Composite plate rolling process monitoring method and system based on high-temperature high-brightness imaging technology

Abstract

The invention discloses a composite plate rolling process monitoring method based on a high-temperature high-brightness imaging technology, and relates to the technical field of composite plate rolling, and the method comprises the following steps: collecting image data of a plate blank rolling process through a high-temperature high-brightness phase machine; constructing a training data set and a verification data set through the collected image data; building a slab cracking detection model based on the convolutional neural network, training the model through the training data set until convergence, and verifying the model through the verification data set; image data of the composite plate blank are collected in real time, the image data are recognized through the plate blank cracking detection model, and whether cracking characteristic data meeting a preset threshold value are detected or not is judged. No-dead-corner monitoring of the whole rolling process of the composite plate is achieved, the problem of high-brightness shielding temperature is effectively solved, the modular design can be matched with existing rolling mill equipment, and the rolling failure rate of the composite plate is reduced.

Description

Technical Field

[0001] This invention relates to the field of composite plate rolling technology, and in particular to a method and system for monitoring the composite plate rolling process based on high-temperature and high-brightness imaging technology. Background Technology

[0002] The composite plates produced by the heavy plate mill are diverse and widely used in various fields. Unlike ordinary steel billets, composite plates are subject to cracking during the rolling process, as the rolled piece is pulled between rotating rolls by friction and undergoes plastic deformation under compression.

[0003] Currently, the composite plate rolling process is generally monitored manually. When operators use ordinary video monitoring, the billet temperature exceeds 1000℃, causing the billet surface to be highly bright, making it impossible to clearly identify the state of the beginning and end of the plate. Furthermore, it is difficult to simultaneously solve multiple interferences such as high temperature interference, moisture obstruction, and mechanical obstruction from equipment, thus making it impossible to accurately detect the potential cracking of the composite plate billet through the system. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method and system for monitoring the rolling process of composite plates based on high temperature and high brightness imaging technology.

[0005] To solve the above technical problems, the technical solution of the present invention is as follows: A method for monitoring the rolling process of composite plates based on high-temperature, high-brightness imaging technology includes: Image data of the slab rolling process is acquired using a high-temperature, high-brightness camera. Training and validation datasets were constructed using the collected image data; A slab crack detection model was constructed based on a convolutional neural network. The model was trained until convergence using a training dataset and validated using a validation dataset. Image data of composite slabs are acquired in real time, and the image data is identified by the slab crack detection model to determine whether crack feature data that meets the preset threshold is detected.

[0006] As a preferred embodiment of the composite plate rolling process monitoring method based on high-temperature and high-brightness imaging technology described in this invention, the real-time acquisition of image data of the composite slab includes: Image data of the head and tail of the composite slab were acquired using a high-temperature, high-brightness camera.

[0007] As a preferred embodiment of the composite plate rolling process monitoring method based on high temperature and high brightness imaging technology described in this invention, four high temperature and high brightness cameras are provided, with two cameras used to collect image data of the composite plate head and two cameras used to collect image data of the composite plate tail.

[0008] As a preferred embodiment of the composite plate rolling process monitoring method based on high temperature and high brightness imaging technology described in this invention, the high temperature and high brightness camera operates at a temperature of -40℃ to 80℃.

[0009] As a preferred embodiment of the composite plate rolling process monitoring method based on high-temperature and high-brightness imaging technology described in this invention, after real-time acquisition of image data of the composite slab, identification of the image data through a slab crack detection model, and determination of whether crack feature data meeting a preset threshold are detected, the method further includes: An alarm is triggered when crack feature data that meets a preset threshold is detected.

[0010] This invention also provides a composite plate rolling process monitoring system based on high-temperature, high-brightness imaging technology, comprising: The image acquisition module is used to acquire image data of the slab rolling process using a high-temperature, high-brightness camera; The data processing module is used to construct training and validation datasets from the acquired image data; The model building module is used to build a slab crack detection model based on a convolutional neural network. The model is trained on the training dataset until it converges, and the model is validated on the validation dataset. The defect identification module is used to acquire image data of composite slabs in real time, identify the image data through the slab crack detection model, and determine whether crack feature data that meets the preset threshold is detected.

[0011] As a preferred embodiment of the composite plate rolling process monitoring system based on high-temperature and high-brightness imaging technology described in this invention, it further includes: The alarm module is used to trigger an alarm when crack feature data that meets a preset threshold is detected.

[0012] The beneficial effects of this invention are: (1) The present invention installs a high-temperature and high-brightness camera at the inlet and outlet of the rolling mill. The high-temperature and high-brightness camera collects high-resolution image data during the slab rolling process. The collected image data is then identified by the slab crack detection model to determine whether crack feature data that meets the preset threshold is detected. An alarm is triggered when crack feature data that meets the preset threshold is detected. This realizes full-process monitoring of composite plate rolling without dead angles, effectively solves the problem of high brightness blocking temperature, and the modular design can be used with existing rolling mill equipment to reduce the failure rate of composite plate rolling.

[0013] (2) In this invention, four high-temperature and high-brightness cameras are set up and deployed in a distributed manner at the head and tail of the board. The two high-temperature and high-brightness cameras at the head and tail of the board collect image data of the composite board from different angles to form multi-angle detection, which ensures the resolution of the collected slab images. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A schematic flowchart of the composite plate rolling process monitoring method based on high temperature and high brightness imaging technology provided by the present invention; Figure 2 High-resolution images of slab rolling process captured by a high-temperature, high-brightness camera; Figure 3 A schematic diagram of the composite plate rolling process monitoring system based on high-temperature and high-brightness imaging technology provided by the present invention. Detailed Implementation

[0016] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0017] Figure 1 This is a flowchart illustrating a method for monitoring the rolling process of composite plates based on high-temperature, high-brightness imaging technology, provided in an embodiment of this application. The method specifically includes the following steps: Step S101: Acquire image data of the slab rolling process using a high-temperature, high-brightness camera.

[0018] Specifically, high-temperature, high-brightness cameras are installed at the inlet and outlet of the rolling mill, and equipped with circulating water-cooling kits to ensure the cameras operate at temperatures ranging from -40℃ to 80℃, guaranteeing stable operation at 70℃. High-resolution image data of the slab rolling process is acquired using these cameras; see [link to relevant documentation]. Figure 2 .

[0019] In this embodiment, four high-temperature and high-brightness cameras are installed in a distributed manner at the head and tail of the composite board. That is, two high-temperature and high-brightness cameras collect image data at the head of the composite board, and two high-temperature and high-brightness cameras collect image data at the tail of the composite board.

[0020] Ideally, two high-temperature, high-brightness cameras at the head and tail of the board acquire image data of the composite board from different angles to form a multi-angle detection.

[0021] Step S102: Construct training and validation datasets using the collected image data.

[0022] Step S103: Construct a slab crack detection model based on a convolutional neural network, train the model using a training dataset until convergence, and validate the model using a validation dataset.

[0023] Specifically, a slab rolling crack detection algorithm model was constructed based on deep learning algorithms such as Convolutional Neural Networks (CNN) and Transformer. CNN utilizes convolutional and pooling layers to automatically extract local and abstract features from images, effectively identifying crack textures, shapes, and other features on the slab surface. Transformer, with its self-attention mechanism, models global image features, solving long-distance dependency problems and achieving accurate capture of complex crack morphologies. Supervised learning was conducted using a large amount of labeled slab crack and normal sample data, and transfer learning and data augmentation optimization strategies were employed to improve the model's generalization ability.

[0024] Step S104: Real-time acquisition of image data of composite slab, identification of image data through slab crack detection model, and determination of whether crack feature data meeting the preset threshold is detected. When crack feature data meeting the preset threshold is detected, an alarm is triggered.

[0025] Specifically, based on real-time image data streams, deep learning inference technology is used to automatically identify surface cracking defects in slabs. When cracking features that meet preset thresholds are detected, the system triggers a graded alarm mechanism to achieve intelligent and automated quality control of the rolling production process.

[0026] See Figure 3 This application also provides a composite plate rolling process monitoring system based on high temperature and high brightness imaging technology. The system includes: an image acquisition module, a data processing module, a model building module, a defect identification module, and an alarm module.

[0027] The image acquisition module is used to acquire image data of the slab rolling process using high-temperature, high-brightness cameras. Specifically, the image acquisition module includes four high-temperature, high-brightness cameras, which are deployed in a distributed manner at the head and tail of the slab. That is, two high-temperature, high-brightness cameras acquire image data of the head of the composite slab, and two high-temperature, high-brightness cameras acquire image data of the tail of the composite slab.

[0028] The data processing module is used to construct training and validation datasets from the collected image data.

[0029] The model building module is used to build a slab crack detection model based on a convolutional neural network. The model is trained using a training dataset until it converges, and validated using a validation dataset.

[0030] The defect identification module is used to acquire image data of composite slabs in real time, identify the image data through the slab crack detection model, and determine whether crack feature data that meets the preset threshold is detected.

[0031] The alarm module is used to trigger an alarm when crack feature data that meets a preset threshold is detected.

[0032] Therefore, the technical solution of this application realizes full-process monitoring of composite plate rolling without blind spots, effectively solves the problem of high brightness blocking temperature, and the modular design can be used with existing rolling mill equipment to reduce the failure rate of composite plate rolling.

[0033] In addition to the above embodiments, the present invention may have other implementation methods; all technical solutions formed by equivalent substitution or equivalent transformation fall within the protection scope claimed by the present invention.

Claims

1. A method for monitoring the rolling process of composite plates based on high-temperature, high-brightness imaging technology, characterized in that: include: Image data of the slab rolling process is acquired using a high-temperature, high-brightness camera. Training and validation datasets were constructed using the collected image data; A slab crack detection model was constructed based on a convolutional neural network. The model was trained until convergence using a training dataset and validated using a validation dataset. Image data of composite slabs are acquired in real time, and the image data is identified by the slab crack detection model to determine whether crack feature data that meets the preset threshold is detected.

2. The method for monitoring the rolling process of composite plates based on high-temperature and high-brightness imaging technology according to claim 1, characterized in that: The image data acquired through the high-temperature, high-brightness camera during the slab rolling process includes: Image data of the head and tail of the composite slab were acquired using a high-temperature, high-brightness camera.

3. The method for monitoring the rolling process of composite plates based on high-temperature and high-brightness imaging technology according to claim 2, characterized in that: Four high-temperature, high-brightness cameras are provided, with two cameras used to collect image data of the front of the composite board and two cameras used to collect image data of the rear of the composite board.

4. The method for monitoring the rolling process of composite plates based on high-temperature and high-brightness imaging technology according to claim 1, characterized in that: The operating temperature of the high-temperature, high-brightness camera is -40℃ to 80℃.

5. The method for monitoring the rolling process of composite plates based on high-temperature and high-brightness imaging technology according to claim 1, characterized in that: After acquiring image data of the composite slab in real time, identifying the image data using the slab crack detection model, and determining whether crack feature data meeting a preset threshold is detected, the process further includes: An alarm is triggered when crack feature data that meets a preset threshold is detected.

6. A monitoring system for the rolling process of composite plates based on high-temperature, high-brightness imaging technology, comprising: The image acquisition module is used to acquire image data of the slab rolling process using a high-temperature, high-brightness camera; The data processing module is used to construct training and validation datasets from the acquired image data; The model building module is used to build a slab crack detection model based on a convolutional neural network. The model is trained on the training dataset until it converges, and the model is validated on the validation dataset. The defect identification module is used to acquire image data of composite slabs in real time, identify the image data through the slab crack detection model, and determine whether crack feature data that meets the preset threshold is detected.

7. The composite plate rolling process monitoring system based on high-temperature and high-brightness imaging technology according to claim 6, characterized in that: Also includes: The alarm module is used to trigger an alarm when crack feature data that meets a preset threshold is detected.

Images