Rolled Steel Plates Research Articles

Steel plate surface defect classification technology based on image enhancement and combination feature extraction

PurposeIn the process of continuous casting and rolling of steel plate, due to the influence of rolling equipment and process, there are scratches, inclusions, patches, scabs and pitted surfaces on the surface of steel plate, which will not only affect the corrosion resistance, wear resistance and fatigue strength of steel plate but also may cause production accidents. Therefore, the detection of steel plate surface defect must be strengthened to ensure the production quality of steel plate and the smooth development of industrial construction.Design/methodology/approach(1) A steel plate surface defect detection technology based on small datasets is proposed, which can detect multiple surface defects and fill in the blank of scab defect detection. (2) A detection system based on intelligent recognition technology is built. The steel plate images are collected by the front-end monitoring device, then transmitted to the back-end monitoring center and processed by the embedded intelligent algorithms. (3) In order to reduce the impact of external light on the image, an improved Multi-Scale Retinex (MSR) enhancement algorithm based on adaptive weight calculation is proposed, which lays the foundation for subsequent object segmentation and feature extraction. (4) According to the different factors such as the cause and shape, the texture and shape features are combined to classify different defects on the steel plate surface. The defect classification model is constructed and the classification results are recorded and stored, which has certain application value in the field of steel plate surface defect detection. (5) The practicability and effectiveness of the proposed method are verified by comparison with other methods, and the field running tests are conducted based on the equipment commissioning field of China Heavy Machinery Institute.FindingsWhen applied to small dataset, the precision of the proposed method is 94.5% and the time is 23.7 ms. In order to compare with deep learning technology, after expanding the image dataset, the precision and detection time of this paper are 0.948 and 24.2 ms, respectively. The proposed method is superior to other traditional image processing and deep learning methods. And the field recognition precision is 91.7%.Originality/valueIn brief, the steel plate surface defect detection technology based on computer vision is effective, but the previous attempts and methods are not comprehensive and the accuracy and detection speed need to be improved. Therefore, a more practical and comprehensive technology is developed in this paper. The main contributions are as follows: (1) A steel plate surface defect detection technology based on small datasets is proposed, which can detect multiple surface defects and fill in the blank of scab defect detection. (2) A detection system based on intelligent recognition technology is built. The steel plate images are collected by the front-end monitoring device, then transmitted to the back-end monitoring center and processed by the embedded intelligent algorithms. (3) In order to reduce the impact of external light on the image, an improved MSR enhancement algorithm based on adaptive weight calculation is proposed, which lays the foundation for subsequent object segmentation and feature extraction. (4) According to the different factors such as the cause and shape, the texture and shape features are combined to classify different defects on the steel plate surface. The defect classification model is constructed and the classification results are recorded and stored, which has certain application value in the field of steel plate surface defect detection. (5) The practicability and effectiveness of the proposed method are verified by comparison with other methods, and the field running tests are conducted based on the equipment commissioning field of China Heavy Machinery Institute.

Ultrasonic examination of a thin, textured metal plate with Penrose tile structure consisting of pitched indentations

A feasibility study is presented on the experimental application of ultrasound to examine rolled stainless steel plates having equidistant surface textures in two directions in the form of Penrose tiles. The specific problem of interest is investigating surface profile quality in terms of its equidistance and depth to monitor the manufacturing process. The goal is to eventually replace current time-consuming optical examination procedures with a reliable and rapid ultrasonic inspection technique. Two practical experimental setups are discussed and compared in this work: examining frequency spectra obtained from normal incidence pulse-echo measurements, and those obtained at Laue angle incidence. A thorough survey of ultrasonic methods precedes the experimental results to investigate such surfaces from a historical perspective.

Online Prediction of Mechanical Properties of the Hot Rolled Steel Plate Using Time-series Deep Neural Network

The prediction of steel’s properties by the process parameters has been of great interest because they can effectively reduce production costs and improve product quality. At present, the extensively used artificial neural network (ANN) can only reveal the correlation between parameters and mechanical properties from the perspective of statistics but loses the critical information on time-series correlation in the steel production process. In this work, time-series neural networks based on long short-term memory (LSTM) were established to predict the steel plate’s yield strength (YS), ultimate tensile strength (UTS), and elongation (EL). The results verified that the proposed LSTM model exhibited sufficient accuracy and outperformed the classical algorithms (SVM, random forest, ANN), with mean squared error reduced by 30% when compared with ANN. Also, the sensitivity analysis proved that LSTM was more capable to make full use of the information contained in input parameters and achieved better generalization performance. Notably, the interpretation of the results was more consistent with the real physical metallurgical process of a hot-rolled steel plate and can provide a better understanding of the physical metallurgical process.

Mechanical behavior of additively and conventionally manufactured 316L stainless steel plates joined by gas metal arc welding

Combining several additive manufactured (AM) parts to larger parts by welding may be required due the limited building volume of powder bed AM methods. Laser powder bed fusion (LPBF) has a great potential because it enables the production of nearly full-density components through AM processes; however, additional residual stresses and production defects are induced by LPBF. These residual stresses affect the residual stress state of welded AM parts. In combination with the production related defects, both alter the mechanical and—in particular—the fatigue behavior of these welded joints. In this study, various tests are performed to characterize the butt joints of 316L AM steel plates made by gas metal arc welding. To this goal, joints are produced with weld seams parallel and vertical to the layer orientation of AM plates. The results are compared to joints of conventionally rolled steel plates produced with the same welding parameter. The residual stress states in initial (unloaded) condition and after cyclic loading were determined by X-ray diffraction techniques for AM and rolled plates. Complex residual stress states were determined at the welds made of AM steel plates compared to the welds made of rolled steel plates; however, the residual stress level in the heat affected zone of the butt-welded AM steel plates was similar to the welds made of hot-rolled steel plates. After cyclic loading with a high load level, high residual stress relaxations were observed in the parent materials. The fatigue design curve for butt joints from international standards is exceeded by all three test series, but the fatigue strength of the butt joints made by LBPF and hot rolling vary significantly. This is thought to be related to differences between the AM and conventional joints in microstructure, static strength, residual stress level, and small crack-like defects that partially interact with stress concentrations at the weld transition.

The impact of annealing on microstructure of cold rolled steel plates detected by eddy current method

ABSTRACT Eddy current (EC) is a non-destructive method, which plays an important role in the quality inspection sector of various metal industries. It has been used for a wide range of inspection purposes, such as detecting cracks and defects, and inspecting the coating quality. The EC method can also be utilised in the material and microstructure characterisation of engineering steels and alloys. In this study, the EC method is used to detect the microstructural changes during the annealing of a cold rolled steel plate. To support the EC findings, the electromagnetic results are compared with the hardness measurements and the microstructure observations. The comparison proves that the EC method is able to separate recovery and recrystallisation stages, and extract useful information from the changes in the microstructure at various recrystallisation levels.

Online magnetic flux leakage detection of inclusions and inhomogeneities in cold rolled steel plate

Abstract Cold rolled steel plate is found in different applications. In manufacturing processes with high stretch ratios like bending or deep drawing, non-metallic inclusions, or segregations, which remain in the material during slab casting, can lead to ruptures of the plate. While not being visible optically, those inclusions are detectable by magnetic flux leakage inspection. Therefore, an online inspection system using giant magnetoresistance gradiometers was developed. The individual components were optimized by researching the magnetic properties of the cold rolled plate and natural as well as artificial defects on a laboratory setup for moving material simulation. Correlations between defect properties, inspection parameters and the resulting signals were determined. By optimizing components and configuration accordingly, a sensor module containing 48 sensors, signal conditioning, digitization, and an adjustable magnet was arrived at. Due to the necessarily small distance to the moving material, attention was paid to mechanical repeatability and ruggedness. An inclusion detection system, fully covering 1344 mm strip width, was installed in a coating line. Operating in production for over a year, the inspection system has shown high sensitivity and is, in addition to internal defects, able to detect various strip damages like roll marks and scratches.

Investigations on the lateral-torsional buckling of an innovative U-shaped steel beam

An innovative solution of a steel-concrete composite beam was developed for buildings with beam spans between 6 and 12 m, taking into consideration the fire situation and the construction stage. The beam is composed of an U-shaped steel part, connected to a reinforced concrete part. In the construction phase, the beam is supporting the slab and constitutes a formwork for the reinforced concrete part. When casting concrete, the steel beam is filled at the same time as the slab, this allows considerable time savings on site. In the exploitation stage, the beam behaves as a steel-concrete composite beam.The developed U-shaped section is composed of three steel parts; a hot rolled steel plate (180x8x6300 mm, and two cold-formed steel plates (470x4x6300 mm) before folding. The section was assembled together with four self-drilling screws longitudinally spaced of 150 mm.In the construction stage, the U-shaped steel beam without restraints is prone to lateral torsional buckling instability. In this paper, the mechanical behaviour of the steel beam at the construction stage, in particular its lateral-torsional buckling stability, was studied. In order to characterize this instability without lateral support, a four-point bending test was carried out on a life-size specimen. The load introduction system was developed especially for the test to allow for all displacements due to LTB. The test clearly showed the lateral torsional buckling behaviour of the steel beam. The test results are compared with numerical simulations and analytical studies. A parametrical study, covering 200 geometrical configurations of the U-shaped beam, is carried out to validate the use of the curve “b” for the design of the steel beam for lateral torsional buckling according to Eurocodes 3.

Microstructure and properties of Nb, V and N alloyed low magnetic hot rolled stainless steel plate

Microstructure and properties of Nb, V and N alloyed low magnetic hot rolled stainless steel plate

Usage of Converter Gas as a Substitute Fuel for a Tunnel Furnace in Steelworks.

Converter gas (BOFG) is a by-product of the steel manufacturing process in steelworks. Its usage as a substitute fuel instead of natural gas for fueling a metallurgical furnace seems to be reasonable due to potential benefits as follows: CO2 emission reduction into the ambient air and savings in purchasing costs of natural gas. Results of theoretical analysis focused on implementing converter gas as a fuel for feeding a tunnel furnace for either steel plate rolling, steel sheet hardening in its real working condition or both, are discussed. The analysis was focused on the combustion chemistry of the converter gas and its potential ecological and economic benefits obtained from converter gas usage to heat up steel in a tunnel furnace. Simulations of combustion were conducted using a skeletal chemical kinetic mechanism by Konnov. The directed relation graph with error propagation aided sensitivity analysis (DRGEPSA) method was used to obtain this skeletal kinetic mechanism. Finally, the model was validated on a real tunnel furnace fueled by natural gas. Regarding exhaust emissions, it was found that nitric oxide (NO) dropped down from 275 to 80 ppm when natural gas was replaced by converter gas. However, carbon dioxide emissions increased more than three times in this case, but there is no possibility of eliminating carbon dioxide from steel manufacturing processes at all. Economic analysis showed savings of 44% in fuel purchase costs when natural gas was replaced by converter gas. Summing up, the potential benefits resulting from substituting natural gas with converter gas led to the conclusion that converter gas is strongly recommended as fuel for a tunnel furnace in the steel manufacturing process. Practical application requires testing gas burners in terms of their efficiency, which should provide the same amount of energy supplied to the furnace when fed with converter gas.

Stochastic model describing evolution of microstructural parameters during hot rolling of steel plates and strips

Enhancing strength-ductility synergy of materials has been for decades an objective of research on structural metallic materials. It has been shown by many researchers that significant improvement of this synergy can be obtained by tailoring heterogeneous multiphase microstructures. Since large gradients of properties in these microstructures cause a decrease of the local fracture resistance, the objective of research is to obtain smoother gradients of properties by control of the manufacturing process. Advanced material models are needed to design such microstructures with smooth gradients. These models should supply information about distributions of various microstructural features, instead of their average values. Models based on stochastic internal variables meet this requirement. Our objective was to account for the random character of the recrystallization and to transfer this randomness into equations describing the evolution of dislocations and grain size during hot deformation and during interpass times. The idea of this stochastic model is described in the paper. Experiments composed of uniaxial compression tests were performed to supply data for the identification and verification of the model in the hot deformation and static recrystallization parts. Histograms of the grain size were measured after hot deformation and at different times after the end of deformation. Identification and validation of the model were performed. The validated model, which predicts evolution of heterogeneous multiphase microstructure, is the main output of our work. The model was implemented in the finite element program for hot rolling of plates and sheets and simulations of these processes were performed. The model’s capability to compare and evaluate various rolling strategies are demonstrated in the paper.

Non-Uniform Deformation of Slab during Heavy Reduction Process

The bulge deformation of slab narrow face can cause surface defects of hot rolled steel plate. A three-dimensional bulging model, was proposed to simulate the evolution of deformation behavior of the continuous casting slab during heavy reduction (HR). The model was taken to investigate the non-uniform deformation of slab during HR process. The bulging deformation behavior of the slab was then calculated in one segment included seven pairs of rollers. To improve the edge defect on hot-rolled steel plates, the relationship between the reduction amount and bulge deformation of slab narrow face has been investigated.

Anisotropy in electrochemical jet texturing of rolled stainless steel SUS304

The influence of the processing method of bulk material on the resultant workpiece surface microstructure in electrochemical jet texturing (EJT) is presented. A rolled stainless steel plate SUS304 workpiece is subjected to EJT for texturing by applying various current densities and jet kinematics. The resultant surface topography, removal characteristics, surface chemistry, and texture pattern are carefully investigated considering material microstructures. Experimental results reveal evident anisotropic behavior of texture which depends on the surface orientation with respect to the rolling direction. Periodic microstructures of aligned ripples are exclusively produced on surfaces parallel to the rolling direction, demonstrating the considerable influence of the rolling process on the EJT. Further, it is clarified by microstructural characterizations that the anisotropic texturing is ascribed to segregation bands of alloying elements resulting from the plastic deformation of the rolling process.

Quenching cracks in medium carbon steel initiated at manganese sulfide inclusions

Abstract Quench crack propensity has been investigated for two identical medium carbon steels (50CrMo4) which only differed in sulfur content. The standard variant of the material was affected by manganese sulfide inclusions whereas the low sulfur material was essentially free of sulfides. 80 mm, cylindrical test specimens were used. The test specimens had been cut out of cross rolled steel plates in which the manganese sulfide were flattened to discoid shape. The specimen orientation was in direction of principal deformation where the flattened sulfides were aligned with the test specimen axis. The largest sulfides reached lengths of 150 μm. Both materials were austenitized at 850 °C for 60 minutes, quenched in a salt-bath at 180 °C and immediately annealed at 425 °C for one hour. It turned out that the standard material developed quench cracks which initiated at manganese sulfide. No cracks could be found in the low sulfur material. Thermal and transformational stresses, which develop during quenching, aggravate the stress situation around the manganese sulfide and cause the cracking. The results are interpreted in fracture mechanical terms.

The effect of cooling transients on tearing resistance of a rolled steel

The fracture toughness in the upper shelf area is determined with J-R curves. Presently, there is no data, nor predictions, regarding the J-R curve development during a decreasing thermal transient. In this study, the effect of rapid cooling on J-R curves is investigated. Fracture toughness specimens of a rolled steel plate S460 were cooled from 300°C to the room temperature at a cooling rate of 2°C/s determined at the center location of the specimen. The cooling rate was selected to be in the same range as during a loss of coolant accident in a nuclear power plant. The following was concluded: 1.) No unstable brittle fracture in the upper shelf region is observed for the investigated material during the cooling transient. 2.) Based on the results, when doing structural integrity analyzes of thermal transients in the upper shelf region, it is recommendable to use the lowest fracture toughness properties in the investigated temperature range, if no other information is available.

Anchorage performance of embedments in anchored prefabricated wall–beam joints of BLC–CCWS

The existing wall–beam joints of bundled lipped channel–concrete composite wall structures (BLC–CCWS) are mainly fabricated by welding thin–walled steel tubes and hot–rolled steel plates, and this application requires further improvement. An innovative anchored prefabricated wall–beam joint that does not require field welding was presented in this paper, and the anchorage performance of the embedment, which is the key force transfer component of the joint, was investigated by experimental research and theoretical analysis. The results indicated that (1) the embedments have good anchorage performance in the “bundled lipped channel–concrete” anchorage area, especially the “steel–stud–anchor plate” type of embedment, which clearly demonstrates a mechanical property evolution stage. (2) Embedment construction has a great influence on the anchorage performance. The ultimate bearing capacity, ultimate slip, ductility and energy storage/dissipation capacity of the “steel–stud–anchor plate” embedment are better than those of the “steel–stud” embedment. (3) The construction of the anchorage area has a great influence on the anchorage performance. The bundled lipped channel can substantially limit the range of damage to concrete. In addition, the anchorage performance of the embedment in the “bundled lipped channel–concrete” anchorage area is better than that of the “circumferential steel plate–concrete” anchorage area in terms of the ultimate bearing capacity, ultimate slip, ductility and energy dissipation capacity. (4) The established formula for calculating the anchorage bearing capacity of the embedment is accurate and conservative, and thus can meet the requirements of engineering practice.

Effect of copper on edge cracking behavior and microstructure of rolled austenitic stainless steel plate

Effect of copper on edge cracking behavior and microstructure of rolled austenitic stainless steel plate

Online prediction of mechanical properties of hot rolled steel plate using machine learning

In industrial steel plate production, process parameters and steel grade composition significantly influence the microstructure and mechanical properties of the steel produced. But determining the exact relationship between process parameters and mechanical properties is a challenging process. This work aimed to devise a deep learning model, to predict mechanical properties of industrial steel plate including yield strength (YS), ultimate tensile strength (UTS), elongation (EL), and impact energy (Akv); based on the process parameters as well as composition of raw steel, and apply it online to a real steel manufacturing plant. An optimal deep neural network (DNN) model was formulated with 27 inputs parameters, 2 hidden layers each having 200 nodes and 4 output parameters (27 × 200 × 200 × 4) with an initial learning rate 0.0001, using Adam optimizer and subjected to Z pre-processing method, to yield an accurate model with R2 = 0.907. The tuned DNN model, had a root mean square error of 21.06 MPa, 16.67 MPa, 2.36%, and 39.33 J, and root mean square percentage error of 4.7%, 2.9%, 7.7%, and 16.2%, for YS, UTS, EL and Akv respectively. Through comparative analysis, it was found that the accuracy of DNN model was higher than other classic machine learning algorithms. To interpret the model assumptions and findings, several local linear models were devised and analyzed to establish the link between process parameters and mechanical properties. Finally the tuned DNN model was deployed in the real-steel plant for online monitoring and control of steel mechanical properties, and to guide the production of targeted steel plates with tailored mechanical properties.

Microstructure, mechanical, ballistic property evaluation of RHA steel produced by continuous-casting route

This paper presents the commercial production of rolled homogeneous armour steel with 5 different thicknesses i.e. 20,30,40,50 and 80 mm through continuous casting route. All the plates display tempered martensitic structure. Tensile and charpy impact properties are obtained for all the different thickness plates. It is observed that the produced steel shows a good combination of strength and impact toughness. For ballistic evaluation 30 mm thick plates are impacted with 30 mm medium caliber armour piercing steel projectiles at a velocity of 460±20 m/s at zero degree angle of impact. Depth of penetration method is used to measure the ballistic performance of the plates. From the post ballistic microstructural observations, an adiabatic shear band induced material removal is detected in the front face of the impacted rolled homogeneous armour steel plates. The microstructure, mechanical and ballistic properties of the continuous cast steel is compared with the steel produced by conventional ingot cast route.

Determination of temperature conditions for steel plate rolling at Vyksa Steel Works (AO VMZ)

This article describes the working process of OMK Heavy Plate Mill 5000. Based on the fabricator's facilities capability, the technological production flow diagram of thick steel sheets of K60 (X 70) grades, for cold-working large diameter pipes has been developed. As the initial blank authors used a 310×2.100×3.170 mm plate slab. The controlled rolling process, described in the article, consists of two stages –roughing-down (when deformation and re-crystallization results in obtaining fine uniform austenite grains) and finishing rolling (when austenite is hammer-hardened and converted into fine-grain ferrite). As a result of the research has been selected and the optimal temperature conditions for rolling in the passes, which would reduce the waste of metal in furnace.

Prediction for Manufacturing Factors in a Steel Plate Rolling Smart Factory Using Data Clustering-Based Machine Learning

A Steel Plate Rolling Mill (SPM) is a milling machine that uses rollers to press hot slab inputs to produce ferrous or non-ferrous metal plates. To produce high-quality steel plates, it is important to precisely detect and sense values of manufacturing factors including plate thickness and roll force in each rolling pass. For example, the estimation or prediction of the in-process thickness is utilized to select the control values (e.g., roll gap) in the next pass of rolling. However, adverse manufacturing conditions can interfere with accurate detection for such manufacturing factors. Although the state-of-the-art gamma-ray camera can be used for measuring the thickness, the outputs from it are influenced by adverse manufacturing conditions such as the high temperature of plates, followed by the evaporation of lubricant water. Thus, it is inevitable that there is noise in the thickness estimation. Furthermore, installing such thickness measurements for each passing step is costly. The precision of the thickness estimation, therefore, significantly affects the cost and quality of the final product. In this paper, we present machine learning (ML) technologies and models that can be used to predict the in-process thickness in the SPM operation, so that the measurement cost for the in-process thickness can be significantly reduced and high-quality steel plate production can be possible. To do so, we investigate most-known technologies in this application. In particular, Data Clustering based Machine Learning (DC-ML), combining clustering algorithms and supervised learning algorithms, is introduced. To evaluate DC-ML, two experiments are conducted and show that DC-ML is well suited to the prediction problems in the SPM operation. In addition, the source code of DC-ML is provided for the future study of machine learning researchers.