4241 results about "Strip steel" patented technology

The invention discloses a strip steel surface defect detection method based on an attention mechanism and an improved UNet network, relates to the field of industrial image processing, and aims to solve the defects of a traditional detection method in defect positioning precision, multi-scale feature fusion and class imbalance. According to the technical scheme, an improved UNet network architecture is adopted, and the method specifically comprises the following steps: in a coding stage, designing a UELAN feature extraction module to improve parameter efficiency; a coordinate attention mechanism (CA) is introduced to enhance the focusing capability on a defect area; and capturing multi-scale high-level semantic features in combination with a cavity space pyramid pooling module (ASPP). In a decoding stage, a convolutional block attention mechanism (CBAM) is integrated, and a channel and spatial feature fusion are dynamically optimized. Besides, a combined loss function is put forward, focus loss and Dice loss are fused, the problem of boundary uncertainty is relieved in combination with a label fuzzification strategy, and category imbalance is coped with through weighted cross entropy. According to the invention, high-precision pixel-level segmentation of strip steel surface defects is realized.

A strip steel surface defect area dimension measurement method relates to a defect dimension measurement method, and comprises the following steps: calibrating a camera coordinate system, and shooting a strip steel surface image by using a camera; identifying defects on the surface of the strip steel in the image by using a defect identification model, dividing a defective area in the image, measuring the size of the defective area, and calculating the actual size of the defective area according to the coordinate conversion relation; the defect identification model is based on YOLOv8s, a linear focusing attention mechanism is fused into a C2F module of a backbone network to form a C2F-FLA module, global features of an image are concerned, and more defect related features are captured; a bidirectional weighted feature pyramid network structure is adopted in the Neck network part, and a wavelet convolution module is utilized to respond to low-frequency information in the image; and performing down-sampling by adopting a Haar wavelet down-sampling module. The feature extraction capability is improved while the model parameter quantity, the large calculation amount and the calculation complexity are reduced, so that the detection result is more accurate.

The invention provides a cold-rolled sheet shape multivariable cooperative control optimization method and system. The method comprises the steps that rolling tension set value parameters are determined on the basis of a dynamic coupling model so as to be matched with dynamic tension fluctuation in the high-speed strip steel rolling process of a cold-rolled sheet, the magnetic conductivity change in the strip steel rolling process is detected through the magnetoelastic effect, and a magnetic conductivity gradient map is generated; residual stress distribution data are inverted based on the magnetic conductivity gradient map, and target axial roll shifting control parameters influencing the roll gap shape are generated; and multivariable cooperative matching is conducted on the rolling tension set value parameter and the target axial roll shifting control parameter, a matching result is obtained, the rolling tension and the roll gap shape are adjusted, and strip steel deviation and fluctuation caused by dynamic coupling of the composite material tension and the cold-rolled sheet shape in the cold-rolled sheet high-speed strip steel rolling process are restrained. According to the technical scheme, through cooperative control of the dynamic coupling model and the magnetic conductivity gradient map, deviation and fluctuation of the strip steel are effectively restrained, and the shape stability and rolling precision of the cold-rolled plate are improved.

The invention provides a cold-rolled strip steel plate shape closed-loop control method and system based on dynamic feedback, and belongs to the technical field of cold-rolled strip steel production.The method comprises the steps that a dynamic incidence matrix and a plate shape deviation signal are generated based on cold-rolled strip steel transverse measuring point data, processing the dynamic incidence matrix to obtain a space correlation defect feature vector of the cold-rolled strip steel, and processing the strip shape deviation signal to obtain a multi-dimensional defect feature vector; determining the strip shape defect type of the cold-rolled strip steel based on the multi-dimensional defect feature vector and the space correlation defect feature vector; and strip steel production control parameters are adjusted based on the strip shape defect type. The plate shape quality stability of the cold-rolled strip steel can be improved.

The invention discloses thin-gauge non-oriented silicon steel and a production method thereof, and a motor iron core and a manufacturing method thereof. The production method comprises the following steps: heating a steel billet to 1080-1120 DEG C, preserving heat for 150-200 minutes, performing rough rolling to obtain an intermediate billet of 35-45 mm, performing finish rolling, and coiling to obtain a hot-rolled coiled plate of 2.30-3.00 mm; the initial rolling temperature of finish rolling is 970 + / -20 DEG C, the final rolling temperature is 850 + / -20 DEG C, the reduction rate is 92% or above, and the coiling temperature is 700 + / -20 DEG C; acid pickling and cold rolling are carried out under the condition that normalizing is not carried out, and cold-rolled strip steel with the thickness ranging from 0.15 mm to 0.35 mm is obtained; the total reduction rate of cold rolling is greater than or equal to 86%; performing primary annealing, cooling and coating to obtain a steel plate finished product; the annealing temperature is 830-870 DEG C, and the temperature is kept for 60-90 seconds. Therefore, the same piece of non-oriented silicon steel can meet the requirements of the stator and rotor iron cores at the same time, the manufacturing cost of the stator and rotor iron cores is low, and material waste is avoided.

The invention relates to the field of deep learning, and provides a strip steel surface defect detection method and system, and electronic equipment, and the method comprises the steps: obtaining the state information of a to-be-detected strip steel, the state information at least comprising a surface image of the to-be-detected strip steel; inputting the state information into a pre-constructed detection model for defect detection, wherein the detection model is constructed by fusing an improved spatial pyramid module and a polarization self-attention mechanism on the basis of a Yolov8 model; the improved spatial pyramid module comprises an SPPCSPC layer and a variable convolutional layer, and the polarization self-attention mechanism comprises a channel branch and a spatial branch; and determining the surface defect of the to-be-detected strip steel based on the detection information output by the defect detection model. The method and the device are used for overcoming the defect of poor detection precision when defect detection is carried out on the surface of the strip steel in the prior art, and compared with an existing Yolov8 model, the scheme provided by the invention has the advantage that the detection precision is obviously improved when the defect detection is carried out on the surface of the strip steel.

The invention provides a method for reducing the transverse same-plate difference of a silicon steel product. The method comprises the following steps that S1, the original roller convexity of a cold-rolled silicon steel rolling mill is increased; s2, upper and lower middle rollers of the cold-rolled silicon steel rolling mill are moved, and the axial movement positions of the upper and lower middle rollers are determined according to the width of the strip steel; s3, adjusting the bending roller of the cold-rolled silicon steel rolling mill according to the rolling reduction rate: increasing the forward bending extension amount of the bending roller of the cold-rolled silicon steel rolling mill; s4, the total reduction rate of the cold-rolled silicon steel rolling mill is increased; s5, the strip steel tension of the cold-rolled silicon steel rolling mill is increased; and S6, the same-plate difference of the cold-rolled strip steel is reduced, and meanwhile the plate shape is improved. According to the method, the transverse same-plate difference of the silicon steel cold-rolled strip steel can be reduced to 0.0048-0.0051, and the requirements of manufacturers of motors, transformers and the like on the transverse same-plate difference of silicon steel products are met.

The invention discloses a strip steel surface defect detection method based on a multi-branch parallel attention model, and belongs to the technical field of industrial defect target detection. On the basis of a YOLOv8 network, space depth conversion convolution is used for replacing all down-sampling convolution in a backbone network CSPDarkNet, a multi-branch parallel attention module is used for replacing a C2f module, and an internal feature interaction module is added behind the backbone network, so that the strip steel surface defect detection precision of the model is improved. The strip steel surface defect detection precision is effectively improved while model parameters are slightly increased, and compared with an existing strip steel surface defect model, the problem that the image detection effect is poor due to the fact that background interference is strong, various defects are large in difference, the number of small defects is large, and defect targets are likely to be confused in an industrial scene can be better solved.

The invention provides a continuous annealing strip steel performance multi-index prediction method based on multi-objective evolution deep learning, and relates to the technical field of data processing in the cold rolling production control process of iron and steel enterprises. A multi-objective evolutionary algorithm is applied to hyper-parameter selection of a deep learning model, so that the precision of the online detection method is further improved; meanwhile, a training set is obtained in a random selection mode, the optimal hyper-parameter combination of the deep learning model can be rapidly obtained through the search strategy of the training set in the hyper-parameter space, and finally the generalization ability of the online detection model is improved. According to the method provided by the invention, the defects of a traditional deep neural network model based on artificial experience are overcome, simultaneous prediction of multiple performance indexes is realized, autonomous optimization construction of a deep learning model and multi-index prediction of strip steel performance are realized, and the cold rolling production can be helped to improve the product quality control level.

The invention provides a strip steel surface defect detection method and system based on a depth map, and the method comprises the steps: S1, carrying out data collection, obtaining steel plate surface point cloud data, and then generating a point cloud data set; s2, after a point cloud data set is obtained, preprocessing is carried out, and defect detection is carried out; s3, judging the category of the defect image; and S4, respectively displaying the defects of the steel plate in combination with detection results. Through the man-machine interaction module provided by the invention, detection results including depth, size, proportion and category conclusions of defects can be visually displayed to a user, so that the user can conveniently know the defect condition of the surface of the steel plate.

The invention provides a post-buckling calculation method for cold-rolled strip steel internally provided with an initial strip shape defect, and relates to the technical field of buckling. The method comprises the following steps: firstly, acquiring geometric dimension parameters and attribute parameters of cold-rolled strip steel with initial plate shape defects in a production field, and then establishing a large deflection balance equation and a deformation coordination equation of the cold-rolled strip steel with initial defect items in the interior to form a post-buckling calculation model of the cold-rolled strip steel with the initial plate shape defect items in the interior; and finally, a perturbation method and a Hamiltonian method are used for solving a large-deflection balance equation and a deformation coordination equation of the cold-rolled strip steel internally provided with the initial plate type defect item, and a general analysis model formula for post-buckling of the cold-rolled strip steel internally provided with the initial plate type defect item is obtained. According to the post-buckling calculation method for solving the cold-rolled strip steel internally provided with the initial plate shape defect, the post-buckling mechanical mechanism of the cold-rolled strip steel internally provided with the initial plate shape defect can be better improved, the plate quality is improved, and the plate defect is eliminated.

The invention provides a hot-rolled strip steel tracking control method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a strip steel tail initial position, a strip steel operation speed, a strip steel entering moment and a prediction compensation parameter, and obtaining a current passing moment to determine tail tracking information when a tail passing signal is received; determining the current passing moment and the tail tracking information as current tail tracking information, determining the steel throwing moment and the actual strip steel tail position in response to a steel throwing completion signal, and determining the predicted strip steel tail position according to the current tail tracking information, the steel throwing moment, the prediction compensation parameter and the strip steel running speed; determining a position difference based on the actual strip steel tail position and the predicted strip steel tail position, and updating the predicted compensation parameter according to the position difference; according to the method, dependence on a detector is reduced through multi-parameter comprehensive calculation, it is ensured that the correction effect is more accurate before the strip steel reaches the rolling mill by circularly updating the prediction compensation parameters, and prediction accuracy and control stability are gradually improved.

The invention discloses a high-speed strip steel defect detection method and system based on anomaly enhancement and image reconstruction, the problem that part of defects are difficult to recognize is solved by adopting a pseudo anomaly enhancement method, the defects difficult to detect are divided into a scratch type and a stain type, a local average gray level floating method is designed to generate scratch type pseudo anomalies, and the defect detection accuracy is improved. And a contour random disturbance method is used to generate stain type pseudo anomalies, and local difference features are provided for the model. And a lightweight image reconstruction model of an encoder-decoder structure is designed to achieve high detection speed and meet the requirement of synchronous detection. According to the method, performance testing is carried out on a strip steel surface image data set, advanced performance evaluation is obtained, and the effectiveness of the method is proved.

The invention relates to the field of metal processing, and particularly discloses a low-carbon steel oxide scale reduction annealing process and a hot galvanizing process, the low-carbon steel oxide scale reduction annealing process comprises the following steps: S1, pretreatment: firstly carrying out water / abrasive sand blasting treatment on a hot-rolled strip steel substrate, then washing with water, then heating to 650-700 DEG C at a heating rate of 50-60 DEG C / s, and keeping the temperature for 2-3 hours; after heat preservation treatment, taking out and immediately putting into a salt solution to obtain a pretreated base material; s2, reduction annealing: carrying out reduction annealing on the prepared pretreated base material in a reducing gas atmosphere; the hot galvanizing process comprises the following steps: carrying out oxide scale reduction annealing treatment on a hot-rolled strip steel base material; and hot galvanizing is conducted on the hot-rolled strip steel obtained after the oxide scale reduction annealing treatment is conducted in zinc liquid, a zinc coating is formed on the surface of the hot-rolled strip steel, and the hot-galvanized plate is prepared. The method has the characteristics that the oxide scale is effectively reduced, good adhesion of a zinc layer and a substrate is ensured, and the quality of a plating layer is improved.

The utility model relates to the field of mechanical manufacturing, and discloses a strip steel stamping device for hose clamp production, which comprises a stamping platform, hose clamp stamping equipment is fixedly arranged at the top of the stamping platform, a micro motor is fixedly mounted on one side of the hose clamp stamping equipment, and two groups of side plates are fixedly arranged on one side, corresponding to the hose clamp stamping equipment, of the top of the stamping platform. A plurality of sets of supporting rollers distributed at equal intervals are rotationally installed between the two sets of side plates, a driving motor is fixedly installed at the top of the stamping platform, and hose clamp pieces are arranged on the stamping platform. Therefore, the two sets of grinding discs rotate and grind the two sides of the hose clamp piece when rotating along with the two sets of rotating shafts, burrs on the two sides of the hose clamp piece are eliminated, the personal safety of operators is protected, potential safety hazards are reduced, and the product quality of the hose clamp piece is improved.

The invention belongs to the technical field of cold-rolled strip steel shearing control, and provides a silicon steel shearing unit coiling tension automatic control method which comprises the steps that S1, a coiling tension preset value preset strategy is compiled for silicon steel according to the strip steel incoming material thickness and the finished product width, and an automatic tension matching model and a taper control model are established; s2, optimizing a current tension preset value by using an automatic tension matching model; s3, outputting a tension taper coefficient in real time according to the actual value of the coiling diameter of the steel coil by using the taper control model; s4, according to the real-time actual rolling diameter value, the tension taper coefficient and the tension preset value, a motor torque set value dynamically changing along with the actual rolling diameter value is output; strict execution of coiling tension technologies of different strip steel products is ensured, and the problems that the electromagnetic performance of a steel plate is degraded due to the fact that the coiling tension of the shearing procedure is too large, and coil collapse is caused due to insufficient tension are solved.

The invention discloses a strip steel surface defect detection and classification method based on a graph neural network, and the method comprises the steps: segmenting a strip steel surface image into different scale sub-blocks, and extracting the node features of each scale sub-block; constructing graph structures with different scales according to the similarity of the node features; constructing a stacked multi-scale image attention network model, taking the image structures of different scales as input for training, inputting the image structures of different scales into the trained stacked multi-scale image attention network model, performing feature extraction on nodes in the image structures of different scales, and performing feature extraction on the nodes of the image structures of different scales; aggregating and updating node features in the different scale graph structures through a graph attention mechanism, and obtaining a feature matrix according to the updated node features in the different scale graph structures; and paving the obtained feature matrix into a one-dimensional vector, inputting the one-dimensional vector into a full connection layer, carrying out nonlinear operation, outputting the probability of each category through a softmax function, and taking the category with the maximum output probability as the category of the strip steel surface defects.

The invention belongs to the technical field of steel production, and relates to a method for improving the accuracy of final rolling temperature prediction based on a deep learning algorithm. According to the method, multiple key data, such as finish rolling inlet temperature, rolling force, cooling water flow and rolling speed, influencing the finish rolling temperature of hot rolled strip steel are selected through correlation analysis, and a generalized regression neural network (COA-GRNN) optimized by a raccoon optimization algorithm is utilized to establish a finish rolling temperature prediction model. The model realizes remote monitoring of the finish rolling temperature, so that production personnel can predict the trend of the finish rolling temperature in advance and adjust the finish rolling temperature in time, thereby ensuring the quality stability of strip steel and the yield of production.

The invention discloses a cold-rolled strip steel rolling control method and related equipment, and relates to the technical field of strip steel rolling, and the method comprises the steps that a first deviation rectification value and a second deviation rectification value in the strip steel running process are obtained in real time; based on the first deviation correction value, the rolling speed of the rolling mill is adjusted to a preset speed; after the rolling speed is adjusted, the tension parameter and the roller inclination parameter of the target rack are dynamically adjusted based on the second deviation correction value; and the tension parameter and the roller inclination parameter are recovered to initial set values so as to maintain the continuous stability of the rolling process. According to the method, the deviation correction values in front of the pickling section outlet and the rolling mill inlet are obtained in real time, the rolling speed and the shearing speed are dynamically adjusted before the strip steel enters the rolling mill, the tension parameters of the rack and the inclination direction of the rollers are accurately adjusted in combination with feedback of the second deviation correction value, and the deviation trend caused by camber defects is effectively restrained; and the rolling continuity and stability are improved.

The invention discloses a deep reinforcement learning-based self-learning deviation correction control method for a finish rolling strip threading process, and belongs to the technical field of steel rolling automatic control. The control method comprises the following steps: constructing a deep reinforcement learning model by taking a finish rolling rack in the finish rolling strip threading process as an intelligent agent; based on a deep reinforcement learning model, an optimal strategy is generated through interaction of an intelligent agent and a rolling environment; and the optimal strategy is embedded into a self-learning system in the finish rolling strip threading process for real-time correction, so that the strip steel inlet deviation amount is compensated in real time, and the strip steel outlet deviation amount is controlled within the process allowable range. By adopting the technical scheme provided by the invention, the deviation degree of the head of the finish-rolled strip steel can be greatly reduced.

The utility model discloses a tension hydraulic compensation adjusting mechanism based on strip steel coiling, and relates to the technical field of strip steel coiling. Comprising a mounting plate, three guide rollers arranged side by side are arranged at the upper end of the front face of the mounting plate, a linear empty groove is formed in the other side of the mounting plate, and an adjusting roller is arranged in the linear empty groove; the clamping assembly is located between the two front guide rollers; and the tension compensation assembly is arranged on the other side of the back surface of the mounting plate. A grating ruler is connected to a first protruding plate at one end of a movable plate, a hydraulic rod is arranged on the back face of a mounting plate, the hydraulic rod is matched with a second protruding plate through a pull block on an output shaft of the hydraulic rod to drive the movable plate to move downwards, an adjusting roller can be driven to move downwards when the tension of a spring is insufficient, and the tension of the adjusting roller is compensated; and the situation that the steel belt is loosened due to insufficient tension is avoided.

The invention relates to the technical field of metal material heat treatment, in particular to a gradient temperature control method for a stainless steel cold rolling annealing process. Comprising the steps that an annealing furnace is divided into independent temperature control sections in the strip steel conveying direction, each temperature section is provided with an electric heating wire set, a forced convection cooling air duct and a high-precision infrared temperature measurement array, temperature data are collected in real time, and a target gradient temperature curve is generated; constructing a phase change power-temperature correlation model, and outputting a theoretical gradient temperature curve; a three-stage MPC is adopted to cooperatively control and output a control signal; and the heating power and the cooling strength are adjusted in real time according to the control signal, if the temperature fluctuation exceeds the limit, the power is compensated through a PID algorithm, the strip steel speed is adjusted, and closed-loop feedback is formed. Through multi-section independent temperature control, thickness-temperature dynamic modeling and a three-level cooperative control framework, accurate tracking of a gradient temperature curve is achieved, and the quality and efficiency of the annealing process are improved.

The invention provides a coiling temperature calculation method and system based on a neural network and a mechanism model.The calculation method comprises the steps that firstly, strip steel slabs with the high coiling temperature hit rate are screened in a historical data set, abnormal values are removed, and layers are divided; segmenting the heat output roller way, and classifying and preprocessing data; performing offline training on the preprocessed data based on a BP neural network to obtain a water-cooling heat transfer coefficient prediction model of each layer; in the actual production process of the strip steel, input parameters of online operation are obtained, the water-cooling heat exchange coefficient of the real-time working condition is predicted online through the water-cooling heat exchange coefficient model, and then the mechanism model is further used for calculating the temperature on the strip steel reheating output roller way.

The utility model discloses a galvanized strip steel cutting device which comprises a cutting table, a transverse driving mechanism is arranged at the rear end of the cutting table, the front end of the transverse driving mechanism is connected with a longitudinal driving mechanism, and a plurality of pressing adsorption plates are arranged at the lower end of the longitudinal driving mechanism. A plurality of cutting grooves are formed in the front end of the cutting table, the pressing adsorption plates and the cutting grooves are arranged at intervals, a cutting mechanism and a pushing mechanism for pushing the cutting mechanism to move in the direction parallel to the cutting grooves are arranged at the bottom of the cutting table, and the cutting mechanism comprises a cutting motor. A plurality of cutting blades corresponding to the cutting grooves in a one-to-one mode are installed on a shaft body of the cutting motor, and the upper end parts of the cutting blades protrude out of the upper surface of the cutting table. According to the galvanized strip steel cutting device, the galvanized strip steel can be cut into multiple sections at the same time, the cutting efficiency is improved, and meanwhile the cut galvanized steel sheets can be effectively transferred.

The invention discloses a strip steel hot continuous rolling loop system modeling and fault detection method and system, and relates to the technical field of industrial steel rolling process modeling and fault detection, and the method comprises the steps: building a state space model of a strip steel hot continuous rolling loop system at a plurality of stable working points based on least square identification; fuzzifying the state space model based on a preset fuzzy rule to obtain a T-S fuzzy model of the strip steel hot continuous rolling loop system; optimizing the T-S fuzzy model based on a neural network, and establishing a target system model in which the neural network and fuzzy kernel representation are combined; and performing fault detection on the strip steel hot continuous rolling loop system based on the sum of residual signals between the model output of the target system model in the preset time period and the real output of the strip steel hot continuous rolling loop system. The technical problems that in the prior art, an accurate mathematical model of a loop is difficult to establish in a model-based fault detection method, and a data-driven fault detection method is low in accuracy and difficult in fault interpretation are solved.

The invention discloses a strip steel surface multiple defect detection method based on fuzzy measurement. The surface of the strip steel is shot and subjected to graying treatment, and a surface grayscale image is obtained; calculating a gray level histogram of the surface gray level image, and performing Gaussian function fitting on the gray level histogram to obtain an image background model; performing calculation and construction according to the image background model to obtain a membership function, and performing defect estimation on the surface grayscale image by adopting the membership function to obtain a membership matrix; and carrying out binarization processing on the surface grayscale image according to the membership matrix to obtain a strip steel surface defect area. The method is accurate and practical in detection, can detect various different types of surface defects, particularly has good detection performance for low-contrast images and fine defects, has relatively high application value for defect detection of the strip steel surface, and is relatively small in calculation amount and relatively high in accuracy due to linear operation mainly concentrated on a single pixel. The realization is easy.

The invention provides an intelligent detection method and device for the deviation amount of cold-rolled strip steel, and belongs to the technical field of data detection.The method comprises the steps that a strip steel edge line and a strip steel center line are determined based on a strip steel image; calculating a first deviation amount of the strip steel image based on the strip steel edge line and the standard edge line, and calculating a second deviation amount of the strip steel image based on the strip steel center line and the standard center line; the first deviation quantity is the deviation quantity of the edge part of the strip steel, and the second deviation quantity is the deviation quantity of the center part of the strip steel; and calculating a target deviation amount based on the first deviation amount of the strip steel image and the second deviation amount of the strip steel image. According to the intelligent detection method and device for the deviation amount of the cold-rolled strip steel, the problem that the deviation data detection accuracy of the cold-rolled strip steel is low can be solved.

The utility model provides a hot galvanizing strip steel edge heating device, and relates to the technical field of hot galvanizing strip steel production, the hot galvanizing strip steel edge heating device comprises a shell, strip steel, an air supply assembly and two air knives, the strip steel penetrates through the shell, and the left side and the right side of the strip steel are each provided with a heating assembly, a first adjusting assembly and a second adjusting assembly; by arranging the heating assembly, the cooling speed of the strip steel edge coating can be slowed down, so that the phenomenon that the strip steel edge coating is too thick is prevented, the strip steel quality is guaranteed, the heating assembly can heat the strip steel edge from the upper side and the lower side at the same time, the heating uniformity is improved, and the heating effect is improved; by arranging the first adjusting assembly, the distance between the first air outlet pipe and the second air outlet pipe can be adjusted according to actual use requirements so as to adapt to strip steel of different thicknesses, and the use stability is improved; and by arranging the second adjusting assembly, the distance between the two heating assemblies located on the left side and the right side of the strip steel can be adjusted according to actual use requirements so as to adapt to the strip steel with different widths, and the use stability is improved.

The utility model relates to the technical field of strip steel rolling, in particular to galvanized strip steel production rolling equipment. According to the technical scheme, the device comprises an air outlet pipe, a fixing frame, rollers and an air cooler, the rollers are symmetrically and rotationally installed in the positions, close to the upper surface and the lower surface, of the interior of the fixing frame, cooling cavities are formed in the two rollers, and the two ends of each roller are fixedly connected with a rotary joint A and a rotary joint B correspondingly; connecting pipes A are fixedly connected to the outer sides of the two rotating joints A, the tail ends of the two connecting pipes A are fixedly connected to the outer surface of an air outlet pipe, and an air inlet pipe is fixedly connected to the tail end of the air cooler output pipe. In the hot rolling process of the galvanized strip steel, the temperature of the roller is reduced, the service life of the roller is prolonged, heat of the roller is converted into hot air, production raw materials needing to be heated can be preheated, meanwhile, the hot air drying device can be used for drying the galvanized strip steel after follow-up cleaning operation, and energy consumption is saved.

The invention discloses a full-automatic deviation rectification control method for strip steel in the finish rolling process, and belongs to the field of hot rolling and finish rolling processes. According to the method, images in the strip steel rolling process are collected through a camera and sent to a server to recognize and calculate the deviation amount of the strip steel, then the deviation amount is sent to an automatic control system, the automatic control system predicts the deviation trend of the strip steel through a deviation amount fitting curve, a rack is leveled most weekly according to the deviation trend of the strip steel and closed-loop control over the deviation amount, and the strip steel is rolled. Therefore, strip steel deviation rectification is realized. Visual identification, artificial intelligence and automatic control can be effectively combined, full-automatic deviation correction of the strip steel in the finish rolling process is achieved, the purpose that a camera and an automatic control system are used for replacing an operator to conduct leveling intervention on deviation of the strip steel in the finish rolling process is achieved, rolling breakage or scrap steel caused by deviation of the finish rolling strip steel can be effectively reduced, and the production efficiency is improved. The loss of the working roller and the risk of manual misoperation are reduced, and the precision rolling stability, the rolling line automation rate, the labor efficiency and the yield are improved.