demonstrated that the proposed method can realize the affine transformation of the distorted rail profile with a high precision and robustness. Additionally, it can achieve an online dynamic calibration and inspection of the rail by comparing the data with the conventional iterative closest point, scaling iterative closest point, and Calipri.

Recognizing Coiled Tubing Defect Characteristics Based on Weak Magnetic Detection Technology

The development of coiled tubing operation technology has continuously improved with the higher requirements of coiled tubing on-line detection technology. The current coiled tubing field weak magnetic on-line detection system calculates the gradient of the signal for improving the locating precision. Quantitative identification of defect shape, based on the weak magnetic detection and magnetic leakage flux systems, is a key issue to evaluate whether coiled tubing can continue to be used. The 2-D finite-element models with different typical defect shapes were established, and the gradient changes of the magnetic-field intensity are calculated. The influence of defect size on signal characteristics is analyzed, and a functional model can be obtained. The signal characteristics of a 2-D self-leakage magnetic field were further analyzed, and a 2-D cloud image was established used to identify the shape of defects. The level of damage as a result of coiled tubing use is evaluated according to the identified defect shape, which is significant for a quantitative description of defects and the accurate evaluation of conditions in the coiled tubing on-line inspection system. This research of identification of defect shape and size makes up the disadvantage that weak magnetic detection can only judge whether the defect of log exists or not.

Automation of On-Line Quality Control of Laminated Glass

Statistical quality control of laminated windshields during bending was performed. Parameter deviations of the articles from technical specifications were uncovered. A logistic regression model was constructed to determine the probability of obtaining high-quality laminated automobile glass as a function of the bending regime. It was shown that logistic regression is effective for on-line inspection of laminated windshields and in making decisions concerning bending regime adjustments.

Surface Quality Assurance Method for Lithium-Ion Battery Electrode Using Concentration Compensation and Partiality Decision Rules

This paper presents a novel method for lithium-ion battery electrode (LIBE) surface quality assurance. First, based on machine vision, an automatic optical inspection system is developed to check defects on LIBE. In addition, a background normalization algorithm is put forward to preprocess the large-scale LIBE with inhomogeneous thickness in uneven illumination. With the help of the auto-concentration compensation algorithm, flaws can be extracted precisely. Moreover, after characterizing the defects, features machine applied partiality parameter automatic adjustment method and partiality decision rules are exerted for defects accurate classification, which provides near-optimal performance and reduces the complexity of tuning parameters. The proposed method is computationally efficient and satisfies real-time online inspection requirement. Experimental results verify the effectiveness and performance of the proposed method according to the inspection speed and recognition rate.

Multispectral Fluorescence Imaging Technique for On-Line Inspection of Fecal Residues on Poultry Carcasses.

Rapid and reliable inspection of food is essential to ensure food safety, particularly in mass production and processing environments. Many studies have focused on spectral imaging for poultry inspection; however, no research has explored the use of multispectral fluorescence imaging (MFI) for on-line poultry inspection. In this study, the feasibility of MFI for on-line detection of fecal matter from the ceca, colon, duodenum, and small intestine of poultry carcasses was investigated for the first time. A multispectral line-scan fluorescence imaging system was integrated with a commercial poultry conveying system, and the images of chicken carcasses with fecal contaminants were scanned at processing line speeds of one, three, and five birds per second. To develop an optimal detection and classification algorithm to distinguish upper and lower feces-contaminated parts from skin, the principal component analysis (PCA) and partial least square discriminant analysis (PLS-DA) were first performed using the spectral data of the selected regions, and then applied in spatial domain to visualize the feces-contaminated area based on binary images. Our results demonstrated that for the spectral data analysis, both the PCA and PLS-DA can distinguish the high and low feces-contaminated area from normal skin; however, the PCA analysis based on selected band ratio images (F630 nm/F600 nm) exhibited better visualization and discrimination of feces-contaminated area, compared with the PLS-DA-based developed chemical images. A color image analysis using histogram equalization, sharpening, median filter, and threshold value (1) demonstrated 78% accuracy. Thus, the MFI system can be developed utilizing the two band ratios for on-line implementation for the effective detection of fecal contamination on chicken carcasses.

Automatic Virtual Metrology for Carbon Fiber Manufacturing

Carbon fiber is currently one of the most popular composite materials in the world, with its wide applications ranging from bikes to space shuttles. However, there is no comprehensive method for the total quality inspection of carbon fiber products so far due to its feature of continuous production and destructive examination of quality inspection. Through years of development, the Automatic Virtual Metrology (AVM) technology has been applied to various industries to turn the offline sampling inspection with metrology delay into online and real-time total inspection. With AVM, every workpiece of the carbon fiber products can be examined to ensure total quality inspection. The biggest issue of continuous production is that it can neither accurately define each unit nor label on the products. To solve this problem, a production data traceback (PDT) mechanism is proposed in this letter. The PDT mechanism can provide a virtual label to each unit for referring back to the process parameters of the production. With PDT, the process data of each workpiece of a complete spin can be acquired so as to accomplish the work-in-process (WIP) tracking requirement of applying AVM. Also, the Advanced Manufacturing Cloud of Things (AMCoT) platform is adopted in this letter to fulfill all the smart manufacturing requirements for achieving the goal of zero defects of carbon fiber manufacturing.

Final optics damage online inspection in high power laser facility*

For the laser-induced damage (LID) in large-aperture final optics, we present a novel approach of damage online inspection and its experimental system, which solves two problems: classification of true and false LID and size measurement of the LID. We first analyze the imaging principle of the experimental system for the true and false damage sites, then use kernel-based extreme learning machine (K-ELM) to distinguish them, and finally propose hierarchical kernel extreme learning machine (HK-ELM) to predict the damage size. The experimental results show that the classification accuracy is higher than 95%, the mean relative error of the predicted LID size is within 10%. So the proposed method meets the technical requirements for the damage online inspection.

Computer vision detection of foreign objects in walnuts using deep learning

Detection of foreign objects is crucial for quantitative image analysis in numerous food quality and safety inspection applications. Rapid detection of foreign objects in walnuts using computer vision still faces challenge due to the irregular shapes and complex features of foreign objects. Some detection methods require application specific transforms, expertly designed constraints and model parameters, and have limited detection performance due to their maintenance costs. In recent years, deep learning has become a focus in different research fields, because methods based on deep learning are able to directly learn features from training data. In this study, we apply two different convolutional neural network structures to walnut images for automatically segmenting images and detecting different-sized natural foreign objects (e.g., flesh leaf debris, dried leaf debris and gravel dust) and man-made foreign objects (e.g., paper scraps, packing material, plastic scraps and metal parts). The proposed deep-learning method is simpler because it avoids extracting features manually, and overcomes the conglomeration phenomenon between walnuts and foreign objects in actual images. The proposed method is able to correctly segment 99.5% of the object regions in the 101 test images and to correctly classify 95% of the foreign objects in the 277 validation images. The segmentation and detection processing time of each image was less than 50 ms. Future work will focus on deep learning using multi-waveband imaging hardware and fast on-line inspection control for the equipment and robots.

A method of steel ball surface quality inspection based on flexible arrayed eddy current sensor

In order to perform the rapid and efficient automatic on-line inspection of the surface quality of steel ball, a new spherical deployment method based on flexible array eddy current sensor (FAECS) is proposed. Full unfolding of the steel ball surface can be performed by combining the flexible array eddy current sensor with the one-dimensional orbital deployment. The inspection coil of array eddy current sensor is fabricated on flexible printed circuit board (FPCB), which is bent into a semicircle to fix on the top of one-dimensional track. When the steel ball rolls on the one-dimensional track, complete inspection of steel ball surface can be performed by the flexible array eddy current sensor. In this method, one-dimensional mechanical motion is used instead of the traditional two-dimensional deployment motion, which reduces the complexity and cost of deployment structure. Three typical defects include linear cracks (0.12 mm width × 0.5 mm depth), cross cracks (0.06 mm width × 0.9 mm depth) and arc cracks (0.05 mm width × 0.1 mm depth) are measured in the experiment to verify the effectiveness of the inspection system. Subsequently, same batch of steel balls are tested at speed of 4 pcs/s, results show that the inspection accuracy of surface defect of steel balls is 95.52%, and the false inspection rate is less than 0.7%, which proves this method efficiency.

A deep-learning-based approach for fast and robust steel surface defects classification

Automatic visual recognition of steel surface defects provides critical functionality to facilitate quality control of steel strip production. In this paper, we present a compact yet effective convolutional neural network (CNN) model, which emphasizes the training of low-level features and incorporates multiple receptive fields, to achieve fast and accurate steel surface defect classification. Our proposed method adopts the pre-trained SqueezeNet as the backbone architecture. It only requires a small amount of defect-specific training samples to achieve high-accuracy recognition on a diversity-enhanced testing dataset of steel surface defects which contains severe non-uniform illumination, camera noise, and motion blur. Moreover, our proposed light-weight CNN model can meet the requirement of real-time online inspection, running over 100 fps on a computer equipped with a single NVIDIA TITAN X Graphics Processing Unit (12G memory). Codes and a diversity-enhanced testing dataset will be made publicly available.

Comparison of rapid techniques for classification of ground meat

Computer vision and near infrared spectroscopy are fast and non-invasive techniques currently available for processing control in the meat industry. These techniques can be used, either separately or combined, for on-line assessment of meat quality parameters. This study aimed to compare a portable near-infrared (NIR) spectrometer, near infrared hyperspectral imaging (NIR-HSI) and red, green and blue imaging (RGB-I) to differentiate ground samples from beef, pork and chicken meat; and to quantify amounts of each in mixtures. Chicken breast meat was adulterated with either pork leg meat or beef round meat from 0 to 50% (w/w). Partial Least Squares regression (PLSR) models were performed using full spectra and after selecting most important wavelengths. The best results were obtained with NIR-HSI, with coefficient of prediction (RP2) of 0.83 and 0.94, ratio performance to deviation (RPD) of 1.96 and 3.56, and ratio of error range (RER) of 10.0 and 18.1, for samples of chicken adulterated with pork and beef, respectively. In addition, the results obtained using NIR spectroscopy and RGB-I confirm that these techniques provide an alternative for rapid, on-line inspection of ground meat in the food industry.