Automatic Steel Research Articles

Application of Intelligent Technology on Tunnel Steel Structure Factory

At present, the construction of tunnel steel structure factories has been common, but most of them are still on the low end. In addition, the manual processing modes face certain problems, such as high waste rate, low efficiency, and discrete welding of steel components. To address these problems, this paper introduces an intelligent management cloud platform, aiming to realize unmanned control methods and construct an intelligent tunnel steel component processing plant. There are four main production lines in the intelligent steel structure factory. First, an intelligent production line of a tunnel grille arch effectively improves the production efficiency, production quality, and dust removal effect of the grille arch through the automatic welding technology of a reinforcement mechanism-based welding robot. Second, a processing line of a steel pipe for tunnel construction improves the overall processing efficiency, saves costs, and reduces work intensity by using an intelligent numerical control system, a pneumatic clamping device, a plasma cutting hole assembly, and a 360-degree automatic rotation system. Third, an automatic feeding platform of section steel arch production line adopts the six-axis robot sensor clamping connecting plate positioning and six-axis robot sensing three-plane automatic welding system, which not only compensates for the defects of manual production but also realizes the reduction, increases efficiency, and improves welding quality and precision. Fourth, an automatic steel mesh production line saves much labor and enhances site management and production efficiency through vertical and horizontal reinforcement pay-off racks. Overall, in this study, the construction goals of informatization, intellectualization, and personnel downsizing of the tunnel steel structure processing have been achieved.

Development and Application of Standard Device for Calibrating Steel Measuring Tape Based on Machine Vision

Aiming to address the disadvantages of the low automation degree of the traditional steel measuring tape verification device, as well as the high labor intensity of metrologists, high time costs, and low measurement accuracy, this paper develops a new automatic steel measuring tape calibration device based on machine vision technology. In this study, the key components, such as the tape-winding mechanism, machine vision system, force measuring device, and tape-retracting device, were designed. Three different scribed line identification and location algorithms, processing the longest line, the closest to the starting line, the highest line, etc., were designed to adapt to different types of steel measuring tape. Lastly, indication error and measurement repeatability tests were carried out. The results showed that the standard device can carry out the verification and calibration of steel measuring tape, and the maximum repeatability of the indication error of the test objects was 0.025 mm, 0.038 mm, and 0.030 mm, respectively, and the indication error of each object was within the allowable error limits. The uncertainty evaluation of the indication error of each test object was 0.16 mm, 0.54 mm, 0.78 mm, which was not more than one third of the maximum permissible error value. Hence, the measuring principle, method, and procedure of the calibration device are shown to be scientific and feasible.

CASPPNet: a chained atrous spatial pyramid pooling network for steel defect detection

Automatic machine vision-based defect detection has been successfully applied to many industrial visual inspection applications. However, automatic steel surface defect detection is still a challenging task due to diverse defect categories, low-contrast between defect and complex texture background. To address these challenges, a chained atrous spatial pyramid pooling network (CASPPNet) is proposed for steel surface defect detection. In CASPPNet, chained atrous spatial pyramid pooling is designed to enlarge receptive field and obtain enrich semantic information. An improved global attention feature fusion module is introduced to achieve feature interaction and salience. Moreover, residual boundary refinement block is introduced to get more complete defect boundary. Comparative experimental results verify that our method is superior to the state-of-the-art segmentation methods on public accessible SD-saliency-900 datasets and can meet the requirement of real-time online detection (the detection efficiency is at over 47 FPS on a single GPU).

Dual attention deep learning network for automatic steel surface defect segmentation

AbstractA dual attention deep learning network is developed to classify three types of steel defects, locate their positions, and depict their shapes on the steel surface in an automatic and accurate manner. The novel pixel‐level detection algorithm called DAN‐DeepLabv3+ integrates a dual attention module into the DeepLabv3+ framework in pursue of more precise segmentation results. For one thing, the dual parallel attention module helps to explicitly model rich contextual dependencies over local feature representations in the spatial and channel dimensions. For another, the popular DeepLabv3+ in an encoder‐decoder architecture is useful in capturing multi‐scale contextual information and sharp object boundaries. The DAN‐DeepLabv3+ is applied to an available dataset containing 6666 images, where three types of steel defects are taken by high‐frequency cameras and have been annotated manually. Experimental results show that, compared with other deep learning models, DAN‐DeepLabv3+ based on the Xception backbone exhibits the best segmentation performance under the mean intersection over union (IoU) of 89.95% and the frequency‐weighted IoU of 97.34%. Besides, the F1‐score for the three kinds of defects can reach 86.90%, 99.20%, and 92.81%. From the comparative study, it has been found that the adoption of the dual attention module and DeepLabv3+ contributes to boosting the segmentation performance. The significance of the proposed hybrid model lies in the enhancement in accurately detecting single or multiple steel defects, which has proven to outperform other classical methods.

Automatic steel grades design for Jominy profile achievement through neural networks and genetic algorithms

The paper proposes an approach to the design of the chemical composition of steel, which is based on neural networks and genetic algorithms and aims at achieving a desired hardenability behavior possibly matching other constraints related to the steel production. Hardenability is a mechanical feature of steel, which is extremely relevant for a wide range of steel applications and refers to the steel capability to improve its hardness following a heat treatment. In the proposed approach, a neural-network-based predictor of the so-called Jominy hardenability profile is exploited, and an optimization problem is formulated, where the optimization function allows taking into account both the desired accuracy in meeting the target Jominy profile and other constraint. The optimization is performed through genetic algorithms. Numerical results are presented and discussed, showing the efficiency of the proposed approach together with its flexibility and easy customization with respect to the user demands and production objectives.

A deep learning approach for real-time rebar counting on the construction site based on YOLOv3 detector

Counting steel bars is a routine daily task for steel manufacturers and most building construction sites. Currently, counting is usually performed manually, which is laborious, time-consuming, and error-prone. This study proposes a deep learning approach based on YOLOv3 detector for automatic steel bars detection and counting through images. Three new measures, including an additional feature pyramid, complete intersection over union (IoU) loss and focal loss, and bag of freebies, were introduced to improve rebar detection and counting performance. A dataset containing 74,824 rebar sections from real construction sites was constructed and utilized to evaluate the proposed approach. The application results demonstrate that the suggested measures can significantly improve the performance of the YOLOv3 detector to high average precision of 99.7% at IoU of 0.5. Comparisons with other detectors show that the proposed approach is fast, accurate, and robust at rebar counting under different construction site conditions. It can be facilitated as the basis of a real-time, cost-effective rebar counting scheme.

Main Components and Neuroanalysis of Automatic Steel Water Deoxygenation Alloy Ingredients

At present, how to establish a mathematical model of deoxygenation alloying link through historical data, predict and optimize the type and quantity of alloy input online, and minimize the production cost of alloy steel while ensuring the quality of steel and water. It is an important issue for major steel companies to improve their competitiveness. Based on the main component and neural network and multivariate nonlinear regression theory, based on the existing alloy historical data, this paper optimizes the ingredients scheme of "deoxygenation and alloying" of steel and water, establishes the mathematical model, calculates the historical harvest rate of the main elements, and predicts the historical harvest rate of the main elements. Combined with the price of the alloy, the automatic ingredient scheme of the steel water deoxygenated alloy is given, and the cost of the steel water deoxygenated alloy is optimized.

An Improved GANs Model for Steel Plate Defect Detection

Automatic steel plate defect detection is very important for it can monitor the product quality. This paper makes a study on steel plate defect detection based on machine learning. The main difficult is that there is not enough data to make powerful detection models. We propose a Generative Adversarial Networks based method to generate synthetic training image. A novel structure is designed with type related variable incorporated in Generator and a classification branch added to Discriminator. With expanded dataset, two detection algorithm, Faster R-CNN and YOLO are adopted. Various model structures, optimization methods, batch sizes and model execution time are evaluated and the influence of parameters are also analyzed. The experimental results show that the proposed novel data generation method can effectively improve the model performance.

Automatic Tunnel Steel Arches Extraction Algorithm Based on 3D LiDAR Point Cloud.

Automation is an inevitable trend in the development of tunnel shotcrete machinery. Tunnel environmental perception based on 3D LiDAR point cloud has become a research hotspot. Current researches about the detection of tunnel point clouds focus on the completed tunnel with a smooth surface. However, few people have researched the automatic detection method for steel arches installed on a complex rock surface. This paper presents a novel algorithm to extract tunnel steel arches. Firstly, we propose a refined function for calibrating the tunnel axis by minimizing the density variance of the projected point cloud. Secondly, we segment the rock surface from the tunnel point cloud by using the region-growing method with the parameters obtained by analyzing the tunnel section sequence. Finally, a Directed Edge Growing (DEG) method is proposed to detect steel arches on the rock surface in the tunnel. Our experiment in the highway tunnels under construction in Changsha (China) shows that the proposed algorithm can effectively extract the points of the edge of steel arches from 3D LiDAR point cloud of the tunnel without manual assistance. The results demonstrated that the proposed algorithm achieved 92.1% of precision, 89.1% of recall, and 90.6% of the F-score.

Study of the Influence of Intermix Conditions on Steel Cleanliness

Modern steel plants produce today a large portfolio of various steel grades, many for end-uses demanding high quality. In order to utilize the maximum productivity of the continuous-casting machine, it is sometimes necessary to cast steel grades with different chemical compositions in one sequence. It is important, therefore, to know the possibilities of a specific continuous-casting machine to make the Intermix connections as short as possible. Any interference with established procedures may, however, have a negative impact on the cleanliness of the cast steel. Using physical and numerical simulation tools, it was found that reducing the steel level in the tundish during the exchange of ladles makes it possible to shorten the transition zone. However, when the steel level is reduced, the flow of steel is impaired, which can have a negative effect on the cleanliness of the cast steel and, in extreme cases, may even lead to entrapment of slag in the mold. The cleanliness of cast steel was evaluated using one of the most advanced tools for automatic steel cleanliness evaluation, AZtecFeature (Oxford Instruments, Abingdon, UK), which enables determination of the type, size, distribution, and shape, as well as the chemical composition, of individual types of non-metal inclusions.

Problems of blackening of steel parts in technical practice

Problems of degradation are mainly solved in finished products, in a number of cases it is manifested during production. A great influence on the degradation processes of the products has the surface quality and surface finish. The article deals with the problem of blackening of steel, which has been treated with alkaline blackening. The throttle shaft made of automatic steel will be analyzed. Degradation was manifested by a change in surface quality after alkaline blackening. The methods of light and electron microscopy in combination with complex analysis of production flow and technology conditions were used for analysis. Based on the results of analyzes it was found that the degradation processes are closely related to the quality of preparation of the surface. Owing to the incorrect pretreatment of the surface, oxides occurred locally. At the sites of oxides, after alkaline blackening, the surface layer was peeled off, resulting in deterioration in the quality of the product.

Review of vision-based steel surface inspection systems

Steel is the material of choice for a large number and very diverse industrial applications. Surface qualities along with other properties are the most important quality parameters, particularly for flat-rolled steel products. Traditional manual surface inspection procedures are awfully inadequate to ensure guaranteed quality-free surface. To ensure stringent requirements of customers, automated vision-based steel surface inspection techniques have been found to be very effective and popular during the last two decades. Considering its importance, this paper attempts to make the first formal review of state-of-art of vision-based defect detection and classification of steel surfaces as they are produced from steel mills. It is observed that majority of research work has been undertaken for cold steel strip surfaces which is most sensitive to customers' requirements. Work on surface defect detection of hot strips and bars/rods has also shown signs of increase during the last 10 years. The review covers overall aspects of automatic steel surface defect detection and classification systems using vision-based techniques. Attentions have also been drawn to reported success rates along with issues related to real-time operational aspects.

Structure Decomposition and Homomorphism Identification of Planar Variable Topology Mechanisms

Structural analysis of variable topology mechanisms (VTMs) is the leading task when studying the topological variability of mechanisms. Of several major concerns, structure decomposition and homomorphism identification are two dominating issues for the structural analysis of VTMs. This paper presents a systematic computational approach for the structure decomposition and homomorphism identification of planar VTMs. Along with the proposed method, a constraint matrix representation, that records the potential motion constraints and the topological structures of a VTM, is introduced for serving as the basis of the approach. In addition, a new index, namely, degrees of homomorphism (DOHs), is suggested for quantifying the topological similarity among VTMs. For illustration, an automatic steel clamping and sawing mechanism and a group of mechanisms with similar topologies are adopted, from which their structure decomposition and homomorphism identification are carried out. As shown, the method is both symbolically readable and computationally considerable. The result is helpful for the automated structural analysis and synthesis of variable topology mechanisms.

Study on Properties and Inclusion of Environmental Automatic Steel Alloyed with Tin Element

A new environment-friendly automatic steel alloyed with tin element Y20Sn has been developed. In test, new steel possess good machinability and mechanical property, which reach Chinese National Standard of corresponding automatic steel GB/T8731-88. In steel, the shape of inclusions is mainly spherical or spindly, which volume is small and distribution is evenly. Such style of inclusion can fairly improve the steel machinability, and doesnt impair its mechanical property at the same time. As the main inclusion in developed automatic steel, MnS inclusion is one important factor which influences the machinability of steel. In the steel alloyed with tin, tin exist around MnS inclusions or are attached to MnS inclusion will influence the formation mechanism of MnS inclusion, that is the important reason why tin element influence machinability of developed automatic steel Y20Sn.

The Application of Image Recognition Based on BP Neural Networks in Automatic Steel Rolling

A BP neural networks is presented for billet character recognition. Firstly, by a series of image processing techniques, the character’feature in the billet character region of the video image gathered by frame grabber is abstracted. Secondly, the BP neural networks algorithm is employed for character recognition. Application results show that the image recognition based BP neural networks can performs well in billet character recognition, and the method presented is speedy, efficient and of high value in practice.

The Analysis of the Automatic Steel Drum Tilter

In order to meet the requirements of a steel drum manufacturing plant on steel drums flip using traditional design methods design the steel drums flipping machine. The finite element analysis method to flip the frame and the strength of the drive shaft. Flip frame analysis results obtained displacement, the stress distribution diagram, the maximum stress of 41.361MPa, and the stress concentration point to flip the frame and at the connection of the link plate. Analysis the shaft torsional displacement, stress and its former fifth-order natural frequencies and mode shapes, after optimization analysis of the shaft, the quality of the drive shaft is reduced by about 13.49%, and to identify the best hollow radius R1 = 15.035mm.

Signal Acquisition Analysis in Hi-Speed and Hi-Precision MFL Testing for Steel Pipe

Dealing with the relationship properly between the sensor scanning and signal acquisition is the base of hi-speed and hi-precision MFL (magnetic flux leakage) testing for steel pipe. Firstly, the MFL wave form characteristic was established using magnetic dipole theory. Further, on that basis, the relationship between signal frequency and sensor scanning was analyzed. Finally, sample frequency was designed according to the requirement of the automatic steel pipe MFL testing. Additionally, the MFL signal acquisition experiment was conducted to verify the influence of the signal sampling frequency. The signal acquisition analysis was of great significant to perform the MFL testing for steel pipe in hi-speed and hi-precision.

Study on Novel Automatic Steel Wire Bundling Machine Hydraulic Control System

This paper introduces a new type of automatic steel wire bundling machine for bundling process, which includes a hydraulic action process, mainly do some researches on the hydraulic control system. The system chooses PLC as the core control component, puts forward the hardware of control system and control flow. Eventually we have been designed the control program.

Study on Novel Automatic Steel Bundling Machine Pneumatic Control System

This paper introduces a new type of automatic steel bundling machine for bundling process, which includes a pneumatic action process, mainly do some researches on the pneumatic control system. The system chooses PLC as the core control component, puts forward the hardware of control system and control flow. Eventually we have been designed the control program.

Analysis of an Automatic Steel-teeming Method Using Electromagnetic Induction Heating in Slide Gate System

A new method (electromagnetic steel-teeming method) using electromagnetic induction heating in slide gate system was proposed to overcome the pollution of traditional nozzle sand on the molten steel and achieve 100% automatic steel teeming. By Joule heat, this new process is to melt part or the whole of a new ladle well-packing materials (i.e. Fe–C alloy particles with similar composition as the molten steel), which is the substitute for traditional nozzle sand. A numerical simulation method based on the integration of both electromagnetic and thermal analysis modules of ANSYS was employed to investigate the outside surface temperature change of the Fe–C alloy. In addition, the effects of the coil structural parameters (the distance between the molten steel and the coil, the length and the diameter of the coil) and electric current parameters (ampere-turns, frequency) on teeming time were investigated. Under the condition of the optimal parameters found according to simulation, the smoothly steel-teeming using induction heating was achieved. The safety of the steel shell under the action of electromagnetic induction heating with the optimal parameters was also checked via the distribution of temperature field on the shell.