Cooperative system for intelligent temperature control and thermal fatigue prediction of H13 steel mold and use method of cooperative system

Abstract

The invention discloses an H13 steel mold intelligent temperature control and thermal fatigue prediction cooperation system and a use method thereof, and relates to the field of steel molds, the H13 steel mold intelligent temperature control and thermal fatigue prediction cooperation system comprises an intelligent temperature control subsystem, a thermal fatigue prediction subsystem, a data processing and storage module, a fault diagnosis module, a data interaction and cooperation module and a user management and authority module. According to the invention, the intelligent temperature control subsystem utilizes dual-controller redundancy and dynamic strategy adjustment to realize high-precision dynamic temperature control, the thermal fatigue prediction subsystem is fused with a multi-scale model and an LSTM neural network to realize high-precision prediction, and the data processing module supports efficient noise reduction and cross-domain data conversion. The fault diagnosis module constructs a multi-level threshold detection and intelligent response system, the data interaction module realizes multi-protocol compatibility and full-process collaboration, and the user management module ensures system safety and global adaptation, so that the temperature control precision, the thermal fatigue prediction reliability and the production collaboration efficiency of the mold are integrally improved, the maintenance cost is reduced, and the service life of the mold is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of steel molds, and in particular to a collaborative system for intelligent temperature control and thermal fatigue prediction of an H13 steel mold and a method for using the system. Background Art

[0002] H13 steel molds are industrial molds made of H13 hot work die steel. This type of steel has excellent thermal strength, wear resistance, thermal fatigue performance and hardenability, and is commonly used in high-temperature working conditions such as die-casting molds, hot forging molds, and extrusion molds.

[0003] However, in traditional H13 steel mold production, temperature control accuracy is insufficient and there is a lack of dynamic adjustment mechanism. Thermal fatigue prediction relies on a single model or offline calculation, and cannot integrate multi-source real-time data. Data interaction between subsystems is isolated and coordination is poor, fault response is delayed, and multi-brand equipment has low compatibility and insufficient data processing capabilities. It is difficult to meet the needs of precise mold temperature control and full life cycle management of thermal fatigue in complex industrial scenarios, resulting in low production efficiency and short mold life. Summary of the Invention

[0004] The purpose of the present invention is to provide a collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds and its use method to solve the problems raised by the above background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solutions: a collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds, comprising an intelligent temperature control subsystem, a thermal fatigue prediction subsystem, a data processing and storage module, a fault diagnosis module, a data interaction and collaboration module, and a user management and authority module; The intelligent temperature control subsystem includes: a temperature monitoring module, which installs high-precision temperature sensors at key locations on the H13 steel mold and uses data acquisition cards and communication technology to collect and transmit temperature data in real time; a temperature control module, which uses control algorithms to formulate control strategies based on process requirements and historical data to control heating and cooling devices; a human-computer interaction interface, which allows operators to view temperature data, set parameters, monitor processes, and receive alarms; and a hardware support platform, including the control layer, sensor layer, and execution layer. The thermal fatigue prediction subsystem includes: a data input module, which collects temperature data from the intelligent temperature control subsystem, mold material / structure / workload parameters, environmental data, and dynamic load data. It uses pressure sensors to collect peak pressure and load cycle frequency. The thermal fatigue prediction module, based on a multi-scale fusion model, establishes a material fatigue damage dynamics model at the micro level. It also uses finite element simulation at the macro level to obtain the equivalent plastic strain range. The data is driven by an LSTM neural network to process time series data and output a predicted crack growth rate. The intelligent temperature control subsystem transmits temperature data through the data interaction and collaboration module and receives thermal fatigue prediction results to adjust the control strategy; the thermal fatigue prediction subsystem obtains multi-source data through the data interaction and collaboration module and feeds back the prediction results.

[0006] Preferably, the control layer adopts a dual-controller redundant architecture; the sensor layer deploys thermocouple sensors in the core area of ​​the cavity; and the execution layer heating system adopts a composite solution of electromagnetic induction heating and resistance heating.

[0007] Preferably, the data processing and storage module includes: The data preprocessing module uses sliding average filtering to perform noise reduction and extract key feature parameters; The data storage module deploys a distributed time series database and supports hierarchical data storage and long-term archiving.

[0008] Preferably, the fault diagnosis module includes the following contents: The abnormality detection system sets multi-level thresholds for key parameters such as temperature, pressure, and flow. The temperature parameter sets the normal working range threshold, warning threshold, and emergency shutdown threshold, and an alarm is triggered when the threshold is exceeded; Fault response mechanism, implements three-level alarm, generates fault logs with timestamps, and pushes maintenance guides; Fault log generation: the system automatically generates a detailed fault log with a timestamp each time a fault occurs. The log content includes the specific time and type of fault. Maintenance guide push: For different types of faults, the system has built-in detailed maintenance guides, which are presented in the form of pictures and texts.

[0009] Preferably, the data interaction and collaboration module includes the following: The data interaction mechanism, with multi-protocol support and device compatibility, is used to achieve seamless data interaction between system modules. The controller and data processing and storage modules of the intelligent temperature control subsystem use the OPCUA protocol to transmit structured data. The thermal fatigue prediction subsystem sends prediction results to the intelligent temperature control subsystem and fault diagnosis module via the MQTT protocol. Data transmission priority and timing, as well as priority division, are also provided. In the parameter setting phase of the collaborative workflow, the operator logs into the system's human-computer interaction interface to set parameters, including the temperature control target of the intelligent temperature control subsystem, using the permissions assigned by the user management and permission module. The parameter setting information is then transmitted to the intelligent temperature control subsystem and thermal fatigue prediction subsystem through the data interaction and collaborative module for initial configuration. During the real-time interaction phase, the intelligent temperature control subsystem collects mold temperature data, transmits it to the data processing and storage module, and sends it to the thermal fatigue prediction subsystem after preprocessing. The thermal fatigue prediction subsystem predicts the thermal fatigue state of the mold based on the temperature data, mold parameters, and environmental data, and feeds the prediction results back to the intelligent temperature control subsystem. Fault-control linkage: the fault diagnosis module receives abnormal data marks and operating status data in real time. When a device fault is detected, it issues an alarm signal and transmits the fault information to the intelligent temperature control subsystem. During the maintenance phase, a mold maintenance plan is generated based on the prediction results of the thermal fatigue prediction subsystem and the historical fault data of the fault diagnosis module. The maintenance information is sent to the equipment management department and the terminal devices of maintenance personnel. The maintenance personnel will feed back the maintenance results to the system, update the mold maintenance records and status information, and optimize the thermal fatigue prediction model and fault diagnosis strategy based on the maintenance results.

[0010] Preferably, the user management and permission module supports three levels of user permissions, integrates electronic signatures and operation log audits, and provides a multi-language interface.

[0011] A method for using a collaborative system for intelligent temperature control and thermal fatigue prediction of an H13 steel mold includes the following steps: S1, initialization configuration phase: complete the user login authority allocation, set the parameters of each module and configure the data interaction protocol; S2, real-time operation control stage: collect and process data and predict thermal fatigue status, perform fault diagnosis response and human-machine monitoring; S3, Maintenance phase: Generate maintenance plan based on prediction and failure data, and optimize system model after execution feedback; S4. Key coordination mechanism: clarify the priority and timing of data interaction and implement closed-loop control.

[0012] Preferably, in step S1, setting the parameters of each module and configuring the data interaction protocol specifically includes: Temperature control parameter setting: set the temperature control target in the human-machine interface of the intelligent temperature control subsystem; Thermal fatigue prediction parameter setting: Configure the prediction model parameters in the thermal fatigue prediction subsystem, and set the fatigue damage rate threshold and remaining life warning threshold; Fault diagnosis threshold setting: Customize multi-level thresholds in the fault diagnosis module; Data interaction configuration: Configure the communication protocol through the data interaction and collaboration module.

[0013] Compared with the prior art, the present invention has the following beneficial effects: In the present invention, the system achieves high-precision dynamic temperature control (dual controller redundancy and dynamic strategy adjustment) through an intelligent temperature control subsystem. The thermal fatigue prediction subsystem integrates multi-scale models and LSTM neural networks to achieve high-precision prediction. The data processing module supports efficient noise reduction and cross-domain data conversion. The fault diagnosis module builds a multi-level threshold detection and intelligent response system. The data interaction module achieves multi-protocol compatibility and full-process collaboration. The user management module ensures system security and global adaptation, thereby improving the mold temperature control accuracy, thermal fatigue prediction reliability and production collaboration efficiency as a whole, reducing maintenance costs and extending mold life. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of a collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds according to the present invention; Figure 2 This is a composition diagram of a collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds according to the present invention. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the implementation regulations described are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0016] Example 1: Reference Figure 1-Figure 2 Shown: A collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds, including an intelligent temperature control subsystem, a thermal fatigue prediction subsystem, a data processing and storage module, a fault diagnosis module, a data interaction and collaboration module, and a user management and authority module.

[0017] 1. Intelligent temperature control subsystem, including the following: (1) Temperature monitoring module: High-precision temperature sensors (such as thermocouple sensors, infrared temperature sensors) are installed at key parts of the H13 steel mold, and temperature data is collected and transmitted in real time through data acquisition cards and communication technology.

[0018] (2) Temperature control module: Using advanced control algorithms (such as fuzzy control and neural network control), the control strategy is formulated in combination with process requirements and historical data to control heating (electric heating rods, induction heating, etc.) and cooling (water cooling, oil cooling, etc.) devices to achieve precise temperature control.

[0019] (3) Human-computer interaction interface: Provides a friendly interface to facilitate operators to view temperature data, set parameters, monitor processes and receive alarm information.

[0020] (4) Hardware support platform, specifically: 1. Control layer: It adopts a dual-controller redundant architecture (main controller + backup controller), with a switching time of less than 10ms, an integrated dedicated motion control card (such as ACSSPiiPlus), and supports precise PID control of cooling water flow (adjustment accuracy ±2%).

[0021] 2. Sensor layer: Thermocouple sensors (type K, accuracy ±1.5°C) are deployed in the core area of ​​the cavity (with a 50mm grid spacing). Infrared thermal imagers (temperature measurement range -20°C to 600°C, resolution 0.1°C) cover 90% of the mold surface. Vibration and acceleration sensors are deployed in combination to identify resonance risks (frequency deviation >5% triggers an early warning).

[0022] 3. Execution layer: The heating system adopts a composite solution of electromagnetic induction heating (efficiency > 90%) and resistance heating (accuracy ±1°C), and supports independent temperature control in different zones (maximum number of zones: 16). The intelligent valve group of the cooling system integrates flow (accuracy ±1%) and pressure (accuracy ±0.5%) sensors to achieve dynamic balance control of the cooling medium flow.

[0023] Specifically, the control strategy is dynamically adjusted: Receive fatigue damage rate data (e.g. >0.02mm / thousand cycles) from the thermal fatigue prediction subsystem and automatically tighten the temperature control accuracy (from ±5°C to ±3°C).

[0024] When the fault diagnosis module reports a cooling valve group leakage fault, the temperature control system actuator layer automatically switches to the backup valve group and sends a sensor calibration request to the data processing module.

[0025] 2. Thermal fatigue prediction subsystem, including the following contents: (1) Data input module: including temperature data of the intelligent temperature control subsystem, material / structure / working load parameters of the mold, environmental data and dynamic load data (peak pressure, load cycle frequency, etc. collected through pressure sensors).

[0026] (2) The thermal fatigue prediction module is based on a multi-scale fusion model. At the micro level, it establishes a material fatigue damage dynamics model based on the Arrhenius equation. At the macro level, it combines finite element simulation to obtain the equivalent plastic strain range. The data-driven LSTM neural network is used to process time series data and output the crack growth rate prediction value. This module has the following advantages: Model update mechanism: supports online incremental learning, and automatically triggers retraining when the prediction error exceeds 15% for three consecutive cycles. The data sources include real-time production and virtual simulation data (generated by the digital twin system).

[0027] Life Management: Generate mold thermal fatigue life curves (remaining life accuracy ±10%), develop preventive maintenance plans, mark high fatigue risk areas, and output local strengthening recommendations.

[0028] Process optimization: Feedback to the production process system to adjust parameters, reduce the thermal fatigue damage rate, establish a process parameter-fatigue damage mapping relationship library, and support pre-evaluation of new mold process solutions (prediction error <15%).

[0029] Specifically, multi-source data fusion input: The temperature data (including the surface temperature field distribution of the infrared thermal imager) and the pressure sensor load data (dynamic stress-strain curve) provided by the data processing module are used for thermal-mechanical coupling calculation in the prediction model.

[0030] The real-time control parameters (such as heating power and cooling flow) output by the edge computing layer are used as environmental variables to input the prediction model to correct the fatigue damage calculation coefficient (correction accuracy ±8%).

[0031] 3. Data processing and storage module, including the following: (1) Data preprocessing module: Sliding average filtering (window size 50ms) and wavelet denoising (db4 wavelet basis) are used to reduce the noise of raw temperature (accuracy ±0.5°C), load (resolution 0.1MPa), and vibration (sensitivity 100mV / g) data, and more than 20 key characteristic parameters are extracted.

[0032] (2) Data storage module: Deploy a distributed time series database (such as InfluxDB) to store high-frequency real-time data (sampling rate 100Hz, storage period 30 days), and a relational database (MySQL) to store structured data, supporting data hierarchical storage and long-term archiving.

[0033] This module has the following advantages: providing data services, API interfaces and visual queries, integrating data quality assessment modules, and real-time monitoring of sensor data validity.

[0034] Cross-domain data conversion: Convert the analog signal of the temperature control system (thermocouple mV value) into an engineering value (°C), and map the fatigue life prediction result (hours) of the prediction subsystem into the number of production cycles (times).

[0035] Provides a standardized data interface for the fault diagnosis module and supports unified processing of sensor signals from different brands.

[0036] 4. Fault diagnosis module, including the following contents: (1) Anomaly detection system, mainly threshold detection: Multi-level thresholds can be set for key parameters such as temperature, pressure, and flow. For example, temperature parameters can be configured with a normal operating range threshold, a warning threshold, and an emergency shutdown threshold. When the cavity temperature of an H13 steel mold exceeds the upper limit of the normal operating range (e.g., exceeding the set value by 10°C), an alarm is triggered. If it continues to rise to the emergency shutdown threshold (e.g., exceeding the set value by 20°C), a shutdown protection is immediately initiated. Regarding pressure parameters, an alarm is triggered when the mold pressure exceeds 110% of the rated pressure, and an emergency shutdown occurs when it exceeds 120%. Regarding the cooling system's flow rate parameters, if the actual flow rate falls below 80% of the set flow rate, an insufficient flow alarm is issued. If it falls below 60%, it is considered a serious fault and the machine is shut down. Threshold settings support user customization. The parameter thresholds can be flexibly adjusted in the human-computer interaction interface according to different mold process requirements and production scenarios. At the same time, the system has a threshold rationality verification function to avoid false alarms or missed alarms due to unreasonable threshold settings.

[0037] (2) Fault response mechanism: implement three-level alarm, generate fault logs with timestamps, and push maintenance guides.

[0038] Level 1 Warning (Yellow): Triggered when a parameter approaches a threshold or shows a slight abnormal trend. For example, if the mold temperature reaches 90% of the warning threshold, or if a small number of abnormal points appear in sensor data but do not affect normal system operation, the system will issue a warning on the human-machine interface by flashing a yellow warning light and pop-up prompts. It will also send a notification via SMS or app push notification to the relevant operator, reminding them to pay attention to the equipment status and prepare for troubleshooting. Level 2 Fault (Orange): Triggered when a parameter exceeds a threshold or the abnormality is severe, potentially impacting normal equipment operation but not yet reaching emergency shutdown conditions. For example, if the mold temperature exceeds the warning threshold or an actuator experiences partial malfunction (e.g., a cooling valve block fails to regulate flow but still maintains a certain cooling effect), the system issues a fault alarm with a flashing orange warning light and an audible buzzer. The system also automatically records the time of the fault and related parameter information, and displays a detailed fault description and preliminary action suggestions on the human-machine interface. At this point, the system automatically switches to a backup device or activates a redundant control strategy (e.g., switching to a backup cooling circuit) to maintain basic equipment operation. Level 3 Emergency Shutdown (Red): Triggered when a serious fault occurs that could threaten the mold, equipment, or personnel safety. For example, if the mold temperature exceeds the emergency shutdown threshold or the actuator fails completely, the system immediately emits a strong red warning light and a high-decibel alarm. It also automatically cuts off the power supply to the equipment, halting mold operation to prevent further escalation of the fault. The system also records the emergency shutdown event in detail and notifies senior management and maintenance teams via email and text message, allowing for prompt troubleshooting. Fault Log Generation: The system automatically generates a detailed, time-stamped fault log for each fault. This log includes the exact time the fault occurred, the fault type (e.g., temperature anomaly, sensor failure, actuator failure), equipment operating parameters prior to the fault (e.g., real-time data on temperature, pressure, load, vibration, etc.), the alarm level at the time of the fault, and the system's response measures. Fault logs are stored in a relational database within the data processing and storage module, allowing operators to query, filter, and export them through a human-computer interface. The system also provides fault log analysis, which uses historical fault data to compile statistics on fault frequency and type distribution, providing data support for equipment maintenance and improvements. Maintenance guide push: The system has built-in detailed maintenance guides for different types of faults. The maintenance guide is presented in a graphic and text format, including fault cause analysis, maintenance step instructions, required tools and parts list, etc. For example, when a sensor fault alarm sounds, the maintenance guide will prompt the operator to check whether the sensor wiring is loose or whether the sensor probe is damaged, and provide specific operating steps and precautions for replacing the sensor. The maintenance guide can be viewed directly through the human-computer interaction interface, or it can be pushed to maintenance personnel through a mobile phone APP to facilitate on-site maintenance operations. In addition, the system also supports maintenance personnel to provide feedback on the accuracy and practicality of the maintenance guide after completing the maintenance, so as to continuously optimize the content of the maintenance guide.

[0039] 5. Data interaction and collaboration module, including the following contents: (1) Data interaction mechanism, specifically: 1. Multi-protocol support and device compatibility: OPCUA protocol application: OPCUA (Open Platform Communications Unified Architecture), as a standard communication protocol in the field of industrial automation, is used to achieve seamless data interaction between various modules of the system. In this system, data is transmitted between the controller of the intelligent temperature control subsystem (such as the Siemens S7-1500) and the data processing and storage module via the OPCUA protocol. It can efficiently transmit structured data, such as the real-time temperature and control parameters of the mold, and supports secure data interaction across platforms and networks, ensuring interoperability between devices of different brands and types. For example, even if the temperature control subsystem uses sensors and actuators from different manufacturers, through the OPCUA protocol, the data generated by these devices can be uniformly formatted and accurately transmitted to the data processing module for subsequent analysis and processing. MQTT Protocol Application: MQTT (Message Queuing Telemetry Transport) is a lightweight, low-bandwidth, and reliable messaging protocol suitable for systems with high real-time requirements and relatively small data volumes. The thermal fatigue prediction subsystem sends prediction results (such as remaining life and crack growth rate) to the intelligent temperature control subsystem and fault diagnosis module via MQTT. This protocol, based on a publish / subscribe model and supporting asynchronous communication, effectively reduces network congestion and ensures timely and accurate transmission of critical data to the target modules even in unstable industrial network environments. 2. Plug-and-Play Implementation: To enable plug-and-play integration for devices from multiple brands, the system has built a device description file library and a driver adaptation layer. When a new device is connected to the system, the operator simply connects the device to the designated network port and imports the device description file (containing information such as device functions, communication parameters, and data format) into the system. The system's driver adaptation layer automatically identifies the device type and loads the corresponding driver, completing the rapid integration of the device into the system. For example, a newly connected infrared thermal imager can quickly transmit collected temperature field data to the data processing and storage module for analysis through this mechanism, without complex programming and configuration. 3. Data transmission priority and timing, including: Prioritization: The system divides data transmission into three priorities. High priority: Emergency control commands (such as fault shutdown signals and safety interlock signals). This type of data requires extremely low latency (response time <50ms) to ensure that the equipment can react quickly in an emergency and protect personnel and equipment safety. The system uses a dedicated priority queue and real-time scheduling algorithm to prioritize the processing and transmission of this type of data. Medium priority: Real-time monitoring data, such as mold temperature, pressure, and vibration sensor data, is transmitted at a high frequency (e.g., temperature data has a 100Hz sampling rate) and requires accurate transmission within a short timeframe (10ms level of interaction) to ensure the system can promptly understand the mold's operating status. The system ensures stable transmission of real-time data by optimizing network bandwidth allocation and using data compression technologies. Low priority: Non-real-time data such as historical data synchronization and system configuration information updates are transmitted asynchronously during network idle periods to avoid occupying the transmission bandwidth of critical data. Timing management: Establish a strict data interaction timing diagram to clearly define the order and time intervals for data transmission between modules. For example, at the beginning of each control cycle (e.g., 100ms), the intelligent temperature control subsystem first transmits the collected real-time temperature data to the data processing and storage module. The data processing module completes data preprocessing within 20ms and then sends the characteristic data to the thermal fatigue prediction subsystem. The thermal fatigue prediction subsystem completes the prediction calculation within 30ms and feeds the results back to the intelligent temperature control subsystem so that the temperature control subsystem can adjust the control strategy in the next control cycle. (2) Collaborative workflow, specifically: 1. Parameter Setup: Before the mold is put into use, the operator logs into the system's human-computer interaction interface (HMI) using the permissions assigned by the User Management and Permissions module to set parameters. These settings include the temperature control targets (such as heating temperature, cooling temperature range, and heating rate) for the intelligent temperature control subsystem, prediction parameters (such as fatigue life calculation model parameters and crack propagation prediction period) for the thermal fatigue prediction subsystem, and alarm thresholds (such as temperature warning thresholds and pressure failure thresholds) for the fault diagnosis module. This parameter setting information is transmitted to the corresponding subsystems for initial configuration via the Data Interaction and Collaboration Module, according to established priorities and timing. The system also verifies the rationality of the set parameters, such as whether the temperature control target exceeds the tolerance of the mold material. If any unreasonable parameters are found, the operator is promptly prompted to make corrections. 2. Real-time Interaction Stage: Temperature-Fatigue Collaborative Control: During mold operation, the intelligent temperature control subsystem collects mold temperature data at a frequency of 100Hz and transmits it in real time to the data processing and storage module via the data interaction and collaboration module. The data processing module performs pre-processing on the temperature data, including filtering and feature extraction, before sending the processed data to the thermal fatigue prediction subsystem. Based on temperature data, mold parameters, and environmental data, the thermal fatigue prediction subsystem uses a multi-scale fusion model to predict the mold's thermal fatigue status (such as fatigue damage degree and remaining life) in real time and feeds the prediction results back to the intelligent temperature control subsystem. If the predicted thermal fatigue damage rate of the mold accelerates (e.g., exceeding 0.02mm / 1,000 cycles) or the remaining life falls below a certain threshold (e.g., 20% of the design life), the intelligent temperature control subsystem automatically adjusts the temperature control strategy, such as reducing the temperature fluctuation range (from ±5°C to ±3°C) and extending the cooling phase by 15%, to mitigate the mold's thermal fatigue progression. Fault-control linkage: The fault diagnosis module receives abnormal data markers and operational status data from each subsystem in real time from the data processing and storage module. When a device fault (such as a sensor failure or actuator anomaly) is detected, the fault diagnosis module immediately issues an alarm signal and transmits fault information (including fault type, occurrence time, and severity) to the intelligent temperature control subsystem via the data interaction and collaboration module. Based on the fault information, the intelligent temperature control subsystem automatically implements appropriate control measures. For example, if a cooling valve group fails, it switches to a backup valve group and adjusts heating power to compensate for temperature deviations. If a sensor fails, it activates an adjacent sensor data fusion algorithm to maintain system operation. Simultaneously, the fault diagnosis module transmits fault information to the production scheduling platform and maintenance management system to facilitate the allocation of maintenance resources and adjust production plans. 3. Maintenance phase: Based on the prediction results of the thermal fatigue prediction subsystem and the historical fault data of the fault diagnosis module, the system generates a maintenance plan for the mold. Maintenance information (such as recommended maintenance time, maintenance content, required parts, etc.) is sent to the terminal devices (such as mobile phone apps and tablets) of the equipment management department and maintenance personnel through the data interaction and collaboration module. When performing maintenance operations, maintenance personnel can view the historical operation data and fault records of the mold in real time through the terminal device to assist in troubleshooting and repair work. After the maintenance is completed, the maintenance personnel will feedback the maintenance results (such as a description of the maintenance process, replacement parts information, equipment operation test results, etc.) to the system through the data interaction and collaboration module. The system updates the maintenance records and status information of the mold, and optimizes the thermal fatigue prediction model and fault diagnosis strategy based on the maintenance results.

[0040] 6. User Management and Permission Module: Supports three levels of user permissions, integrates electronic signatures and operation log audits, provides a multi-language interface, and adapts to global production.

[0041] Example 2: According to Figure 1-Figure 2 As shown, a method for using a collaborative system for intelligent temperature control and thermal fatigue prediction of an H13 steel mold includes the following steps: S1, initialization configuration phase, specifically including: S11. User login and authority allocation: Log in to the system through the user management and authority module. The system supports three levels of authority (administrator, engineer, operator). Different authorities correspond to different operating scopes (for example, administrators can modify system parameters, while operators can only view data). Electronic signatures and operation log audits are integrated to ensure traceability of operations. The system supports multi-language interfaces to adapt to global production needs.

[0042] S12, parameter setting and system initialization: Temperature control parameter setting: Set the temperature control target (such as heating temperature, cooling temperature range, and heating rate) in the human-machine interface of the intelligent temperature control subsystem, and support independent temperature control of each partition (up to 16 partitions).

[0043] Thermal fatigue prediction parameter setting: Configure the prediction model parameters (such as Arrhenius equation parameters and LSTM neural network training cycle) in the thermal fatigue prediction subsystem, set the fatigue damage rate threshold (such as >0.02mm / thousand cycles) and the remaining life warning threshold (such as <20% of the design life).

[0044] Fault diagnosis threshold setting: Customize multi-level thresholds in the fault diagnosis module (e.g., the temperature warning threshold is the set value + 10°C, the emergency shutdown threshold is the set value + 20°C; if the pressure exceeds 110% of the rated value, a warning will be issued, and if it exceeds 120%, a shutdown will be issued). The system automatically verifies the rationality of the thresholds to avoid false alarms / missing alarms.

[0045] Data interaction configuration: Configure the communication protocol through the data interaction and collaboration module (OPCUA for structured data transmission, MQTT for real-time prediction result push). When connecting a new device, import the description file to achieve "plug and play" (such as connecting an infrared thermal imager).

[0046] S2, real-time operation control stage, specifically including: S21. Data collection and collaborative processing: Real-time temperature monitoring: K-type thermocouples (accuracy of ±1.5°C, distributed on a 50mm grid) are deployed in the core area of ​​the mold cavity at the sensor level. Infrared thermal imagers (measuring -20°C to 600°C, with a resolution of 0.1°C) cover 90% of the mold surface, collecting temperature data in real time at a sampling rate of 100Hz. During data processing, the raw data undergoes sliding average filtering (50ms window) and wavelet denoising (DB4 basis) in the data processing and storage module. Over 20 characteristic parameters (such as temperature field distribution and temperature gradient) are extracted and stored in InfluxDB (high-frequency data) and MySQL (structured data).

[0047] Thermal Fatigue State Prediction: Input data: Temperature data from the temperature control subsystem and load data (dynamic stress-strain curves) from the pressure sensor are fed into the thermal fatigue prediction subsystem. Multi-scale model calculations are performed, using the Arrhenius equation to calculate fatigue damage dynamics at the micro level. Finite element simulations are used to obtain plastic strain at the macro level. Combined with an LSTM neural network to process time series data, the system outputs the crack growth rate and remaining life (with an accuracy of ±10%). During the dynamic feedback process, if the predicted fatigue damage rate exceeds 0.02 mm / 1,000 cycles, the intelligent temperature control subsystem automatically tightens the temperature control accuracy (from ±5°C to ±3°C) or extends the cooling time by 15% to mitigate fatigue progression.

[0048] S22. Fault diagnosis and emergency response, specifically including: Anomaly detection, including threshold triggering, triggers alarms when key parameters (temperature, pressure, flow) exceed preset thresholds. For example, a yellow warning is triggered when the temperature exceeds the warning threshold (setpoint + 10°C), and a red shutdown is triggered when it exceeds the emergency threshold (setpoint + 20°C). A warning is triggered when the cooling flow falls below 80% of the setpoint, and a shutdown is triggered when it falls below 60%.

[0049] A three-level alarm system: Level 1 Warning (yellow): Parameters approaching thresholds or experiencing minor anomalies (e.g., sensor data fluctuations) trigger a flashing warning light and a text message to alert operators. Level 2 Fault (orange): Parameters exceeding limits or partial actuator failure (e.g., cooling valve block regulation failure) triggers automatic switching to a backup device (e.g., a backup valve block), and the interface displays a description of the fault and recommended actions. Level 3 Emergency Shutdown (red): A serious fault threatening safety (e.g., actuator failure) triggers immediate power outage, visual and audible alarms, and email notifications to senior management and the maintenance team.

[0050] Fault linkage control: The fault diagnosis module transmits fault information (type, time, and severity) to the temperature control subsystem via the data exchange module, triggering emergency measures (such as switching to a backup valve group if a cooling valve group fails, or adjusting heating power to compensate for temperature deviation). The production scheduling platform and maintenance system are also notified simultaneously to adjust production plans and allocate maintenance resources.

[0051] S23, human-computer interaction and real-time monitoring. The intelligent temperature control subsystem's human-computer interface allows users to view temperature curves, pressure-strain dynamic graphs, and thermal fatigue life curves in real time; receive alarm information (pop-up windows, audio and visual prompts); and manually intervene in control parameters (such as temporarily adjusting cooling flow). The data processing module provides an API interface and visual query, supporting real-time data charting (such as temperature field distribution cloud maps and fatigue damage trend curves) to assist operators in decision-making.

[0052] S3, maintenance phase, specifically including: S31. Preventive Maintenance Plan Generation: Based on the data, the remaining life prediction results (accuracy ±10%) from the thermal fatigue prediction subsystem, and the historical fault statistics (such as fault frequency and type distribution) from the fault diagnosis module, the system automatically generates maintenance recommendations (such as recommended maintenance time, high fatigue risk areas requiring reinforcement, and a list of replacement parts). These recommendations are then pushed to the equipment management department and maintenance personnel terminals (mobile app, tablet) via the data interaction module.

[0053] S32, Maintenance execution and feedback: Maintenance personnel view mold historical data, fault logs and maintenance guides (graphic steps, tool lists) through the terminal, and perform maintenance tasks (such as sensor calibration, valve group replacement, and local strengthening treatment). After maintenance is completed, the maintenance record (replacement part information, test results) is uploaded through the data interaction module. The system updates the mold status information and optimizes the thermal fatigue prediction model and fault diagnosis strategy based on the maintenance data.

[0054] S33. System Optimization and Model Updates: The thermal fatigue prediction subsystem supports online incremental learning. When the prediction error exceeds 15% for three consecutive cycles, model retraining is automatically triggered. Data sources include real-time production data and digital twin simulation data. Based on a database of mapping relationships between fatigue damage and process parameters, the system pre-evaluates process plans for new molds (prediction error <15%) and provides feedback to the production system to adjust parameters (such as reducing load frequency and optimizing cooling paths), thereby reducing thermal fatigue damage at the source.

[0055] S4. Key coordination mechanisms prioritize data interaction, ensuring a response time of less than 50ms for emergency control commands (such as shutdown signals), 10ms for real-time monitoring data (temperature and pressure), and asynchronous transmission of historical data. Within a control cycle (e.g., 100ms), the following steps are followed: temperature control subsystem → data processing module (20ms preprocessing) → thermal fatigue prediction module (30ms calculation) → temperature control subsystem (next-cycle adjustment strategy), ensuring real-time closed-loop control.

[0056] Through the above process, the system realizes the precise temperature control of H13 steel molds and the dynamic coordinated management of thermal fatigue status, thereby improving mold life and production efficiency and reducing the risk of failure.

[0057] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds, characterized by: It includes intelligent temperature control subsystem, thermal fatigue prediction subsystem, data processing and storage module, fault diagnosis module, data interaction and collaboration module and user management and authority module; The intelligent temperature control subsystem includes: a temperature monitoring module, which installs high-precision temperature sensors at key locations on the H13 steel mold and uses data acquisition cards and communication technology to collect and transmit temperature data in real time; a temperature control module, which uses control algorithms to formulate control strategies based on process requirements and historical data to control heating and cooling devices; a human-computer interaction interface, which allows operators to view temperature data, set parameters, monitor processes, and receive alarms; and a hardware support platform, including the control layer, sensor layer, and execution layer. The thermal fatigue prediction subsystem includes: a data input module, which collects temperature data from the intelligent temperature control subsystem, mold material / structure / workload parameters, environmental data, and dynamic load data. It uses pressure sensors to collect peak pressure and load cycle frequency. The thermal fatigue prediction module, based on a multi-scale fusion model, establishes a material fatigue damage dynamics model at the micro level. It also uses finite element simulation at the macro level to obtain the equivalent plastic strain range. The data is driven by an LSTM neural network to process time series data and output a predicted crack growth rate. The intelligent temperature control subsystem transmits temperature data through the data interaction and collaboration module and receives thermal fatigue prediction results to adjust the control strategy; the thermal fatigue prediction subsystem obtains multi-source data through the data interaction and collaboration module and feeds back the prediction results.

2. The collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds according to claim 1 is characterized by: The control layer adopts a dual-controller redundant architecture; the sensor layer deploys thermocouple sensors in the core area of ​​the cavity; the execution layer heating system adopts a composite solution of electromagnetic induction heating and resistance heating.

3. The collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel mold according to claim 1 is characterized in that: The data processing and storage modules include: The data preprocessing module uses sliding average filtering to perform noise reduction and extract key feature parameters; The data storage module deploys a distributed time series database and supports hierarchical data storage and long-term archiving.

4. The collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel mold according to claim 1 is characterized in that: The fault diagnosis module includes the following: The abnormality detection system sets multi-level thresholds for key parameters such as temperature, pressure, and flow. The temperature parameter sets the normal working range threshold, warning threshold, and emergency shutdown threshold, and an alarm is triggered when the threshold is exceeded; Fault response mechanism, implements three-level alarm, generates fault logs with timestamps, and pushes maintenance guides; Fault log generation: the system automatically generates a detailed fault log with a timestamp each time a fault occurs. The log content includes the specific time and type of fault. Maintenance guide push: For different types of faults, the system has built-in detailed maintenance guides, which are presented in the form of pictures and texts.

5. The collaborative system of intelligent temperature control and thermal fatigue prediction for H13 steel mold according to claim 1 is characterized in that: The data interaction and collaboration module includes the following: The data interaction mechanism, multi-protocol support and device compatibility are used to achieve seamless data interaction between the various modules of the system. The controller of the intelligent temperature control subsystem and the data processing and storage module use the OPCUA protocol to transmit data and structured data. Through the MQTT protocol, the thermal fatigue prediction subsystem sends the prediction results to the intelligent temperature control subsystem and fault diagnosis module; Data transmission priority and timing, prioritization; In the parameter setting phase of the collaborative workflow, the operator logs into the system's human-computer interaction interface to set parameters, including the temperature control target of the intelligent temperature control subsystem, using the permissions assigned by the user management and permission module. The parameter setting information is then transmitted to the intelligent temperature control subsystem and thermal fatigue prediction subsystem through the data interaction and collaborative module for initial configuration. During the real-time interaction phase, the intelligent temperature control subsystem collects mold temperature data, transmits it to the data processing and storage module, and sends it to the thermal fatigue prediction subsystem after preprocessing. The thermal fatigue prediction subsystem predicts the thermal fatigue state of the mold based on the temperature data, mold parameters, and environmental data, and feeds the prediction results back to the intelligent temperature control subsystem. Fault-control linkage: the fault diagnosis module receives abnormal data marks and operating status data in real time. When a device fault is detected, it issues an alarm signal and transmits the fault information to the intelligent temperature control subsystem. During the maintenance phase, a mold maintenance plan is generated based on the prediction results of the thermal fatigue prediction subsystem and the historical fault data of the fault diagnosis module. The maintenance information is sent to the equipment management department and the terminal devices of maintenance personnel. The maintenance personnel will feed back the maintenance results to the system, update the mold maintenance records and status information, and optimize the thermal fatigue prediction model and fault diagnosis strategy based on the maintenance results.

6. The collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel mold according to claim 1 is characterized in that: The user management and permission module supports three levels of user permissions, integrates electronic signatures and operation log audits, and provides a multi-language interface.

7. A method for using a collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds, characterized in that: A collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel molds according to any one of claims 1 to 6 is used, comprising the following steps: S1, initialization configuration phase: complete the user login authority allocation, set the parameters of each module and configure the data interaction protocol; S2, real-time operation control stage: collect and process data and predict thermal fatigue status, perform fault diagnosis response and human-machine monitoring; S3, Maintenance phase: Generate maintenance plan based on prediction and failure data, and optimize system model after execution feedback; S4. Key coordination mechanism: clarify the priority and timing of data interaction and implement closed-loop control.

8. The method for using the collaborative system for intelligent temperature control and thermal fatigue prediction of H13 steel mold according to claim 7, characterized in that: In step S1, set the parameters of each module and configure the data interaction protocol, including: Temperature control parameter setting: set the temperature control target in the human-machine interface of the intelligent temperature control subsystem; Thermal fatigue prediction parameter setting: Configure the prediction model parameters in the thermal fatigue prediction subsystem, and set the fatigue damage rate threshold and remaining life warning threshold; Fault diagnosis threshold setting: Customize multi-level thresholds in the fault diagnosis module; Data interaction configuration: Configure the communication protocol through the data interaction and collaboration module.

Images