Special steel production-oriented furnace number full-process tracking system and method

Abstract

The invention discloses a furnace number full-process tracking system and method for special steel production, and belongs to the technical field of crossing of ferrous metallurgy intelligent manufacturing and industrial network safety. The problems that a traditional tracking system of an existing special steel enterprise depends on manual recording, so that the cross-process data intercommunication rate is low, the positioning and tracing precision is poor, and the confidentiality is low are solved, through a three-level architecture of a sensing layer, a network layer and an application layer, an industrial sensor is integrated, SM4 encryption transmission is carried out, and an isolated forest algorithm and a PBFT block chain consensus mechanism are improved. And full-process data communication and active security defense in the special steel manufacturing process are realized. The system has the advantages of equipment credibility authentication (SM2 algorithm), data tampering detection rate of 100%, block chain rapid response and the like, and is suitable for special steel production lines with high secrecy and quality tracing requirements.

Description

Technical Field

[0001] The present invention relates to the interdisciplinary technical field of intelligent manufacturing in iron and steel metallurgy and industrial network security, and in particular to a full-process tracking system and method for furnace numbers in special steel production. Background Art

[0002] Specialty steel, with its precise composition, high purity, and microstructure, forms a core pillar of modern industrial infrastructure. Its high-end products are used in aerospace, nuclear power equipment, precision manufacturing, and other fields. To meet the demands of intelligent metallurgical manufacturing and carbon neutrality, the next generation of specialty steel is developing towards functional integration, nanoscale strengthening, and high-entropy alloys. However, technological breakthroughs in these areas still require companies to develop refined and stable manufacturing processes.

[0003] Special steel production requires full-process tracking of furnace numbers to ensure quality control, but traditional production systems rely on manual records and paper circulation, resulting in serious data silos between processes and difficulty in interoperating data across processes. In addition, the traditional identification methods of steel such as coloring, marking, and partitioned stacking result in a very high data loss rate during transportation and subsequent processing, making steel mixing prone to occur. Paper records cannot be encrypted and transmitted, and the risk of data theft, tampering, and destruction is extremely high, which cannot meet the real-time defense needs of production data.

[0004] In view of this, there is an urgent need to develop a full-process tracking system and method for furnace numbers in special steel production, aiming to ensure that normal production is not affected while meeting the requirements of full-process tracking of furnace numbers in the production process of special steel to ensure quality controllability, and at the same time achieve full-process data penetration, high-precision positioning and active safety defense, providing a more solid guarantee for production quality and process information security, thereby promoting the intelligent development of special steel. Summary of the Invention

[0005] To address the problems of existing traditional production systems, such as difficulty in intercommunication of cross-process data, easy occurrence of steel mix-up, and inability to meet real-time production data protection requirements, the present invention provides a full-process tracking system for furnace numbers in special steel production, including:

[0006] The perception layer is used to collect process parameters and physical status data in real time through a multi-source sensor array, transmit the process parameters and physical status data to the network layer in a P2P mode, and laser print high-temperature resistant RFID QR code labels containing brand number, furnace number, production date, and contract number on the end face of the continuous casting billet and the head of the rolled material;

[0007] The network layer is used to set up 11 independent network paths (N1-N11), implement data encryption transmission based on IEEE802.3at dual-ring industrial Ethernet, use the Sec-OPC UA protocol to integrate SM4 encryption and SM3 hash check, and dynamically clean abnormal traffic through a CNN-LSTM hybrid model to ensure that data cannot be tampered with. Clean data is obtained and unidirectionally transmitted to the application layer through the independent network paths according to production needs;

[0008] The application layer is used to uniformly store the clean data in a temporary storage area, synchronize it to the database after a delay for artificial intelligence training and new product development, and push production inspection information to the mobile or PC end in a one-way manner through the real-time streaming module. The PBFT consensus mechanism is used to generate blockchain data blocks, combined with the improved isolation forest algorithm to achieve millisecond-level anomaly detection, and a digital twin of the physical-virtual space mapping is constructed through dynamic compensation equations.

[0009] Furthermore, the multi-source sensor array of the sensing layer includes: thermocouples, infrared thermometers, pressure sensors, flow sensors, laser displacement sensors, speed sensors, industrial cameras and infrared sensors.

[0010] Furthermore, in the network layer, data information is encapsulated through the MTSP v2.0 protocol, and the automatic redundancy switching time is less than 50ms.

[0011] Furthermore, in the application layer, the structure of the blockchain data block is: block hash + signature + Nonce + brand + furnace number + production date + contract number + timestamp.

[0012] It also provides a full-process tracking method for heat numbers in special steel production, including:

[0013] The sensing step is used to collect process parameters and physical status data in real time through a multi-source sensor array, transmit the process parameters and physical status data to the network layer using a P2P mode, and laser print high-temperature resistant RFID QR code labels containing the brand number, furnace number, production date, and contract number on the end face of the continuous casting billet and the head of the rolled material;

[0014] The network processing step is used to set up 11 independent network paths (N1-N11), implement data encryption transmission based on IEEE802.3at dual-ring industrial Ethernet, use the Sec-OPC UA protocol to integrate SM4 encryption and SM3 hash check, and dynamically clean abnormal traffic through a CNN-LSTM hybrid model to ensure that data cannot be tampered with. Clean data is obtained and unidirectionally transmitted to the application layer through the independent network paths according to production requirements;

[0015] The application steps are used to uniformly store the clean data in a temporary storage area, synchronize it to the database after a delay for artificial intelligence training and new product development, and at the same time push production inspection information to the mobile terminal or PC terminal in a unidirectional manner through the real-time streaming module, use the PBFT consensus mechanism to generate blockchain data blocks, combine the improved isolation forest algorithm to achieve millisecond-level anomaly detection, and construct a digital twin of physical-virtual space mapping through dynamic compensation equations.

[0016] Beneficial effects of the present invention:

[0017] This invention utilizes a three-tiered architecture encompassing perception, network, and application layers, integrating industrial sensors, SM4 encrypted transmission, an improved Isolation Forest algorithm, and the PBFT blockchain consensus mechanism to achieve data connectivity and proactive security defense throughout the entire process of special steel manufacturing. Based on IEEE802.3 (Ethernet) and the Internet of Things (IoT), digital twin tracking is achieved. By leveraging the two-way mapping of "physical-cyberspace" to establish trusted device authentication (SM2 algorithm), achieving a 100% data tampering detection rate, and achieving ultra-fast blockchain response, this approach completely addresses the challenges of existing traditional production systems, such as the difficulty of interoperable cross-process data, the susceptibility to steel mix-up, and the inability to meet real-time production data security requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a hierarchical architecture diagram of a full-process tracking system for heat numbers in special steel production as described in Example 1; DETAILED DESCRIPTION

[0019] The technical solution of the present invention is further described below with reference to the embodiments, but is not limited thereto. Any modification or equivalent replacement of the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention shall be included in the scope of protection of the present invention. The process equipment or devices not specifically noted in the following examples are all conventional equipment or devices in the art. Unless otherwise specified, the raw materials used in the examples of the present invention can be obtained commercially; unless otherwise specified, the technical means used in the examples of the present invention are all conventional means well known to those skilled in the art.

[0020] Example 1, combined Figure 1 This embodiment is described by Figure 1 It can be seen that:

[0021] This example is based on the tooling conditions of a steel plant's production line, integrating the ISO 16949-2016 quality control standard with an intelligent manufacturing system to develop a full-process tracking system for furnace numbers in special steel production, covering the entire smelting process.

[0022] During the perception layer deployment process in this embodiment:

[0023] The sintering trolley is equipped with an infrared thermometer (range 20-1500°C, accuracy ±0.01%), a speed sensor (range 0-500m / s, accuracy ±0.5Fs), and an industrial camera (1920×1080 pixels, 0.1mm level defect recognition).

[0024] The blast furnace is equipped with thermocouples (range 1200-2000℃, accuracy ±0.5%), infrared thermometers (range 20-1500℃, accuracy ±0.01%), pressure sensors (range 0-100MPa, accuracy ±0.05Fs), flow sensors (range 0-1000m 3 / h, accuracy ±0.25Fs), industrial cameras (image resolution 1920x1080 pixels, defect recognition range not less than 0.1mm).

[0025] Electric furnace, converter, LF furnace, VD furnace are equipped with thermocouples (range 1200-2000℃, accuracy ±0.5%), pressure sensors (range 0-100MPa, accuracy ±0.05Fs), flow sensors (range 0-1000m 3 / h, accuracy ±0.25Fs), monitoring molten steel temperature and equipment gas flow.

[0026] The continuous casting machine outlet is equipped with an infrared thermometer (range 20-1500℃, accuracy ±0.01%), a pressure sensor (range 0-100MPa, accuracy ±0.05Fs), a flow sensor (range 0-1000m 3 / h, accuracy ±0.25Fs), speed sensor (range 0-500m / s, accuracy ±0.5Fs), infrared sensor (wavelength 800nm-14μm).

[0027] The end face of the continuous casting billet is marked with a laser-printed high-temperature resistant RFID label, which is laser-printed with the brand name, furnace number, production date, contract number, and a QR code for image recognition (size 20×20mm, resolution ≥300dpi).

[0028] An industrial camera (1920×1080 pixels, 0.1mm level defect recognition), a laser displacement sensor (range ±100mm, accuracy ±0.5mm), and a speed sensor (range 0-500m / s, accuracy ±0.5Fs) are installed at the entrance of the heating furnace to identify the QR code, actual length, and surface condition of the continuous casting billet.

[0029] An industrial camera (1920×1080 pixels, 0.1mm level defect recognition) and an infrared thermometer (range 20-1500℃, accuracy ±0.01%) are installed at the entrance of the rolling mill to identify the degree of descaling of the continuous casting billet, whether there are cracking defects, and the starting rolling temperature.

[0030] The finishing mill is equipped with an infrared thermometer (range 20-1500℃, accuracy ±0.01%), a pressure sensor (range 0-100MPa, accuracy ±0.05Fs), a flow sensor (range 0-1000m 3 / h, accuracy ±0.25Fs), speed sensor (range 0-500m / s, accuracy ±0.5Fs), infrared sensor (wavelength 800nm-14μm), which facilitates the application of controlled rolling and controlled cooling technology.

[0031] The head of the rolled material is marked, and a laser-printed high-temperature resistant RFID label is affixed to it. The brand number, furnace number, production date, contract number, and a QR code for image recognition (size 10×10mm or 20×20mm, resolution ≥300dpi) are laser-printed.

[0032] During the network layer deployment process of this embodiment:

[0033] Data information is encapsulated through the MTSPv2.0 protocol and transmitted using IEEE 802.3at dual-ring industrial Ethernet. The redundancy switching mechanism can be automatic or manual, and the automatic redundancy switching time is <50ms.

[0034] IEEE 802.3at dual-ring industrial Ethernet covers sintering plants, blast furnace lines, primary smelting lines, refining lines, continuous casting lines, heating furnaces, continuous rolling lines, finishing lines, chemical testing, physical testing, and flaw detection.

[0035] Data information is transmitted one-way along four independent network paths according to production needs, avoiding data tampering between branches.

[0036] The network layer includes a dynamic traffic cleaning mechanism based on the CNN-LSTM hybrid model to identify abnormal traffic data information (source IP, destination IP, port, packet size, timestamp, etc.), passively defend against DDoS attacks, and ensure the rational allocation of network resources.

[0037] The CNN-LSTM hybrid model included in the network layer combines CNN (convolutional neural network) and LSTM (long short-term memory network) in architecture, uses time windows to convert time series data into a two-dimensional data matrix, and adopts cross entropy as the loss function.

[0038] Data information transmission integrates SM4 encryption and SM3 hash check via the Sec-OPC UA protocol, achieving 100% detection of data tampering.

[0039] During the application layer deployment process of this embodiment:

[0040] The data transmitted by the four network layers are uniformly stored in a temporary storage area. The data in the temporary storage area is delayed and transmitted to the database (data storage end) to facilitate the development of new products and the training of artificial intelligence. Production and sales personnel can receive real-time production and inspection information through the mobile end in a one-way manner, which facilitates smooth production and market development.

[0041] The application layer blockchain deploys 5 consensus nodes: steelmaking MES, steel rolling MES, ERP, quality inspection system, and warehousing OA.

[0042] In the application layer, the digital twin of physical-virtual space mapping is realized through dynamic compensation equations.

[0043] The PBFT consensus mechanism is used in the application layer, generating data blocks every 10 seconds. When nodes fail to reach consensus, abnormal isolation is triggered.

[0044] The data block structure of the application layer blockchain: block hash (32B) + signature (64B) + Nonce (4B) + brand (4B) + furnace number (12B) + production date (4B) + contract number (12B) + timestamp (8B).

[0045] The application layer uses an improved isolation forest algorithm, which can achieve millisecond-level anomaly detection in the face of real-time threat responses.

[0046] The application layer includes Android and PC platforms, and the permission levels are production (read-only), sales staff (read-only), R&D (data labeling, backtracking), and management (full view).

[0047] Some code implementation examples prepared in this embodiment are as follows:

[0048]

[0049]

Claims

1. A full-process tracking system for furnace numbers of special steel production, characterized by: include: The perception layer is used to collect process parameters and physical status data in real time through a multi-source sensor array, transmit the process parameters and physical status data to the network layer in a P2P mode, and laser print high-temperature resistant RFID QR code labels containing brand number, furnace number, production date, and contract number on the end face of the continuous casting billet and the head of the rolled material; The network layer is used to set up 11 independent network paths (N1-N11), implement data encryption transmission based on IEEE802.3at dual-ring industrial Ethernet, use the Sec-OPC UA protocol to integrate SM4 encryption and SM3 hash check, and dynamically clean abnormal traffic through a CNN-LSTM hybrid model to ensure that data cannot be tampered with. Clean data is obtained and unidirectionally transmitted to the application layer through the independent network paths according to production needs; The application layer is used to uniformly store the clean data in a temporary storage area, synchronize it to the database after a delay for artificial intelligence training and new product development, and push production inspection information to the mobile or PC end in a one-way manner through the real-time streaming module. The PBFT consensus mechanism is used to generate blockchain data blocks, combined with the improved isolation forest algorithm to achieve millisecond-level anomaly detection, and a digital twin of the physical-virtual space mapping is constructed through dynamic compensation equations.

2. The full-process tracking system for heat number of special steel production according to claim 1 is characterized in that: The multi-source sensor array of the sensing layer includes: thermocouples, infrared thermometers, pressure sensors, flow sensors, laser displacement sensors, speed sensors, industrial cameras and infrared sensors.

3. The full-process tracking system for heat number of special steel production according to claim 1 is characterized in that: In the network layer, data information is encapsulated through the MTSP v2.0 protocol, and the automatic redundancy switching time is less than 50ms.

4. The full-process tracking system for heat number of special steel production according to claim 1 is characterized in that: In the application layer, the structure of the blockchain data block is: block hash + signature + Nonce + brand + furnace number + production date + contract number + timestamp.

5. A full-process tracking method for furnace numbers in special steel production, characterized in that: include: The sensing step is used to collect process parameters and physical status data in real time through a multi-source sensor array, transmit the process parameters and physical status data to the network layer using a P2P mode, and laser print high-temperature resistant RFID QR code labels containing the brand number, furnace number, production date, and contract number on the end face of the continuous casting billet and the head of the rolled material; The network processing step is used to set up 11 independent network paths (N1-N11), implement data encryption transmission based on IEEE802.3at dual-ring industrial Ethernet, use the Sec-OPC UA protocol to integrate SM4 encryption and SM3 hash check, and dynamically clean abnormal traffic through a CNN-LSTM hybrid model to ensure that data cannot be tampered with. Clean data is obtained and unidirectionally transmitted to the application layer through the independent network paths according to production requirements; The application steps are used to uniformly store the clean data in a temporary storage area, synchronize it to the database after a delay for artificial intelligence training and new product development, and at the same time push production inspection information to the mobile terminal or PC terminal in a unidirectional manner through the real-time streaming module, use the PBFT consensus mechanism to generate blockchain data blocks, combine the improved isolation forest algorithm to achieve millisecond-level anomaly detection, and construct a digital twin of physical-virtual space mapping through dynamic compensation equations.

Images