386 results about "Steel quality" patented technology

The invention discloses a steel fireproof door plate polishing device, which belongs to the technical field of fireproof door plate processing, and comprises a rack, rotating rollers uniformly and rotatably mounted along the length direction of the rack, and a housing mounted at one end of the top of the rack, conveying mechanisms are arranged at two ends in the housing, and second hydraulic cylinders are symmetrically and vertically mounted on two sides of the top in the housing. A reciprocating mechanism composed of a first shaft sleeve, a first spline shaft, a reciprocating lead screw, a second sliding block, a connecting rod and a first belt wheel assembly is arranged at the end of the grinding roller, in the process that the grinding motor is used for controlling the grinding roller to rotate, the grinding roller can be controlled to reciprocate in the axial direction, and axial movement and rotation are combined; a crossed net-shaped grinding track is formed, covering is more comprehensive, surface textures are more uniform, subsequent polishing procedures are reduced, spraying or assembling requirements are directly met, in addition, the grinding roller axially moves to enable dust to be distributed more dispersedly, the dust concentration of a working area is greatly reduced in cooperation with a negative pressure dust suction mechanism, and potential safety hazards are reduced.

The present application relates to the technical field of metallurgy, and particularly relates to a production method of a large-wall-thickness pipeline steel with improved carbon dioxide corrosion resistance and strength and toughness, and specific measures include: (1) controlling the bottom blowing mode and end point requirement in the converter process; (2) controlling the tapping time and replacing the tapping hole opportunity in the tapping process; (3) feeding the CAS aluminum wire and controlling the CAS argon blowing time; (4) controlling the top slag addition amount, heating time and outgoing mode in the LF refining process; (5) controlling the wire feeding parameters, soft blowing time and outgoing requirement of RH in the RH refining process; (6) controlling the precision requirement, casting speed fluctuation and crystallizer liquid level fluctuation of the continuous casting equipment in the continuous casting process to obtain the casting blank. Through controlling the parameters of each process in the smelting and rolling process, the present application realizes the whole-process steel cleanliness control and ladle grain refinement, and further improves the strength, low-temperature toughness, HIC resistance and carbon dioxide corrosion resistance of the large-wall-thickness pipeline steel.

The invention provides a steel surface defect detection and identification system and method based on a deep learning YOLOv8 model, and the system comprises an industrial camera which is disposed in a to-be-detected region of a steel production line and is provided with a high-brightness light source. The system is characterized by further comprising an image acquisition module, a fault detection module, a defect identification module and a man-machine interaction module which run on a computer and are connected with the industrial camera control system, and the method comprises the following steps: S1, system initialization, S2, image acquisition, S3, image preprocessing, S4, feature extraction and fusion, S5, defect identification and classification, and S6, result output and interaction. Automatic and precise steel surface defect detection and recognition of the whole process from image acquisition, defect detection to result output are achieved, manual visual judgment is not needed, detection standards are unified, the problem of unstable detection caused by manual experience differences is avoided, the labor intensity of operators is reduced, enterprise losses caused by defective steel circulation are avoided, and the production efficiency is improved. Reliable data support is provided for production process optimization, and the steel quality is guaranteed.

The invention relates to an automatic control method for electromagnetic stirring parameters of a slab continuous casting crystallizer, which comprises the following steps of: 1, selecting a proper water gap immersion depth according to the condition of a continuous casting production field, 2, networking to obtain technological parameters such as a casting blank section, a casting speed, the water gap immersion depth and an argon blowing flow, and 3, automatically controlling and calculating the electromagnetic stirring parameters. The method comprises the following steps: step 1, calculating the most preferential stirring parameter under the current working condition through table look-up and interpolation calculation, step 4, outputting an electromagnetic stirring current parameter, step 5, executing the output electromagnetic stirring parameter, step 6, tracking the quality of a continuous casting slab product, repeating the steps 1-6 at regular time, and repeating the steps 2-6 under the condition that the step 1 is not changed. According to the scheme, casting blank surface defects are effectively removed, and the product quality is improved. After the optimized method is adopted, the occurrence rate of longitudinal cracks on the surface of the steel grade is reduced to 1.96% from 4.73%, and the defect rate of the steel grade is reduced to 3.93% from 8.41%.

The invention provides constructional engineering reinforcing steel bar quality detection equipment, and relates to the technical field of reinforcing steel bar quality detection.The constructional engineering reinforcing steel bar quality detection equipment comprises a detection table, clamping assemblies are arranged at the two ends of the top of the detection table, each clamping assembly comprises a vertical plate, the vertical plates are fixed to the two ends of the top of the detection table, and shaft rods are rotationally installed on the inner side surfaces of the vertical plates through bearings; a driving motor is fixed to the outer end of a vertical plate on one side of the detection table, the output end of the driving motor is fixedly connected with the shaft rod, two limiting assemblies are symmetrically arranged on the surface of the middle of the detection table, each limiting assembly comprises a base, and the bases are fixed to the detection table. Two steel bars of the same specification can be fixed at the same time through the clamping assembly and the limiting assembly on the detection table and are used for bending resistance detection and tensile test respectively, independent tests of the two steel bars do not affect each other, the time for replacing the steel bars is saved, and the quality detection efficiency of the steel bars of the same specification is higher.

The present application relates to steel structure erection technical field, concretely is a kind of steel beam jacking equipment for steel structure factory building assembly.It includes mainframe, adjusting assembly, adjusting track group;The adjusting assembly of the present application can effectively fix steel purlin and support it on roof, and the adjusting assembly of the present application can adjust the front and back positions of steel purlin, can rotate steel purlin by a certain angle to avoid interference with roof frame during jacking process, can also tilt purlin in a row by a certain angle to adapt to the angle of roof frame, under the support of adjusting assembly, workers only need to adjust purlin to installation position to start installation, and can install multiple purlin at a time, it is convenient and fast, greatly improve installation efficiency;The adjusting track group of the present application can use the support column of original building as reference to accurately position the whole system;The adjusting assembly of the present application can move front and back purlin along its length direction, make its head and tail connection without gap, to ensure its installation accuracy.

Disclosed in the present invention is a production and preparation method for a steel bar for a steam turbine impeller. The steel bar comprises the following chemical components in percentages by mass: C: 0.31-0.36%, Si: 0.22-0.35%, Mn: 0.45-0.60%, P: ≤0.013%, S: ≤0.008%, Cr: 1.05-1.20%, Ni: ≤0.20%, Cu: ≤0.10%, Mo: 0.21-0.28%, and V: 0.11-0.18%. The method comprises: converter smelting → LF refining → RH refining → continuous casting → a heating furnace heating process → a rolling process → a heat preservation process. The present invention satisfies the quality requirements of a steel bar for a steam turbine impeller, and indexes, such as mechanical properties, a high hardness and a high steel purity, can also satisfy high-standard requirements.

The application discloses a high-phosphorus molten iron extremely-low-sulfur and low-phosphorus hydrogen-doped pipeline steel steelmaking process control method, which reduces the burden of LF desulfurization through post-furnace slag system treatment, effectively improves the refining rate, improves the production efficiency and guarantees continuous casting through over-treatment or re-treatment of LF refining; compared with conventional production, the desulfurization efficiency is improved through converter slag system and LF refining slagging process, the cleanliness of the steel is also greatly improved, and the final total oxygen content can be stably controlled within 0.0015%. The application aims to provide a steelmaking process control method for realizing low-phosphorus and extremely-low-sulfur chemical composition in the process of smelting hydrogen-doped pipeline steel from high-phosphorus molten iron, especially for smelting low-phosphorus steel and extremely-low-sulfur steel from high-phosphorus molten iron, the cleanliness of the steel is also greatly improved, and the final total oxygen content can be stably controlled within 0.0015%.

The invention discloses a multi-objective optimization design method for a PBL shear connector based on machine learning, and the method is characterized in that the method comprises the following steps: collecting the shear bearing capacity of a plurality of PBL shear connectors containing input parameters and output parameters, and deducing test data; firstly, a Catboost machine learning model is trained by utilizing the deduced test data, then the trained Catboost machine learning model is adopted to predict the shear bearing capacity of the PBL shear connector, the sensitivity of input parameters is analyzed in combination with an SHAP method, and finally, a multi-target genetic algorithm is introduced into the verified Catboost machine learning model, so that the shear bearing capacity of the PBL shear connector is predicted. And a multi-objective optimization model with the minimum hole opening number and the minimum steel quality as objective functions is established, and PBL shear connector optimization design is conducted through the multi-objective optimization model. The lightweight and efficient construction of the PBL shear connector can be realized on the premise of meeting the shear bearing capacity.

The invention provides a buried steel pipeline magnetic stress quantitative detection system based on multi-physics field coupling inversion and PINN, and relates to the technical field of oil and gas pipeline structure health monitoring and integrity management. The invention discloses a buried steel pipeline magnetic stress quantitative detection system based on multi-physics field coupling inversion and a PINN. The system comprises a buried steel pipeline magnetic stress quantitative detection method based on the multi-physics field coupling inversion and the PINN. Accurate inversion from a non-contact MMM signal to continuous stress field distribution (MPa) is achieved. A lift-off height real-time measurement and compensation model is introduced, the problem of signal attenuation caused by buried depth change in buried steel pipeline detection is solved, and the detection precision and consistency are improved; according to the PINN architecture, a multi-physics field coupling model is deeply embedded into a learning process, and the problems that a pure data driving method is poor in generalization and physical interpretation are solved; according to the method, high integration of a detection result and a management system is realized, a mode of'rapid general survey + accurate review 'is realized, the detection efficiency is improved, and resource allocation is optimized.

The invention relates to a method for producing a wire converter square billet for producing an ultralow-resistance wire, which comprises the following steps of: smelting by adopting a 150t converter + LF + RH + 230mm * 1250mm * 3600mm continuous casting smelting process, and producing and manufacturing a converter process industrial pure iron slab; the produced industrial pure iron comprises the following chemical components in percentage by mass: less than or equal to 0.003% of C, less than or equal to 0.005% of Si, less than or equal to 0.030% of Mn, less than or equal to 0.005% of P, less than or equal to 0.005% of S, less than or equal to 0.005% of O, less than or equal to 0.02% of Cr, less than or equal to 0.02% of Ni, less than or equal to 0.02% of Cu, less than or equal to 0.020% of Al and less than or equal to 0.01% of Ti. According to the method, indexes such as chemical components and high steel purity of the converter process industrial pure iron plate blank are guaranteed, and the surface quality level of the converter plate blank industrial pure iron is also guaranteed.

The invention relates to a production process of an alloy structural steel bar smelted by an ecological electric furnace. The alloy structural steel bar comprises the following chemical components: 0.39%-0.44% of C, 0.17%-0.37% of Si, 0.50%-0.80% of Mn, less than or equal to 0.020% of P, less than or equal to 0.020% of S, 0.90%-1.20% of Cr, less than or equal to 0.30% of Ni, 0.15%-0.25% of Mo, 0.02%-0.03% of Nb, 0.05%-0.10% of V, less than or equal to 0.20% of Cu and the balance of Fe and impurities. The production process route comprises ecological electric furnace smelting, LF refining, RH refining, continuous casting, heating, rolling, heat preservation and heat treatment. The smelting process flow of ecological electric furnace smelting, LF refining and RH refining is adopted, the high steel purity index is achieved, and the alloy structural steel bar has the excellent comprehensive mechanical property and meanwhile has the high surface quality level.

The invention discloses a full penetration joint composite liner for a closed small space and a welding method thereof. The composite material liner comprises a structural stress steel layer A, a composite ceramic layer B and an adhesive layer C for compounding the structural stress steel layer A and the composite ceramic layer B; the composite ceramic layer B faces the joint groove and can be provided with a groove / groove for controlling the shape of a molten pool and an air hole for exhausting gas and preventing a root air hole, and the side, away from the steel layer, of the composite ceramic layer B is provided with an acidic or alkaline welding powder layer G. The problems that a traditional metal gasket is prone to beading and incomplete penetration and a traditional ceramic gasket is prone to downwarping and difficult to assemble are solved, the welding seam forming is excellent, the mechanical property is good, and full penetration welding of closed small spaces such as a bridge steel anchor beam and a steel box beam support area is adapted.

The invention relates to the technical field of special steel, in particular to a steel bar for a heavy-load automobile crankshaft and a production and preparation method of the steel bar, and the steel bar comprises the following chemical components in percentage by mass: 0.18-0.22% of C, 0.20-0.35% of Si, 0.60-0.90% of Mn, less than or equal to 0.012% of P, less than or equal to 0.005% of S, 0.40-0.60% of Cr, 0.20-0.30% of Mo, 0.15-0.20% of V, 0.01-0.02% of Nb, less than or equal to 0.15% of Cu, less than or equal to 0.25% of Ni and the balance of iron and inevitable impurities. The production and preparation method comprises the steps of converter smelting, LF refining, RH refining, continuous casting, casting blank annealing, heating, rolling process and heat treatment. According to the steel bar for the heavy-load automobile crankshaft, reasonable component design is adopted, and unique converter + LF + RH + continuous casting smelting process and rolling process are matched, so that the prepared steel bar for the heavy-load automobile crankshaft has the advantages that indexes such as mechanical property, impact property, high hardness and high steel purity of a finished steel product of the steel bar for the heavy-load automobile crankshaft are guaranteed; and the surface quality level of the steel bar for the heavy-load automobile crankshaft is guaranteed, and the requirement of a high-end product can be met.

The invention discloses a steel gas cylinder plate-type bottom collecting machine, and relates to the technical field of bottom collecting machines, the steel gas cylinder plate-type bottom collecting machine comprises a steel cylinder clamping machine, a waste box and a mounting shell, a steel cylinder main body is mounted in the steel cylinder clamping machine, a bottom collecting assembly and a scraper assembly are mounted on the side surface of the mounting shell, and the scraper assembly is correspondingly arranged below the bottom collecting assembly; the bottom collecting assembly comprises a servo motor fixedly mounted on the side face of the mounting shell, a first rotating column is rotationally connected to the interior of the mounting shell, a first rotating rod is fixedly connected to the side face of the first rotating column and extends to the side face of the mounting shell, and a bottom collecting mold is fixedly connected to the end of the first rotating rod and is made of 45 steel; according to the steel gas cylinder plate-type bottom closing machine, the bottom closing mold adopts the inlaid composite ceramic modules, so that the bottom closing machine can be used on two sides and is easy to replace, the service life is remarkably prolonged, the replacement frequency is reduced compared with that of traditional surfacing welding, and the steel gas cylinder plate-type bottom closing machine has the properties of shock resistance, crack resistance, high temperature resistance and wear resistance.

The utility model discloses a kind of steel pipeline weld spray walking mechanism, in along the inner wall or outer wall of steel pipe walking process, the position of weld is detected by the distance sensor that is located above steel pipe weld and continuously does reciprocating movement, and whether weld is located in the middle position of pre-set moving range is distinguished, then weld position information is passed to PLC, differential is generated by PLC control walking wheel, and then walking track is adjusted, guarantee the center line of walking track and the center line of steel pipe weld coincide, the walking mechanism of this steel pipeline weld spray can be realized in the automatic spraying process of anticorrosive coating to steel pipe weld, walking route is always located in the center position of weld, so that anticorrosive coating is accurately sprayed on weld, improve work efficiency.

The invention belongs to the technical field of industrial automation and intelligent manufacturing, and particularly discloses a steel analysis laboratory intelligent monitoring management system based on multi-source data fusion, which comprises a parameter monitoring module, a process judgment module, a path decision module, a deviation analysis module, a deviation propagation judgment module and a health assessment module. According to the method, the equipment operation state parameters and the material quality parameters of the steel sample in each process are collected in real time, and the preset steel quality detection standard is combined to judge whether the current process is qualified or not, so that the problem that equipment monitoring and quality detection are mutually disjointed is solved; meanwhile, the state deviation degree of the previous procedure and the real-time monitoring parameters of the subsequent procedure are subjected to fusion analysis, cross-procedure quality influence evaluation is established, then the propagation path of the quality deviation between the procedures is dynamically tracked, the process set value of the subsequent procedure is accurately adjusted based on the deviation of the previous procedure, and the quality of the subsequent procedure is improved. And beforehand intervention on the quality risk is realized.

The invention relates to a method for producing casings, in particular artillery gun and / or grenade casings, from a steel material. Also disclosed is a system for producing such casings, or a system configured to carry out the method according to the invention. The method and the system enable the efficient production of casings with a significantly shortened cycle time.

This application discloses a method, apparatus, and medium for joint prediction of multiple steel quality indicators under small sample conditions, relating to the field of material performance prediction. The method includes: determining a standardized dataset comprising steel chemical composition, process parameter characteristics, and multi-dimensional mechanical performance indicators; configuring the target order of the multi-dimensional mechanical performance indicators; constructing a regression chain model based on the target order, wherein the predicted values ​​of preceding mechanical performance indicators serve as additional input features for the regressors corresponding to subsequent mechanical performance indicators; employing a nested cross-validation strategy to search and optimize the hyperparameters of the regression chain model to determine the target hyperparameter combination; performing full retraining of the regression chain model based on the target hyperparameter combination to obtain a trained joint prediction model for steel performance; and inputting the chemical composition and production process parameters of the steel to be predicted into the joint prediction model to obtain the joint prediction results for each steel performance indicator corresponding to the steel to be predicted.

The utility model belongs to the technical field of steel cabinet processing, and discloses a steel cabinet body connected in a mortise and tenon mode, which comprises a cabinet body frame, at least one vertical partition plate and a plurality of transverse partition plates, and the at least one vertical partition plate is vertically connected with the plurality of transverse partition plates and divides the inside of the cabinet body frame into a plurality of accommodating spaces. A plurality of first splicing grooves are sequentially formed in the rear end of the vertical partition plate at intervals from top to bottom, the transverse partition plates are sequentially distributed in parallel at intervals from top to bottom, a second splicing groove is formed in the front end of each transverse partition plate, and the first splicing grooves and the second splicing grooves are connected in a matched mode to form a cross tenon-and-mortise structure. The integral transverse partition plates and the vertical partition plates are connected in a mortise and tenon joint mode, high connection strength is achieved, the number of workpieces of the transverse partition plates and the vertical partition plates is effectively reduced, the field environment of a workshop is improved, the workload of welding of the workpieces is reduced, production efficiency is improved, and welding cost is reduced.

This utility model discloses a deformation-resistant steel fire door, comprising a steel fire door body, a front panel, a rear panel, and a heat insulation shell. The front and rear panels are arranged in parallel. A positioning groove is provided on the inner side of the rear panel. The heat insulation shell is disposed inside the positioning groove, and its opening is bent outward to form a flange extending into a buffer space and fitted with a sealing frame. The front panel has multiple countersunk holes, and screw holes are provided on the flange and the rear panel directly opposite the countersunk holes. Buffer bolts for fixing the front and rear panels together are threaded into the corresponding screw holes and countersunk holes. This utility model has a simple structure. The steel fire door consists of a front panel, a rear panel, and a heat insulation shell. The heat insulation shell reduces heat transfer, greatly reducing the probability of the front panel deforming due to high temperatures. Furthermore, the front and rear panels are replaceable; if any one panel deforms, only the damaged panel can be replaced, avoiding the need for complete replacement and reducing costs.

The invention provides a scrap steel quality inspection APP pressing block two-end dismounting prompting method, and relates to the technical field of scrap steel quality inspection in the iron and steel industry, and the method comprises the following steps: S1, material type selection: an acceptor selects a to-be-inspected scrap steel pressing block type in a scrap steel quality inspection APP system according to an actual condition; s2, inputting to-be-detected information: inputting specific information of a to-be-detected briquetting vehicle; and S3, two-end unpacking prompting: according to the input layer number, row number and column number of the to-be-detected briquette vehicle, dynamically generating an unpacking instruction by the APP, and prompting an acceptor to carry out unpacking detection. According to the scrap steel quality inspection APP pressing block two-end-disassembly prompting method, an unpredictable and tamper-resistant random unpacking mark generation system is constructed, dynamic generation of an unpacking instruction is achieved, the sampling rule is transparent, man-made interference is resisted, and the unpacking efficiency is improved. Therefore, the problems that the inspection efficiency is reduced due to full-amount disassembly of the two ends and the internal quality of the pressing block is possibly covered due to lack of randomness are solved, and the accuracy and the high efficiency of scrap steel inspection are effectively improved.

1. The name of the design product: corner structure (steel fireproof curtain wall 02 - inner arc). 2. The use of the design product: used as a connection of building decoration profile. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

This invention relates to the field of iron and steel metallurgy technology, and discloses a digital short-process steelmaking quality prediction method and system. The method includes: acquiring standard temperature deviation data and standard chemical fluctuation data; obtaining an initial disturbance distribution map using a correlation propagation model; constructing a global coupling matrix through graph structure mapping; determining the disturbance accumulation trend through time-series analysis and fusing it with chemical fluctuation analysis to obtain a fluctuation state identifier; adaptively correcting the global coupling matrix based on the fluctuation state identifier to obtain an optimized description of the disturbance influence range; and finally determining the final quantified value of steel quality by combining quality index mapping logic. This invention achieves accurate quantification and forward-looking prediction of the dynamic propagation and nonlinear accumulation effects of disturbances in short-process steelmaking by dynamically correcting the coupling matrix.

The utility model discloses a kind of split shaving cutter for machining gear, to solve the problem of high cost of traditional integrated structure, waste resources, realize tool body replacement alone, tool holder reuse, it includes tool body and steel tool holder, tool body coaxial sleeve is connected in tool holder, tool holder is equipped with root cleaning, first reference surface and the blocking ring of second reference surface with inclination, tool body is positioned by lower end face adhering first reference surface, connecting through-hole wall abutting second reference surface, and is connected with cooperation positioning bolt.The stress concentration is reduced by being equipped with avoidance ring groove, the stability of positioning member and positioning groove adaptation enhances, positioning member adopts rectangular part and semicircular part adhering structure, inside side angle is equipped with avoidance cutting surface to facilitate assembly and disassembly.The design reduces cost, improves machining precision stability and structural reliability, optimizes assembly efficiency.

The invention provides a steel quality diagnosis method and device, equipment and a storage medium, and the method comprises the steps: obtaining technological parameters and quality parameters of a steel production process, obtaining domain knowledge information, building a quality defect knowledge graph based on the domain knowledge information, and injecting the quality defect knowledge graph into a preset data diagnosis model, thereby obtaining an enhanced diagnosis model; performing predictive diagnosis on the process parameters and the quality parameters based on the enhanced diagnosis model to obtain a predictive diagnosis result, performing reasoning diagnosis on the process parameters and the quality parameters based on the quality defect knowledge graph to obtain a reasoning diagnosis result, and if the predictive diagnosis result and the reasoning diagnosis result have conflicts, judging whether the process parameters and the quality parameters are conflicted or not. If yes, processing according to a preset conflict resolution strategy to obtain a steel quality diagnosis result; according to the invention, based on cooperative work of prediction diagnosis and reasoning diagnosis, diagnosis result differences are processed in combination with a conflict resolution strategy, the accuracy and interpretability of steel quality diagnosis are improved, and the adaptability of a diagnosis system under complex working conditions is enhanced.

A convenient lattice type pile pulling device comprises a steel square ground breaking module, a lattice column, a lattice column connecting unit and a lattice column extending frame which are sequentially connected from bottom to top, the lattice column extending frame is connected with a vibration head of a crawler crane, and the lower portion of the steel square ground breaking module is a square ground breaking head of a rectangular hollow structure. An inclined plate is arranged on the front face of the upper portion of the steel square ground breaking module to form a variable cross-section, the minimum position of the variable cross-section is connected with the latticed column, and the minimum cross-section of the variable cross-section is the same as the cross-section of the latticed column. Inclined sliding grooves are formed in the left side face and the right side face of the square ground breaking head, the two inclined sliding grooves are sleeved with the steel bottom-carrying pile pulling sliding bolt in an up-down sliding mode, the highest positions of the inclined sliding grooves are close to the front face of the square ground breaking head, namely the maximum section position of the inclined plate, and the lowest positions of the inclined sliding grooves are located in the middle of the square ground breaking head. The structure is easy and convenient to construct, high in construction speed, low in cost and small in influence on the environment.

The application discloses a small-sample steel quality multi-index joint prediction method and device and a medium, relates to the field of material performance prediction, and comprises the following steps: determining a standardized data set comprising steel chemical composition, process parameter characteristics and multi-dimensional mechanical performance index data; configuring a target order of the multi-dimensional mechanical performance index; constructing a regression chain model based on the target order; in the regression chain model, the prediction value of a previous mechanical performance index is used as an additional input feature of a corresponding regressor of a subsequent mechanical performance index; adopting a nested cross-validation strategy to perform hyperparameter search and optimization on the regression chain model, and determining a target hyperparameter combination; performing full-amount retraining on the regression chain model based on the target hyperparameter combination, and obtaining a trained steel performance joint prediction model; and inputting the chemical composition and production process parameters of the steel to be predicted into the steel performance joint prediction model, and obtaining a joint prediction result of each steel performance index corresponding to the steel to be predicted.

The application discloses a detachable arc striking and extinguishing device based on a box pipeline and a welding method, which comprises a copper main body, a positioning groove is formed in the copper main body, an arc striking assembly is arranged in the positioning groove, clamping assemblies are arranged on the two sides of the copper main body, and the clamping directions of the two clamping assemblies are perpendicular to each other. The device is convenient to install and detach, can provide a stable arc striking and extinguishing point for a weld during a welding process, can easily separate the weld after the welding is completed, and has a size and characteristics which do not interfere with other welding positions, so that pipeline operation is facilitated. The device does not need to additionally add a steel arc striking plate, thereby reducing the negative influence of the labor work amount and a carburized layer, and improving production efficiency and weld quality.