Hot press molding-quality identification integrated system and method for self-bonding motor iron core

Abstract

The invention relates to the technical field of intelligent manufacturing, in particular to a self-bonding motor iron core hot-press forming-quality identification integrated system and method, and the method comprises the steps that a hot-press parameter sensing module collects real-time hot-press parameters, binder state data, lamination characteristic parameters and environment variables in the self-bonding motor iron core hot-press forming process; the dynamic twinning modeling module constructs a self-bonding motor iron core hot-press forming total-factor digital twinning body, and simulates a heat-force-curing coupling effect by using a finite element-curing coupling algorithm; the intelligent quality identification center locates a quality anomaly source through multi-physics field simulation data in combination with an improved Transform model; the integrated regulation and control unit predicts the residual life of the key forming part and dynamically adjusts hot press forming parameters and a quality correction strategy; and the visual interaction module synchronously displays the execution effect of the regulation and control instruction. Therefore, the problems that in the prior art, parameter perception in the hot pressing process is not comprehensive, and quality identification means are lagged are solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, specifically to an integrated system and method for hot pressing and quality identification of self-adhesive motor cores. Background Technology

[0002] Existing technologies have formed a core technology system covering hot pressing and basic quality testing, laying the foundation for the mass production and performance assurance of motor cores. In the forming process, existing hot pressing technology, through precise control of parameters such as temperature, pressure, and holding time, promotes the curing reaction of the self-adhesive silicon steel sheet interlayer coating, achieving integrated stacking of the core. Compared with traditional riveting and welding processes, this effectively avoids physical damage to the core structure, preserves the magnetic properties of the electrical steel to the greatest extent, and significantly improves the energy efficiency and noise reduction of the motor. In the quality testing process, multi-dimensional testing methods have been developed, including pre-testing of the viscosity and solid content of the self-adhesive coating, intermediate testing of the adhesion and insulation performance of the silicon steel sheets, and post-testing of indicators such as the bonding strength, dimensional accuracy, and iron loss of the finished core. Some companies have also built professional laboratories to achieve full-process testing, providing technical support for basic product quality screening.

[0003] Existing technologies related to hot pressing molding and quality assessment of self-bonding motor cores suffer from numerous key deficiencies, making it difficult to meet the demands of high-precision and intelligent production. The sensing of hot pressing process parameters is incomplete and lacks real-time performance. Current technologies cannot simultaneously collect comprehensive data on the entire temperature field of the mold, the temperature gradient between core layers, the curing degree of the adhesive, the alignment of the laminates, and the molding environment. They can only monitor some basic hot pressing parameters, resulting in blind spots in the control of the molding process. The lack of full-element digital twin modeling capabilities prevents the construction of a digital twin encompassing the core, mold, and hot pressing mechanism. Furthermore, it cannot simulate the thermo-mechanical-curing coupling effect using finite element-curing coupling algorithms, making it difficult to predict molding quality risks under different working conditions. Quality assessment methods are outdated and lack accuracy. Without combining multiphysics simulation data and AI algorithms, they cannot comprehensively identify various defect modes such as uneven density and insufficient bonding strength, and it is even more difficult to accurately locate the physical location, scope of influence, and inducing factors of quality anomalies. The disconnect between molding control and component lifespan management is a significant issue. Existing technologies often employ fixed hot-pressing parameters, lacking the ability to dynamically adjust parameters such as temperature gradients and pressure loading rates based on quality assessment results. Furthermore, they cannot predict the remaining lifespan of critical components like molds and heating elements, hindering the prevention of quality issues caused by component aging. The lack of visual interactive mechanisms linking molding and quality control prevents real-time rendering of the hot-pressing process, temperature / stress distribution, and curing progress. Operators struggle to intuitively grasp the quality status and the effectiveness of control commands, hindering timely intervention and process optimization. Summary of the Invention

[0004] This application provides an integrated system and method for hot pressing molding and quality identification of self-adhesive motor cores, in order to solve the problems of incomplete perception of hot pressing process parameters and outdated quality identification methods in the prior art.

[0005] The first aspect of this application provides an integrated system for hot pressing and quality identification of self-bonded motor cores, comprising: a hot pressing parameter sensing module, a dynamic twin molding modeling module, an intelligent quality identification center, an integrated control unit, and a visualization interaction module. The hot pressing parameter sensing module collects real-time hot pressing parameters, adhesive state data, lamination characteristic parameters, and environmental variables during the hot pressing process of the self-bonded motor core. The dynamic twin molding modeling module constructs a digital twin of all elements of the hot pressing process of the self-bonded motor core, simulating the thermo-mechanical-curing coupling effect based on the physical properties of hot pressing and the curing kinetic parameters of the adhesive, using a finite element-curing coupling algorithm. The intelligent quality identification center identifies core forming quality defect patterns and locates quality anomaly sources by combining multi-physics simulation data with an improved Transformer model. The integrated control unit predicts the remaining lifespan of key formed components based on the quality identification results and real-time monitoring data, and dynamically adjusts hot pressing parameters and quality correction strategies. The visualization interaction module constructs an immersive three-dimensional forming-quality linkage scene, rendering the hot pressing process and quality status in real time, and synchronously displaying the execution effect of control commands.

[0006] Preferably, the hot-pressing parameter sensing module includes an infrared temperature measurement array, a high-precision pressure sensor group, an ultrasonic adhesive curing detector, a lamination alignment sensor, and an environmental monitoring unit. The infrared temperature measurement array is used to capture the global temperature field of the mold and the temperature gradient between the iron core layers in real time. The high-precision pressure sensor group is used to collect dynamic pressure curves and holding pressure stability data during the hot-pressing process. The ultrasonic adhesive curing detector is used to obtain the degree of adhesive curing and the interface bonding state. The lamination alignment sensor is used to monitor the radial offset and interlayer misalignment of the iron core laminations. The environmental monitoring unit is used to collect data on the molding environment temperature, humidity, and dust concentration.

[0007] Preferably, the dynamic twin forming modeling module includes a forming twin construction unit and a thermo-mechanical-curing coupling simulation unit. The forming twin construction unit is used to construct a full-element digital twin containing the core, mold, and hot pressing mechanism based on the core design drawings, lamination parameters, and mold 3D scanning data. The thermo-mechanical-curing coupling simulation unit is used to simulate the temperature conduction, stress distribution, and adhesive curing reaction process of the core under different hot pressing conditions by integrating finite element analysis and adhesive curing kinetic models, and using a finite element-curing coupling algorithm, based on the physical properties of hot pressing and the curing kinetic parameters of the adhesive, and outputting thermo-mechanical-curing coupling effect data.

[0008] Preferably, the intelligent quality identification center includes a defect pattern recognition unit and a quality tracing module. The defect pattern recognition unit is used to identify quality defect types such as uneven core density, insufficient bonding strength, lamination misalignment, dimensional deviation, and incomplete curing by using an improved Transformer model combined with multiphysics simulation data and measured parameters. The quality tracing module is used to locate the physical location, influence range, and inducing factors of the quality anomaly source based on the temporal-spatial mapping relationship between defect features and twin forming process.

[0009] Preferably, the integrated control unit includes a component life prediction module and a molding-quality coordinated adjustment unit. The component life prediction module is used to predict the remaining service life of key components such as the mold cavity, heating element, and pressure actuator based on Weibull distribution and molding process performance degradation data. The molding-quality coordinated adjustment unit is used to dynamically adjust the parameters of hot pressing temperature gradient, pressure loading rate, holding time, and stack alignment correction amount according to quality identification results, life prediction data, and real-time molding accuracy, and simultaneously generate quality defect remediation strategies.

[0010] Preferably, the visualization interaction module includes a linked scene rendering engine, a quality-control correlation display unit, and a multimodal operation module. The linked scene rendering engine is used to render the hot pressing process, the core temperature / stress distribution cloud map, the adhesive curing progress, and quality defect visualization markers in real time. The quality-control correlation display unit is used to simultaneously display the quality identification results, defect cause analysis, and control parameter adjustment trajectory. The multimodal operation module is used to provide touch control, voice commands, and virtual sectioning functions, enabling multi-angle observation of the 3D scene, backtracking of the forming process, and viewing of component status details.

[0011] The second aspect of this application provides an integrated method for hot pressing molding and quality identification of self-bonded motor cores, comprising: acquiring real-time hot pressing parameters, adhesive state data, lamination characteristic parameters, and environmental variables during the hot pressing molding process of self-bonded motor cores; filtering and normalizing the hot pressing parameters, adhesive state data, lamination characteristic parameters, and environmental variables to generate a molding-quality correlation dataset; constructing a full-element digital twin based on the molding-quality correlation dataset; and, according to the molding-quality correlation dataset, integrating a finite element-curing coupling algorithm to simulate the thermo-mechanical-curing coupling effect in real time, and comparing the deviation between the measured data and the virtual simulation results to determine the quality identification. An improved Transformer model is used to assess the molding quality status, identify potential quality defect patterns, and locate anomaly sources, obtaining quality assessment results and anomaly location data. Based on the quality assessment results, combined with the 3D virtual mapping of the dynamic twin and historical molding data, an immersive 3D molding-quality linkage scene is constructed, rendering the hot pressing process, quality status, and defect distribution in real time. A molding-quality collaborative control strategy is generated in conjunction with the production plan. Integrated control is performed based on the collaborative control strategy, real-time correction of twin physical parameters and curing kinetic model parameters, prediction of key component lifespan using Weibull distribution, and dynamic adjustment of hot pressing molding parameters and quality correction strategies.

[0012] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the integrated method for hot pressing and quality identification of self-adhesive motor cores as described in the above embodiments.

[0013] The fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the integrated method for hot pressing and quality identification of self-adhesive motor cores as described in the above embodiments.

[0014] The fifth aspect of this application provides a computer program product, including a computer program or instructions, for implementing the integrated method of hot pressing molding and quality identification of self-adhesive motor cores as described in the above embodiments.

[0015] Therefore, this application has the following beneficial effects: This application embodiment acquires full-dimensional process data through a hot-pressing parameter sensing module. Relying on the finite element-curing coupling algorithm of the dynamic twin molding modeling module, it accurately simulates the thermo-mechanical-curing coupling effect. Combined with an improved Transformer model's intelligent quality identification center, it efficiently identifies defect patterns and locates anomaly sources. Then, an integrated control unit dynamically optimizes molding parameters and predicts the remaining lifespan of the component. An immersive 3D visualization interactive module enables real-time rendering of the molding process and quality status, and synchronizes the control effects. This not only overcomes the limitations of traditional hot-pressing molding and quality identification being disconnected, significantly improving the core molding accuracy and quality stability, but also effectively reduces production losses and costs and significantly improves production efficiency through enhanced process controllability, improved anomaly response speed, and component lifespan prediction. Simultaneously, it provides operators with an intuitive and convenient interactive experience, facilitating refined and intelligent management of the production process. Therefore, it solves the problems of incomplete hot-pressing process parameter sensing and lagging quality identification methods in existing technologies.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the integrated system for hot pressing and quality identification of self-adhesive motor cores according to an embodiment of this application. Figure 2 This is a schematic diagram of a hot-pressure parameter sensing module according to an embodiment of this application; Figure 3 This is a schematic diagram of a dynamic twin prototyping modeling module provided according to an embodiment of this application; Figure 4 This is a schematic diagram of an intelligent quality identification center provided according to an embodiment of this application; Figure 5 This is a schematic diagram of an integrated control unit provided according to an embodiment of this application; Figure 6 This is a schematic diagram of a visual interactive module provided according to an embodiment of this application; Figure 7 This is a flowchart of the integrated system for hot pressing and quality identification of self-adhesive motor cores according to an embodiment of this application; Figure 8 This is a flowchart of an integrated method for hot pressing and quality identification of self-adhesive motor cores according to an embodiment of this application; Figure 9This is a schematic diagram of an integrated method for hot pressing and quality identification of self-adhesive motor cores according to an embodiment of this application; Figure 10 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] The following describes an integrated system and method for hot pressing molding and quality identification of self-bonded motor cores according to embodiments of this application, with reference to the accompanying drawings. Addressing the issue of incomplete parameter perception during the hot pressing process mentioned in the background section, this application provides an integrated system for hot pressing molding and quality identification of self-bonded motor cores. In this system, a hot pressing parameter perception module collects full-dimensional process data. The finite element-curing coupling algorithm of the dynamic twin molding modeling module accurately simulates the thermo-mechanical-curing coupling effect. Combined with an improved Transformer model's intelligent quality identification center, defect patterns are efficiently identified and anomaly sources are located. An integrated control unit dynamically optimizes molding parameters and predicts the remaining lifespan of the component. An immersive 3D visualization interactive module enables real-time rendering of the molding process and quality status, and synchronizes the display of control effects. This not only overcomes the limitations of traditional hot pressing molding and quality identification being disconnected, significantly improving core molding accuracy and quality stability, but also effectively reduces production losses and costs and significantly improves production efficiency through enhanced process controllability, improved anomaly response speed, and component lifespan prediction. Simultaneously, it provides operators with an intuitive and convenient interactive experience, facilitating refined and intelligent management of the production process. This solves the problems of incomplete perception of hot pressing process parameters and outdated quality identification methods in existing technologies.

[0020] Figure 1 This is a schematic diagram of the structure of the self-adhesive motor core hot pressing molding-quality identification integrated system provided in the embodiments of this application.

[0021] This application embodiment provides an integrated system for hot pressing and quality identification of self-adhesive motor cores, the system 10 including: Hot pressing parameter sensing module 100, dynamic twin modeling module 200, intelligent quality identification center 300, integrated control unit 400, and visual interaction module 500.

[0022] The system includes the following components: a hot-pressing parameter sensing module 100, which collects real-time hot-pressing parameters, adhesive status data, stacking characteristic parameters, and environmental variables during the hot-pressing process of the self-bonded motor core; a dynamic twin molding modeling module 200, which constructs a digital twin of all elements of the self-bonded motor core hot-pressing process, and simulates the thermo-mechanical-curing coupling effect based on the physical properties of hot pressing and the curing dynamics parameters of the adhesive, using a finite element-curing coupling algorithm; an intelligent quality identification center 300, which identifies the core molding quality defect patterns and locates the source of quality anomalies by combining multi-physics simulation data with an improved Transformer model; an integrated control unit 400, which predicts the remaining life of key molding components based on the quality identification results and real-time monitoring data, and dynamically adjusts the hot-pressing parameters and quality correction strategies; and a visualization interaction module 500, which constructs an immersive three-dimensional molding-quality linkage scene, renders the hot-pressing process and quality status in real time, and synchronously displays the execution effect of control commands.

[0023] It is understood that in this embodiment, the hot-pressing parameter sensing module collects full-dimensional process data, and the finite element-curing coupling algorithm of the dynamic twin forming modeling module accurately simulates the thermo-mechanical-curing coupling effect. Combined with the improved Transformer model's intelligent quality identification center, it efficiently identifies defect patterns and locates anomaly sources. Then, the integrated control unit dynamically optimizes forming parameters and predicts the remaining lifespan of the component. An immersive 3D visualization interactive module enables real-time rendering of the forming process and quality status, and synchronizes the display of control effects. This not only overcomes the limitations of traditional hot-pressing forming and quality identification being disconnected, significantly improving the core forming accuracy and quality stability, but also effectively reduces production losses and costs and significantly improves production efficiency through enhanced process controllability, improved anomaly response speed, and component lifespan prediction. Simultaneously, it provides operators with an intuitive and convenient interactive experience, facilitating refined and intelligent management of the production process. Therefore, it solves the problems of incomplete hot-pressing process parameter sensing and lagging quality identification methods in existing technologies.

[0024] In this embodiment of the application, the hot-pressing parameter sensing module 100 includes: as follows Figure 2 As shown, it includes an infrared temperature measurement array, a high-precision pressure sensor group, an ultrasonic adhesive curing detector, a stack alignment sensor, and an environmental monitoring unit.

[0025] Among them, the infrared temperature measurement array is used to capture the temperature field of the mold and the temperature gradient between the iron core layers in real time; the high-precision pressure sensor group is used to collect the dynamic pressure curve and pressure holding stability data during the hot pressing process; the ultrasonic adhesive curing detector is used to obtain the curing degree of the adhesive and the interface bonding state; the lamination alignment sensor is used to monitor the radial offset of the iron core laminations and the amount of interlayer misalignment; and the environmental monitoring unit is used to collect data on the molding environment temperature, humidity and dust concentration.

[0026] It is understood that the embodiments of this application capture the temperature field of the entire mold and the temperature gradient between the iron core layers in real time through an infrared temperature measurement array, collect dynamic pressure and pressure holding stability data of hot pressing through a high-precision pressure sensor group, obtain the curing degree of the adhesive and the interface bonding state through an ultrasonic adhesive curing detector, monitor the radial offset and interlayer misalignment of the iron core laminations through a stack alignment sensor, and collect ambient temperature, humidity and dust concentration data through an environmental monitoring unit. It can accurately collect key parameters of the entire hot pressing process from multiple dimensions, control the molding state in real time, identify various molding hazards in a timely manner, avoid iron core quality defects, ensure the molding accuracy and bonding quality of the iron core, and provide accurate data support for the quality identification process, effectively improving the accuracy of system process control and quality identification.

[0027] For example, this hot-pressing parameter sensing module includes an infrared temperature measurement array, a high-precision pressure sensor group, an ultrasonic adhesive curing detector, a stack alignment sensor, and an environmental monitoring unit. The infrared temperature measurement array captures the temperature field of the mold (130~220℃) and the temperature gradient between the core layers (≤3℃) in real time with a temperature measurement accuracy of ±1℃. The high-precision pressure sensor group acquires dynamic pressure curves of 0.2~4N / mm² and stability data of holding pressure fluctuations (≤±0.3MPa) during the hot-pressing process with an acquisition accuracy of 0.1MPa. The ultrasonic adhesive curing detector accurately detects the adhesive's 0~100% curing degree (accuracy ±2%) and interface bonding state, ensuring the bonding strength between the core layers is ≥3N / m². The lamination alignment sensor can monitor radial offset of ≤0.15mm and interlayer misalignment of ≤0.05mm for the laminations of the iron core. The environmental monitoring unit simultaneously collects data on the forming environment temperature of 20~35℃, humidity of 30~60%RH, and dust concentration of ≤0.5mg / m³. The modules work together to achieve accurate quantitative acquisition and real-time monitoring of key process parameters throughout the hot pressing process. It can promptly identify problems such as uneven temperature, pressure fluctuation, insufficient bonding and curing, lamination misalignment, and environmental interference, thereby avoiding iron core quality defects from the source, ensuring the forming accuracy and structural stability of the iron core, and providing detailed measured data support for the quality identification process, effectively improving the accuracy of the system's process control and quality identification.

[0028] In this embodiment of the application, the dynamic twin prototyping modeling module 200 includes: as follows Figure 3 As shown, the molded twin building block and the thermo-mechanical-curing coupled simulation unit are shown.

[0029] Among them, the molding twin construction unit is used to construct a full-element digital twin containing the iron core, mold, and hot pressing mechanism based on the iron core design drawings, lamination parameters, and mold 3D scanning data; the thermo-mechanical-curing coupling simulation unit is used to simulate the temperature conduction, stress distribution, and adhesive curing reaction process of the iron core under different hot pressing conditions based on the physical properties of hot pressing and the curing dynamics parameters of the adhesive, integrating finite element analysis and adhesive curing dynamics model, and using the finite element-curing coupling algorithm, and outputting thermo-mechanical-curing coupling effect data.

[0030] It is understood that the molding twin construction unit in this application relies on the core design drawings, lamination parameters, and mold 3D scanning data to construct a full-element digital twin of the core, mold, and hot pressing mechanism, accurately restoring the entire molding structure and actual working conditions, laying the foundation for high-precision simulation; the thermo-mechanical-curing coupling simulation unit combines the physical properties of hot pressing with the curing dynamics parameters of the binder, and simulates the temperature conduction, stress distribution, and binder curing process of the core under different working conditions through a coupling algorithm, and outputs coupling effect data, which can predict the molding quality, optimize process parameters, avoid trial molding and error costs, and ensure the stability of molding and curing.

[0031] For example, in the hot-pressing production of non-oriented silicon steel self-bonding motor cores for new energy vehicles, this dynamic twin molding modeling module can play a significant role. Its molding twin construction unit, based on the core's design drawings, the layer thickness parameters of the 0.35mm silicon steel laminations, and precise 3D scanning data of the hot-pressing mold, builds a full-element digital twin including the core blank, hot-pressing mold, and hot-pressing actuator. This accurately replicates the physical structure of the core laminations and the mold fitting gap, etc., of the actual molding process, laying a solid foundation for high-precision simulation analysis. The thermo-mechanical-curing coupled simulation unit imports the hot-pressing physical properties of the core and the curing kinetic parameters of the epoxy self-bonding agent, integrating finite element analysis and the adhesive curing kinetic model. Through the finite element-curing coupled algorithm, the model... Under multiple working conditions with different hot-pressing temperatures ranging from 180℃ to 220℃ and holding pressures from 2 to 5MPa, the uniformity of temperature conduction inside the iron core, the stress concentration areas of the laminates, and the reaction process of the adhesive from semi-curing to complete curing are analyzed. The corresponding thermo-mechanical-curing coupling effect data are output simultaneously. Based on this, quality problems such as uneven curing of the adhesive at the edge of the iron core and local stress warping of the laminates that are prone to occur during short-term hot pressing at 220℃ can be quickly identified. Without the need for multiple physical mold tests, the optimal hot pressing process parameters of 200℃ and 3MPa holding pressure for 2 hours can be directly optimized and determined, effectively avoiding iron core forming defects, significantly shortening the process debugging cycle, and ensuring that the bonding strength, lamination coefficient, and dimensional accuracy of the iron core all meet the standards, thereby effectively improving the yield and production efficiency of self-bonded motor iron core products.

[0032] In this embodiment of the application, the intelligent quality identification center 300 includes: as follows Figure 4As shown, the defect pattern recognition unit and the quality traceability module are shown.

[0033] Among them, the defect pattern recognition unit is used to identify quality defect types such as uneven core density, insufficient bonding strength, lamination misalignment, dimensional deviation, and incomplete curing by using an improved Transformer model combined with multiphysics simulation data and measured parameters; the quality traceability module is used to locate the physical location, influence range, and inducing factors of quality anomaly sources based on the temporal-spatial mapping relationship between defect features and twin forming process.

[0034] It is understood that the defect pattern recognition unit in this application, through an improved Transformer model combined with multiphysics simulation data and measured parameters, can accurately identify various quality defects such as uneven core density, insufficient bonding strength, lamination misalignment, dimensional deviations, and incomplete curing, ensuring the accuracy and comprehensiveness of defect identification. The quality traceability module, based on the temporal-spatial mapping relationship between defect features and the twin forming process, can accurately locate the physical position of the quality anomaly source, clarify the scope of influence, and trace the inducing factors, achieving precise traceability of quality anomalies. The synergistic effect of these two modules can quickly pinpoint the root cause of quality problems and guide rectification, effectively improving the core forming yield rate, reducing rework costs, and providing data support for optimizing the hot pressing forming process, ensuring core forming quality and production stability.

[0035] For example, in the actual production application of hot-pressing self-bonding motor cores, when quality abnormalities occur during core molding, the defect pattern recognition unit of this intelligent quality judgment center, through an improved Transformer model, integrates multi-physics field simulation data such as hot-pressing temperature field and pressure field with measured core parameters. This allows for accurate identification of issues such as density unevenness (core density below 7.60 g / cm³), insufficient bonding strength (bonding strength below 5.5 N / mm²), lamination misalignment exceeding 0.15 mm, dimensional inaccuracies (outer diameter exceeding ±0.03 mm), and curing degree below [specific value missing]. 95% of the defects are incomplete curing or other quality defects. The quality traceability module then quickly locates the source of the quality anomaly to the mold cavity of zone 3 in the hot pressing molding station based on the identified defect characteristics and the temporal-spatial mapping relationship of the digital twin molding process. It clarifies that the defect affects a 5cm area around the iron core and accurately traces the core-causing factors such as a 2MPa lower hot pressing holding pressure, a local heating temperature deviation of ±5℃, and a stacking positioning gap exceeding 0.05mm. This achieves an efficient closed loop from accurate defect identification to root cause location, and can immediately guide the correction of process parameters to quickly solve molding quality problems.

[0036] In this embodiment, the integrated control unit 400 includes, for example: Figure 5 As shown, the component life prediction module and the molding-quality coordinated adjustment unit are shown.

[0037] The component life prediction module is used to predict the remaining service life of key components such as mold cavity, heating element, and pressure actuator based on Weibull distribution and molding process performance degradation data; the molding-quality coordinated adjustment unit is used to dynamically adjust the parameters of hot pressing temperature gradient, pressure loading rate, holding time, and stack alignment correction amount according to quality identification results, life prediction data and real-time molding accuracy, and simultaneously generate quality defect remediation strategies.

[0038] It is understood that the component life prediction module in this application embodiment is based on Weibull distribution and molding process performance degradation data to accurately predict the remaining service life of key components such as mold cavity, heating element, and pressure actuator. It can predict the wear status of components in advance, avoid molding failures and quality fluctuations caused by component failure, ensure stable equipment operation, and reduce the risk of unplanned downtime. The molding-quality collaborative adjustment unit combines quality identification results, life prediction data, and real-time molding accuracy to dynamically adjust various process and correction parameters, and simultaneously generate quality defect remediation strategies. It can accurately correct molding deviations, resolve quality problems in a timely manner, realize the linkage optimization of process and equipment, and ensure stable and controllable core molding quality.

[0039] In this embodiment of the application, the visual interaction module 500 includes, as follows: Figure 6 As shown, it includes a linked scene rendering engine, a quality-control related display unit, and a multimodal operation module.

[0040] Among them, the linkage scene rendering engine is used to render the hot pressing process, the core temperature / stress distribution cloud map, the adhesive curing progress and quality defect visualization mark in real time; the quality-control association display unit is used to synchronously display the quality identification results, defect cause analysis and control parameter adjustment trajectory; the multi-modal operation module is used to provide touch control, voice command and virtual sectioning functions, to observe the three-dimensional scene from multiple angles, trace back the forming process and view the details of the component status.

[0041] It is understood that the linked scene rendering engine in this application embodiment can render the hot pressing process, core temperature stress cloud map, adhesive curing progress, and quality defect markers in real time, intuitively presenting the entire molding state; the quality-control correlation display unit synchronously displays the quality identification results, defect causes, and parameter adjustment trajectory, realizing the linkage and traceability of quality and control; the multi-modal operation module supports multi-angle observation of the three-dimensional scene, molding retrospection, and detailed viewing through touch, voice, and virtual sectioning functions. The three work together to improve the efficiency of human-computer interaction and visual control, making it easier for staff to intuitively grasp the molding status, accurately control quality problems, and efficiently adjust process parameters, ensuring the intuitiveness and accuracy of quality control.

[0042] For example, in the actual production management of hot pressing molding of self-adhesive motor cores, the linked scene rendering engine of this visual interactive module can render the entire hot pressing molding process at 180~220℃ and 2~4N / mm² in real time, simultaneously generating temperature distribution cloud maps of ±4℃ inside the core, stress distribution cloud maps of 0~150MPa, and 93% adhesive curing progress. It also provides visual and precise marking of quality defects such as 0.12mm lamination misalignment at the core edge and density unevenness with a local stacking coefficient of 96.2%. The quality-control correlation display unit can simultaneously display the identification results of the aforementioned quality defects, the formation rate, and the final product. The system analyzes and adjusts the parameters, including the hot pressing pressure from 2.8 N / mm² to 3.2 N / mm² and the molding temperature from 200℃ to 205℃, to achieve real-time linkage between quality and process control. The multi-modal operation module allows for multi-angle observation of the 3D core molding scene from 0° to 360° via touch control, voice command retrieval, and virtual sectioning. It also allows for tracing back the hot pressing molding sequence of the previous 20 cores and clearly viewing the detailed state of the 0.04mm gap between the core layers. This enables a comprehensive understanding of the working conditions and quality details of each stage of core molding without disassembling the physical object.

[0043] The self-bonding motor core hot-pressing forming-quality identification integrated system proposed in this application collects full-dimensional process data through a hot-pressing parameter sensing module. It accurately simulates the thermo-mechanical-curing coupling effect using a finite element-curing coupling algorithm based on a dynamic twin forming modeling module. Combined with an improved Transformer model's intelligent quality identification center, it efficiently identifies defect patterns and locates anomaly sources. An integrated control unit dynamically optimizes forming parameters and predicts the remaining lifespan of the component. An immersive 3D visualization interactive module enables real-time rendering of the forming process and quality status, and synchronizes the control effects. This system not only overcomes the limitations of traditional hot-pressing forming and quality identification being disconnected, significantly improving core forming accuracy and quality stability, but also effectively reduces production losses and costs and significantly improves production efficiency through enhanced process controllability, improved anomaly response speed, and component lifespan prediction. Simultaneously, it provides operators with an intuitive and convenient interactive experience, facilitating refined and intelligent management of the production process. Therefore, it solves the problems of incomplete hot-pressing process parameter sensing and lagging quality identification methods in existing technologies.

[0044] The following will illustrate the integrated system of hot pressing molding and quality identification for self-adhesive motor cores through a specific embodiment, such as... Figure 7 As shown, it includes: In the mass production workshop of a core component manufacturer for new energy vehicles, an automated production line with an annual output of 1.2 million high-performance iron cores is operating stably. Its core support system is an integrated system for hot-pressing and forming self-bonded motor iron cores and quality assessment. This production line primarily manufactures the rotor iron cores for the drive motors of a high-end new energy vehicle model. The iron cores are made of 0.35mm thick self-bonded silicon steel sheets, designed with 280 layers. The finished product has an outer diameter of 180mm and an inner diameter of 80mm, requiring a bonding strength ≥25MPa, iron loss ≤2.8W / kg, and no defects such as excess glue, weak bonding, or misaligned laminations. Previously, due to fluctuations in hot-pressing parameters and delays in quality inspection, this production line experienced batches of iron cores with insufficient bonding strength and localized excess glue, resulting in a pass rate of only 89%. Furthermore, defect tracing was difficult, severely impacting production and delivery schedules. Since the introduction of the integrated system, precise control and quality closed-loop management of the entire hot pressing process have been achieved through the coordinated operation of five modules. The production line pass rate has increased to 99.2%, production efficiency has increased by 15%, and defect traceability time has been shortened from 2 hours to 5 minutes. The specific application process is as follows: The hot pressing parameter sensing module starts working first, building a comprehensive parameter sensing network throughout the entire hot pressing cycle of the iron core, providing accurate data support for subsequent modules.

[0045] This module features 16 sets of high-precision platinum resistance temperature sensors evenly distributed around the upper and lower molds and circumference of the hot-pressing die. The sampling frequency is set to 100Hz, allowing real-time acquisition of temperature data from various areas of the mold during the hot-pressing process, with a measurement accuracy of ±0.5℃, effectively capturing subtle fluctuations in the mold's temperature field. A high-frequency dynamic pressure sensor is integrated at the pressure head to monitor inter-plate pressure changes in real time, covering a range of 0-10 N / mm² with an accuracy of 0.01 N / mm², accurately recording pressure curves at each stage of pressurization, holding, and depressurization. An online infrared spectroscopy sensor collects binder state data in real time, monitoring the bonding process. The characteristic spectral changes of the agent during the heating process are used to inversely determine the trend of curing degree changes, with a sampling interval of 500ms. At the same time, a laser thickness gauge and a vision inspection device are set up at the stacking feeding station to automatically collect stacking characteristic parameters such as thickness deviation, surface flatness, and coating uniformity of each batch of silicon steel sheets, with a thickness measurement accuracy of 0.001mm. In addition, the system also collects environmental variables such as workshop ambient temperature, humidity, and compressed air pressure. The ambient temperature collection range is 0-50℃, and the humidity collection range is 20%-95%RH, ensuring that the impact of environmental factors on molding quality is taken into account. During the production of a certain batch of iron cores, the module collected the following initial parameters in real time: preset mold heating temperature 200℃, target inter-sheet pressure 3N / mm², holding time 8min, average silicon steel sheet thickness 0.3502mm, surface coating thickness uniformity deviation ≤5%, workshop ambient temperature 25℃, humidity 45%RH, and compressed air pressure 0.6MPa. These parameters were transmitted to the system backend in real time via industrial Ethernet with a transmission delay ≤10ms, providing a precise data foundation for subsequent modeling and identification. Based on the above-mentioned sensing data, the dynamic twin forming modeling module quickly constructed a full-element digital twin of the self-bonding motor iron core hot pressing forming, realizing real-time mapping between the physical process and virtual simulation.

[0046] This module first imports the 3D geometric model of the iron core, and combines it with the collected characteristic parameters of the silicon steel sheets and the curing kinetic data of the adhesive (curing rate curves of this type of adhesive at different temperatures obtained through previous experiments). A full-scale digital twin model, including the mold, the stacked silicon steel sheets, the adhesive coating, and the surrounding heat transfer medium, is constructed in the ANSYS simulation platform. Adaptive meshing technology is used for model mesh generation, with mesh refinement applied to key areas such as the silicon steel sheet stack interface and mold corners to ensure simulation accuracy. Subsequently, the module calls the finite element-curing coupling algorithm, using real-time temperature and pressure data transmitted from the thermo-pressure parameter sensing module as boundary conditions to dynamically simulate the thermo-mechanical-curing coupling effect. The simulation focuses on the temperature field distribution, stress-strain evolution, and adhesive curing degree changes in different regions. During the simulation, the model updates data every 100ms, outputting real-time curing degree distribution cloud maps, temperature gradient curves, and stress distribution data at different locations of the iron core at various times. The curing degree simulation error is ≤3%, and the temperature field simulation error is ≤2℃, demonstrating extremely high consistency with the physical process. At the 3-minute mark of hot pressing of this batch of iron cores, the digital twin model showed that the temperature of the edge area near the mold inlet was 4.2°C lower than other areas, with a curing degree of only 42%, while the curing degree of other areas had reached 58% at the same time. The model simultaneously predicted that if the current parameters were maintained, the final curing degree of this area would be lower than the acceptable threshold of 85%, which may lead to poor adhesion. This simulation warning was pushed to the intelligent quality identification center in real time through the data interface, providing key support for the early prediction of quality problems.

[0047] After receiving simulation data from the dynamic twin modeling module and real-time monitoring data from the hot-pressing parameter sensing module, the intelligent quality assessment center initiates a multi-dimensional data fusion analysis and intelligent defect identification process. The center's built-in improved Transformer model (based on PPLA-Transformer optimization, introducing a pyramid pooling linear attention mechanism to improve the extraction efficiency and recognition accuracy of local defect features) first preprocesses the input data, converting time-series data such as temperature, pressure, and curing degree into feature tensors. Simultaneously, it extracts features from infrared spectral data, constructing an input matrix containing 128-dimensional feature vectors. The model then compares and analyzes a pre-trained defect sample library (containing over 100,000 samples of six common defects, including glue overflow, poor adhesion, misalignment of laminates, and local cracking) to quickly identify quality defect patterns. Regarding the edge region curing lag issue warned by the aforementioned digital twin model, the judgment center, combining real-time temperature sensor data for the area (consistently below the set value of 3.8℃), adhesive infrared spectral characteristics (no characteristic peak indicating completion of curing), and stacking characteristic parameters (the thickness deviation of the silicon steel sheet coating in this area is within the allowable range), accurately identified the quality anomaly mode as "adhesive curing lag caused by insufficient local temperature" after 0.8 seconds of calculation. The anomaly source was located as power attenuation of the heating element on the mold inlet side. Simultaneously, model reasoning indicated that without timely intervention, 35% of the iron cores in this batch would exhibit poor bonding defects, with the defects concentrated in the 10-20mm edge area on the inlet side. Furthermore, the judgment center performed correlation analysis on historical data, discovering that the heating element had been continuously operating for 1200 hours, far exceeding the average stable operating time of similar components, posing a potential failure risk. The relevant judgment results and risk assessment report were pushed to the integrated control unit in real time.

[0048] After receiving the output from the intelligent quality assessment center, the integrated control unit immediately initiates a dynamic control process to achieve precise correction of quality defects and ensure stable production. First, the unit's built-in remaining life prediction model (based on an LSTM neural network, combined with historical data such as the heating element's operating time, temperature fluctuation amplitude, and voltage stability) predicts the remaining life of the faulty heating element. The prediction shows that the remaining life is only 28 hours, far below the 72 hours of stable operating time required by the production plan. Therefore, a critical component maintenance warning is triggered first, reminding maintenance personnel to replace the heating element after the current batch of production is completed. Meanwhile, for the iron cores currently in production, the unit, combining real-time simulation data from the dynamic twin molding modeling module, formulated a phased hot-pressing parameter adjustment strategy: In the first phase, the power supply voltage of the heating pipe on the mold inlet side was increased from 220V to 230V, while the target temperature in that area was slightly adjusted from 200℃ to 205℃ to quickly compensate for insufficient temperature; in the second phase, after reaching 205℃, this temperature was maintained for 1.5 minutes to ensure a steady increase in the adhesive's curing degree; in the third phase, the inter-sheet pressure was slightly adjusted from 3N / mm² to 3.2N / mm², and the holding time was extended by 1 minute to enhance bonding strength; in the fourth phase, during the cooling phase, the pressure was adjusted back to 0.3N / mm² to ensure that the iron core is not prone to internal stress after molding. Control commands are transmitted in real-time to the temperature control system, hydraulic system, and other actuators on the production line via the PROFINET bus. The response time of the actuators is ≤50ms, ensuring timely and accurate parameter adjustments. Meanwhile, the unit also tracks the effects of the adjustments in real time. Data from the hot-pressing parameter sensing module shows that after 1 minute of adjustment, the temperature in the edge area of ​​the iron core rose to 202.3℃, and after 3 minutes, the curing degree reached 65%, narrowing the curing progress difference with other areas to within 5%, meeting the qualification standard. In addition, the control unit also stores the parameter adjustment plan and the quality improvement effect in association, forming a closed-loop process knowledge base to provide parameter references for the subsequent production of iron cores of similar materials and specifications.

[0049] The visualization and interaction module creates an immersive 3D molding-quality linkage monitoring scenario for on-site operators and managers, achieving transparent management of the production process. Based on the Unity3D engine, this module constructs a full 3D virtual scene of the production line, accurately mapping the simulation data from the dynamic twin molding modeling module to the real-time status of the physical production line. Operators can intuitively view the entire process of hot pressing of the iron core through the 3D visualization terminal in the central control room, including the dynamic distribution cloud map of the mold temperature field, the real-time status of silicon steel sheet stacking, and the evolution of the binder curing degree. All data is rendered in real-time at a 1:1 scale with a rendering frame rate of ≥30fps, ensuring smooth and lag-free visuals. For quality anomalies identified by the intelligent quality judgment center, the module highlights the abnormal areas in red in the 3D scene and displays the defect type, anomaly source location, and risk level through pop-up windows. Simultaneously, it automatically plays the timeline of the anomaly's occurrence, helping operators quickly understand the cause of the defect. During the parameter adjustment phase, the module simultaneously displays the execution progress of the control commands, including the heating element voltage adjustment curve, pressure change trend, and temperature recovery trajectory. Different colored curves distinguish the parameter differences before and after adjustment, making the adjustment effect readily apparent. Furthermore, the module supports remote monitoring and data analysis by management personnel. Through a web interface, they can view key indicators such as production line pass rate, defect type statistics, and remaining lifespan of critical components in real time, generating daily and weekly statistical reports to provide data support for production scheduling and equipment maintenance.

[0050] In summary, the embodiments of this application intuitively demonstrate the core value of the visualization interaction module's "real-time monitoring - precise early warning - remote collaboration." Through an immersive 3D scene, operators can quickly discover subtle defects, avoiding batch scrap losses caused by delayed defect detection in traditional production, and significantly reducing production costs. The defect tracing and handling time is reduced from 2 hours to 5 minutes, highlighting the integrated system's role in improving production efficiency. At the same time, it verifies the effectiveness of the system's closed-loop mechanism of "quality identification - control response - personnel collaboration," further demonstrating the system's practical ability to solve pain points such as difficult quality control and slow defect tracing, and providing feasible intelligent support for the high-precision manufacturing of core components for new energy vehicles.

[0051] Next, referring to the accompanying drawings, a method for integrating hot pressing molding and quality identification of self-adhesive motor cores according to embodiments of this application is described.

[0052] like Figure 8 As shown, the integrated method for hot pressing and quality assessment of self-adhesive motor core includes the following steps: In step S101, real-time hot pressing parameters, adhesive status data, lamination characteristic parameters, and environmental variables are obtained during the hot pressing process of the self-adhesive motor core.

[0053] It is understood that the embodiments of this application obtain real-time hot pressing parameters, adhesive state data, lamination characteristic parameters and environmental variables during the hot pressing process of self-bonding motor cores. This allows for the comprehensive and accurate capture of all the feature information of the entire hot pressing process of the core, providing real and complete original data support for subsequent data processing and the construction of a molding-quality correlation dataset. This ensures the accuracy and reliability of subsequent digital twin construction, simulation of thermo-mechanical-curing coupling effects, molding quality assessment and defect anomaly location, and is the core data foundation for realizing the integration of molding and quality identification.

[0054] In step S102, the hot pressing parameters, adhesive state data, stacking characteristic parameters and environmental variables are filtered and normalized to generate a molding-quality correlation dataset.

[0055] Among them, the molding-quality correlation dataset is a structured dataset that integrates product molding process parameters, molding process status data and finished product quality inspection indicators to characterize the corresponding correlation between molding process conditions and product quality level.

[0056] It is understood that the molding-quality correlation dataset in this application embodiment is integrated after filtering and normalization, which can effectively eliminate noise and dimensionality differences in the original multi-source data, and realize the accurate correlation mapping between all elements of hot pressing molding and the quality of core molding. It provides standardized data support for the construction of full-element digital twins and the simulation of thermo-mechanical-curing coupling effects, and provides reliable data basis for improving the quality assessment, defect identification and anomaly location of Transformer models. At the same time, it supports the construction of three-dimensional linkage scenes, the generation of collaborative control strategies and the dynamic correction of hot pressing parameters, and opens up the data link from process parameters to quality identification and control optimization. It is the core data foundation for realizing the integration of molding-quality identification.

[0057] In step S103, a full-element digital twin is constructed based on the molding-quality correlation dataset. According to the molding-quality correlation dataset, the finite element-curing coupling algorithm is integrated to simulate the thermo-mechanical-curing coupling effect in real time. By comparing the deviation between the measured data and the virtual simulation results, the molding quality status is evaluated in combination with the improved Transformer model, potential quality defect patterns are identified and anomaly sources are located, and quality assessment results and anomaly location data are obtained.

[0058] Among them, the finite element-curing coupling algorithm is a numerical analysis algorithm that couples the finite element numerical discretization method with the physical and chemical process of material curing, and simultaneously solves the interaction between multiple field variables of heat, force and chemical in the curing process, so as to realize the numerical analysis algorithm of co-calculation of curing degree, temperature field, stress field and structural deformation.

[0059] It is understood that the finite element-curing coupling algorithm in this application embodiment can rely on the forming-quality correlation dataset to accurately simulate the thermo-mechanical-curing coupling effect of hot pressing of iron core, restore the coupling law of physical field and binder curing reaction in the whole forming process, quantify the intrinsic correlation between process parameters and iron core forming quality, and provide a scientific simulation analysis basis for improving the Transformer model to carry out forming quality assessment, defect pattern recognition and anomaly source location by comparing the deviation between measured data and simulation results, ensuring the accuracy and reliability of quality identification, and improving the simulation fit and physical mapping accuracy of digital twin.

[0060] For example, the finite element-curing coupling algorithm, relying on real-time hot-pressing parameters from the molding-quality correlation dataset, simultaneously constructs a finite element model of the hot-pressing thermodynamic field of the iron core and a curing kinetic model of the binder within a process range of 130~220℃ hot-pressing temperature, 0.2~4N / mm² holding pressure, and 5℃ / min heating rate. This accurately simulates the spatiotemporal distribution of the internal temperature field and stress-strain field of the iron core, as well as the gradient evolution of the binder's curing degree from its initial value to ≥95% qualified value. It quantifies the coupling effect of the thermo-mechanical physical field and the curing reaction on the iron core stacking process. The simulation results accurately identify potential quality defects and corresponding anomalies, such as local temperatures exceeding 180℃ leading to excessively rapid adhesive curing and local pressure below 1N / mm² causing stress concentration in the iron core. This provides a precise quantitative simulation analysis basis for molding quality assessment and anomaly location, effectively improving the physical mapping accuracy and quality identification accuracy of the digital twin.

[0061] In step S104, based on the quality assessment results, combined with the three-dimensional virtual mapping of the dynamic twin and historical molding data, an immersive three-dimensional molding-quality linkage scene is constructed, and the hot pressing process, quality status and defect distribution are rendered in real time. A molding-quality collaborative control strategy is generated in combination with the production plan.

[0062] Among them, the molding-quality linkage scenario refers to an application scenario in industrial production that establishes a real-time correlation and two-way feedback between the process parameters of the product molding process, the molding process status and the quality inspection results of the molded product. The process control at the molding end directly drives the quality to meet the standards, while the quality results are used to optimize the molding process parameters in reverse.

[0063] It is understood that the molding-quality linkage scenario in this application is constructed based on the quality assessment results, combined with the three-dimensional virtual mapping of the dynamic twin and historical molding data. It can provide immersive visualization rendering of the entire hot pressing molding process, the real-time quality status of the iron core, and the distribution of defect locations. It intuitively presents the linkage relationship between molding process parameters and iron core quality. Combined with the production plan, it can quickly generate a scientific molding-quality collaborative control strategy, providing a visualized and precise decision-making basis for the real-time correction of process parameters and the dynamic adjustment of quality correction strategies. It effectively connects the implementation link between quality identification results and molding process control, improves the efficiency of quality anomaly handling and the accuracy of hot pressing molding process control, and ensures the stability and consistency of iron core molding quality.

[0064] For example, taking the hot pressing molding of self-adhesive motor cores as an example, an immersive 3D molding-quality linkage scene is constructed based on quality assessment results, the 3D virtual mapping of dynamic twins, and historical molding process data. This scene can visualize and render the real-time fluctuations of hot pressing temperature (180~220℃) and holding pressure (1.0~3.5N / mm²), simultaneously presenting core quality data such as the gradient distribution of adhesive curing degree (88%~98%), core stacking density (98.5%~99.5%), and warpage deformation (0.03~0.09mm). It can also accurately highlight and locate defect areas and their specific locations caused by localized hot pressing temperatures exceeding 200℃ and holding pressures below 1.0N / mm², resulting in curing degree ≤90% and warpage deformation ≥0.07mm. This method intuitively quantifies the linkage between process parameter fluctuations and core forming quality. Combined with production scheduling plans, it quickly generates a forming-quality coordinated control strategy. It precisely cools the overheated area to 190±5℃ and pressurizes the low-pressure area to 1.2±0.2N / mm². Simultaneously, it corrects the physical parameters of the twin and the parameters of the curing kinetic model. Combined with Weibull distribution, it predicts the remaining service life of key components of the hot pressing mold to be ≥8500 cycles and dynamically adjusts the hot pressing parameters and quality correction strategy. This achieves visualized closed-loop control of forming process and quality management. Ultimately, the solidification degree of the finished core is stable at ≥95%, the warpage deformation is ≤0.05mm, and the stacking density is ≥99.2%. The first-pass yield of core forming is increased to 99.7%, effectively avoiding the generation of batch quality defects.

[0065] In step S105, integrated control is performed according to the collaborative control strategy, the physical parameters of the twin and the parameters of the curing kinetic model are corrected in real time, the lifespan of key components is predicted by combining the Weibull distribution, and the hot pressing molding parameters and quality correction strategy are dynamically adjusted.

[0066] Among them, Weibull distribution prediction is a life prediction method that uses the probabilistic and statistical properties of the Weibull distribution to fit the distribution parameters using sample life / failure data, and then quantitatively predicts the product life pattern, failure probability and reliability.

[0067] It is understood that the Weibull distribution prediction in this application can accurately fit the life decay law and failure probability of key components based on the process load, operating parameters and component operation loss data of the entire hot pressing process of iron core. It can scientifically predict the remaining service life and failure threshold of core components such as hot pressing mold and pressure head. Combined with the molding quality assessment results and twin parameter correction requirements, it provides a reliable life prediction basis for dynamically adjusting hot pressing process parameters and optimizing quality correction strategies. It effectively avoids problems such as hot pressing parameter drift and iron core molding quality fluctuation caused by the performance decay and failure of key components. It can formulate component maintenance and replacement plans in advance, ensure the continuity, stability and process reliability of integrated control of hot pressing and quality identification, and reduce the occurrence rate of quality defects and production operation and maintenance costs.

[0068] For example, in the integrated process of hot pressing molding and quality assessment of self-adhesive motor cores, the specific application of life prediction based on Weibull distribution is as follows: Historical operating loads, working condition losses, and failure frequency data of core components such as hot pressing molds and pressure heads are collected. Weibull shape parameters β=2.5 and scale parameters η=9500 are fitted, accurately calculating the B10 life of these key components to be 8200 cycles, meaning a failure probability of 10% after 8200 cycles. Simultaneously, the remaining effective service life of currently in-service hot pressing molds is predicted to be ≥8600 cycles. When the number of cycles reaches 8000, the failure probability rises to 9.2%, and when the number of cycles exceeds 9000, the failure probability sharply increases to 63.2%. Based on this life... Based on the predicted data and combined with the real-time correction requirements of the twin's physical parameters and curing kinetics model, the hot pressing parameters are dynamically fine-tuned in advance. The hot pressing temperature of 180~220℃ is corrected to 185±3℃, and the holding pressure of 1.0~3.5N / mm² is corrected to 1.2±0.1N / mm². At the same time, the quality correction strategy is optimized to effectively avoid the drift of hot pressing parameters caused by the performance degradation and failure of key components. This eliminates quality defects such as core curing degree below 95% and warping deformation exceeding 0.05mm, ensuring that the first-pass yield of core molding is stable at over 99.7%. In addition, component maintenance and replacement plans are formulated in advance to achieve coordinated matching of process parameter control, quality management and equipment operation and maintenance.

[0069] The integrated method for hot pressing molding and quality identification of self-adhesive motor cores proposed in this application collects full-dimensional process data through a hot pressing parameter sensing module. It accurately simulates the thermo-mechanical-curing coupling effect using a finite element-curing coupling algorithm based on a dynamic twin molding modeling module. Combined with an improved Transformer model's intelligent quality identification center, it efficiently identifies defect patterns and locates anomaly sources. An integrated control unit dynamically optimizes molding parameters and predicts the remaining lifespan of the component. An immersive 3D visualization interactive module enables real-time rendering of the molding process and quality status, and synchronizes the control effects. This method not only overcomes the limitations of traditional hot pressing molding and quality identification being disconnected, significantly improving the core molding accuracy and quality stability, but also effectively reduces production losses and costs and significantly improves production efficiency through enhanced process controllability, improved anomaly response speed, and component lifespan prediction. Simultaneously, it provides operators with an intuitive and convenient interactive experience, facilitating refined and intelligent management of the production process. Therefore, it solves the problems of incomplete hot pressing process parameter sensing and outdated quality identification methods in existing technologies.

[0070] The following will illustrate the integrated method of hot pressing molding and quality identification of self-adhesive motor cores through a specific embodiment, such as... Figure 9 As shown, it includes: A new energy vehicle component manufacturer, when mass-producing self-bonded stator cores (outer diameter 280mm, inner diameter 180mm, stack thickness 60mm, formed by stacking 50W470 self-bonded silicon steel sheets) for 200kW new energy vehicle drive motors, introduced an integrated method of hot-pressing forming and quality identification for self-bonded motor cores. This effectively solved problems such as reliance on experience-based hot-pressing parameter settings, delayed detection of quality defects, and slow control response in traditional production, achieving a dual improvement in production efficiency and product qualification rate. The specific application process is as follows: During the mass production of this stator core (planned batch output of 5000 pieces, daily capacity of 250 pieces), the hot-pressing... In the hot pressing process, a full-parameter acquisition system is first established to achieve real-time capture of multi-source data during the hot pressing process. Hot pressing parameters are acquired in real-time by high-precision sensors installed on a 400t servo hot pressing machine, including the upper mold temperature (acquisition accuracy ±0.5℃), lower mold temperature (acquisition accuracy ±0.5℃), hot pressing pressure (acquisition range 0-5N / mm², accuracy ±0.01N / mm²), holding time (timing accuracy ±0.1s), and heating rate (acquisition accuracy ±0.2℃ / s). The target range for hot pressing temperature is set at 180-220℃, and the target range for inter-plate pressure is set at 2-4N / mm². The target holding time is 2-3 hours, matching the curing characteristics of the self-adhesive silicon steel sheet adhesive. Adhesive state data is monitored in real-time using an online infrared spectral sensor, focusing on collecting the adhesive viscosity (range 500-5000 mPa·s), degree of cure (range 0-100%), and volatile matter content (accuracy ±0.1%). The degree of cure is calculated based on the intensity change of characteristic peaks in the infrared spectrum. Stacking characteristic parameters are acquired using a front-end vision inspection system and a thickness measuring instrument, including the thickness of a single silicon steel sheet (accuracy ±0.001 mm), the total thickness of the stack (accuracy ±0.01 mm), and the flatness of the stack. The data collection accuracy is ±0.005mm, and the surface roughness of the silicon steel sheet (Ra, collection range 0.1-1.0μm) is recorded. At the same time, the batch number of each batch of silicon steel sheet and the material parameters provided by the manufacturer are recorded. Environmental variables are collected through the workshop environmental monitoring station, including the production workshop temperature (collection range 15-35℃, accuracy ±0.5℃), relative humidity (collection range 30%-70%, accuracy ±2%), and compressed air pressure (collection range 0.6-0.8MPa, accuracy ±0.01MPa). All data are transmitted to the edge computing node at a frequency of 100Hz via industrial Ethernet to achieve synchronous aggregation of multi-source data.

[0071] After data acquisition, the Kalman filter algorithm was first used to filter the time-series data such as hot-pressing temperature and pressure to remove abnormal fluctuations caused by equipment vibration and electromagnetic interference (the filtering threshold was set to 3 times the standard deviation, and the data removal rate was about 1.2%). Then, the min-max normalization method was used to normalize the parameters of different dimensions and units to the [0,1] interval, where 180℃ hot-pressing temperature corresponds to 0 and 220℃ corresponds to 1, 2N / mm² hot-pressing pressure corresponds to 0 and 4N / mm² hot-pressing pressure corresponds to 1, and 0% adhesive curing degree corresponds to 0 and 100% hot-pressing degree corresponds to 1. At the same time, combined with the offline core quality data (including interlayer bonding strength, dimensional accuracy, and surface defects), a molding-quality correlation dataset containing 12 input features and 3 output labels was constructed. The dataset contains 100,000 samples (covering historical data of 50 production batches and real-time data of 2 batches), and is divided into training set and test set in an 8:2 ratio for subsequent digital twin construction and model training.Based on the constructed molding-quality correlation dataset, a full-element digital twin was built using ANSYS TwinBuilder and Unity3D. The physical entity modeling was based on 3D design drawings of core components such as the hot press forming machine, stator core blank, and mold, importing material physical parameters (silicon steel sheet density 7850 kg / m³, specific heat capacity 460 J / (kg·K), binder curing enthalpy change 80 J / g) and geometric parameters. A finite element-curing coupling algorithm was integrated, and a thermo-mechanical-curing coupling simulation model was established using ANSYS Mechanical. The heat conduction equation considered the difference in thermal conductivity between the silicon steel sheet and the binder, while the mechanical equation considered the layers of the laminate. The mechanical properties of the in-situ contact were investigated, and the curing kinetics equation adopted a first-order reaction model (dx / dt=k(1-x), where k is the reaction rate constant and x is the degree of curing). The distribution and evolution of the temperature field, stress field, and adhesive curing degree during hot pressing were simulated in real time, with a simulation step size of 1 second, synchronized with the real-time data acquisition frequency. To improve the accuracy of quality assessment, an improved Transformer model was used to evaluate the molding quality status. This model introduces a semantically adaptive weight adjustment mechanism into the residual connection structure of the traditional Transformer, and uses a lightweight MLP scorer to score the semantic quality of each layer's output features (scoring range 0-1), dynamically adjusting the original output... The weight ratio of input to output of the current layer is determined. For shallow layers (layers 1-4), the original semantic weight α=0.7 is set to prioritize the retention of basic parameter features. For deeper layers (layers 9-12), α=0.3 is set to enhance high-level semantic association features. The model input consists of normalized multi-source time-series parameters, and the output is the core quality grade (qualified, requiring rectification, unqualified) and the probability of potential defects. By comparing the deviation between measured data and virtual simulation results in real time, an anomaly warning is triggered when the deviation exceeds 3%. Combined with the output of the improved Transformer model, potential quality defect patterns are identified. For example, when the measured hot-pressing temperature is more than 10℃ lower than the simulated value, or the adhesive curing degree is less than 70%, it is identified as "unbonded". "Insufficient bonding" defects; when the measured hot pressing pressure is more than 0.5 N / mm² higher than the simulated value and the volatile content of the binder is less than 1.5%, it is identified as an "overflowing glue" defect. At the same time, the abnormal source is located through the parameter traceability function of the twin. For example, the "insufficient bonding" defect is mostly located as insufficient local heating power of the heating plate, and the "overflowing glue" defect is mostly located as excessive thickness of the stacked sheets or calibration deviation of the pressure sensor. In a certain batch of production, the system found through deviation comparison that the measured hot pressing temperature of the No. 3 hot pressing station was consistently lower than the simulated value by 8℃. Combined with the model output, it was identified that 15 iron cores had "insufficient bonding" defects. After tracing the source, it was located that the power of the No. 2 heating tube of the heating plate of the No. 3 station was attenuated, and the abnormal source was marked in time.

[0072] Based on the above quality assessment results, an immersive 3D molding-quality linkage scene was constructed using Unity3D combined with VR technology. Visual interaction was achieved through an HTC Vive Pro VR device. The scene rendered in real-time the dynamic process of hot pressing (including mold closing, temperature rise, pressure application, etc.), the quality status of the core (qualified parts marked in green, parts requiring rectification marked in yellow, and unqualified parts marked in red), and the defect distribution (the specific locations of areas with insufficient interlayer bonding and excess adhesive were displayed through a transparent model). Simultaneously, a production planning management module was integrated, importing the production plan for the current batch of 5000 pieces (daily capacity 250 pieces, delivery time 15 days) and equipment maintenance plan; the system... Based on real-time quality data, historical molding data (pass rate of 92.3% for nearly 50 batches, main defect types and rectification effects), and production plans, a multi-objective optimization algorithm is used to generate a molding-quality collaborative control strategy. For the "insufficient bonding" defect, the control strategy is to increase the hot pressing temperature by 10-15℃, extend the holding time by 30 minutes, and arrange maintenance personnel to replace and calibrate the heating tube. For the "overflowing glue" defect, the control strategy is to reduce the hot pressing pressure by 0.3-0.5 N / mm², adjust the stacking order of the sheets to ensure that the total thickness of the sheets is controlled within ±0.1mm, and optimize the production schedule, prioritizing the rectified workpieces for subsequent processing steps to avoid affecting the delivery time. Integrated control is implemented based on the generated collaborative control strategy. Parameter adjustment commands are sent to the hot press molding machine via the industrial PLC system, real-time correction of parameters such as hot pressing temperature, pressure, and holding time. For example, to address the "insufficient bonding" defect at station 3, the hot pressing temperature is increased from 200℃ to 212℃, and the holding time is extended from 2.5h to 3h. Simultaneously, based on measured data and control effects, the physical parameters of the twin (such as updating the actual power parameters of the heating plate) and curing kinetic model parameters (such as adjusting the reaction rate constant k) are corrected in real time to ensure consistency between the twin and the physical entity. To ensure production continuity, the lifespan of key components of the hot press molding machine is predicted using the Weibull distribution, by collecting data from the past three years on hot press molds and heating tubes. Failure data for key components such as pressure sensors are used to fit Weibull distribution parameters using the maximum likelihood estimation method. For example, the shape parameter β=2.3 and dimensional parameter η=8000 hours for the hot pressing mold are predicted to have a reliable lifespan of 7500 pressing cycles. When the mold reaches 7000 cycles, the system issues a maintenance warning. For the pressure sensor, β=1.8 and η=12000 hours are predicted to have a remaining lifespan of 1500 hours, allowing for advance replacement of spare parts based on the production plan. Based on the lifespan prediction results and real-time quality status, the hot pressing parameters and quality correction strategies are dynamically adjusted. For instance, when the mold is nearing its service life, the hot pressing pressure is appropriately reduced by 0.2 N / mm², while the frequency of dimensional inspections is increased to ensure stable product quality.

[0073] By applying this integrated method of hot pressing molding and quality identification for self-bonding motor cores, the company's stator core production qualification rate increased from 92.3% to 99.1%, with major defects such as "insufficient bonding" and "excess glue" reduced by more than 85%. Production efficiency increased by 27%, the hot pressing molding cycle of a single core was shortened from 2.5 hours to 2.1 hours, and labor costs were reduced by 78% (reducing the personnel required for manual parameter adjustment and offline quality inspection). Unplanned downtime for key components was reduced by 60%, and the service life of hot pressing molds was extended by 10%, effectively reducing production and maintenance costs. At the same time, it ensured the quality stability of core components for new energy vehicle drive motors. This application case has been selected as a typical case of "artificial intelligence + manufacturing" in the local area, providing replicable practical experience for the intelligent production of motor cores.

[0074] In summary, this application, based on an industrial-grade mass production scenario, fully verifies the feasibility and practicality of the integrated method for hot pressing molding and quality identification of self-bonded motor cores. It effectively addresses industry pain points in traditional production, such as reliance on experience for hot pressing parameters, delayed detection of quality defects, and slow control response. Through the practical application of multi-source data acquisition and processing, full-element digital twin simulation, improved Transformer model quality identification, and immersive visual control, it achieves intelligent control of the entire production process, resulting in significant economic benefits such as improved pass rates, increased efficiency, reduced costs, and more precise maintenance of key components. Furthermore, its clear application logic, detailed parameter settings, and replicable control scheme provide a concrete reference for the intelligent production of new energy vehicle motor cores and similar self-bonded laminated parts, playing a typical demonstration role and helping to promote the deep implementation of "artificial intelligence + manufacturing" in the field of precision parts production.

[0075] Figure 10 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 1001, the processor 1002, and the computer program stored on the memory 1001 and capable of running on the processor 1002.

[0076] When the processor 1002 executes the program, it implements the integrated method of hot pressing molding and quality identification of self-adhesive motor core provided in the above embodiments.

[0077] Furthermore, electronic devices also include: Communication interface 1003 is used for communication between memory 1001 and processor 1002.

[0078] The memory 1001 is used to store computer programs that can run on the processor 1002.

[0079] The memory 1001 may include high-speed RAM (Random Access Memory) and may also include non-volatile memory, such as at least one disk storage.

[0080] If the memory 1001, processor 1002, and communication interface 1003 are implemented independently, then the communication interface 1003, memory 1001, and processor 1002 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 10 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0081] Optionally, in a specific implementation, if the memory 1001, processor 1002, and communication interface 1003 are integrated on a single chip, then the memory 1001, processor 1002, and communication interface 1003 can communicate with each other through an internal interface.

[0082] The processor 1002 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.

[0083] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described integrated method for hot pressing and quality identification of self-adhesive motor cores.

[0084] Furthermore, this application also provides a computer program product, including a computer program or instructions, which, when executed, implement the above-described integrated method for hot pressing and quality identification of self-adhesive motor cores.

[0085] In the description of this specification, the references to "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0086] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0087] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0088] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any of the following techniques known in the art, or a combination thereof: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0089] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0090] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. An integrated system for hot pressing and quality assessment of self-adhesive motor cores, characterized in that: include: The system includes a hot-pressing parameter sensing module, a dynamic twin modeling module, an intelligent quality identification center, an integrated control unit, and a visual interaction module; among these components... The hot pressing parameter sensing module is used to collect real-time hot pressing parameters, adhesive status data, lamination characteristic parameters and environmental variables during the hot pressing process of self-bonded motor core; The dynamic twin forming modeling module is used to construct a full-element digital twin of the self-adhesive motor core hot-press forming. Based on the hot-pressing physical properties and the curing dynamic parameters of the adhesive, the finite element-curing coupling algorithm is used to simulate the thermo-mechanical-curing coupling effect. The intelligent quality identification center is used to identify the defect patterns of iron core forming quality and locate the source of quality anomalies by using multi-physics simulation data and an improved Transformer model. The integrated control unit is used to predict the remaining life of key molding components based on the quality identification results and real-time monitoring data, and dynamically adjust the hot pressing molding parameters and quality correction strategies. The visualization and interaction module is used to construct an immersive 3D molding-quality linkage scene, render the hot pressing process and quality status in real time, and synchronously display the execution effect of control commands.

2. The integrated system for hot pressing and quality identification of self-adhesive motor cores according to claim 1, characterized in that, The hot-pressing parameter sensing module includes an infrared temperature measurement array, a high-precision pressure sensor group, an ultrasonic adhesive curing detector, a lamination alignment sensor, and an environmental monitoring unit. The infrared temperature measurement array is used to capture the global temperature field of the mold and the temperature gradient between the iron core layers in real time. The high-precision pressure sensor group is used to collect dynamic pressure curves and holding pressure stability data during the hot-pressing process. The ultrasonic adhesive curing detector is used to obtain the degree of adhesive curing and the interface bonding state. The lamination alignment sensor is used to monitor the radial offset and interlayer misalignment of the iron core laminations. The environmental monitoring unit is used to collect data on the molding environment temperature, humidity, and dust concentration.

3. The integrated system for hot pressing and quality identification of self-adhesive motor cores according to claim 1, characterized in that, The dynamic twin forming modeling module includes a forming twin construction unit and a thermo-mechanical-curing coupling simulation unit. The forming twin construction unit is used to construct a full-element digital twin containing the core, mold, and hot pressing mechanism based on the core design drawings, lamination parameters, and mold 3D scanning data. The thermo-mechanical-curing coupling simulation unit is used to simulate the temperature conduction, stress distribution, and adhesive curing reaction process of the core under different hot pressing conditions by integrating finite element analysis and adhesive curing kinetic models, and using a finite element-curing coupling algorithm, based on the physical properties of hot pressing and the curing kinetic parameters of the adhesive, and outputting thermo-mechanical-curing coupling effect data.

4. The integrated system for hot pressing and quality identification of self-adhesive motor cores according to claim 1, characterized in that, The intelligent quality identification center includes a defect pattern recognition unit and a quality tracing module. The defect pattern recognition unit is used to identify quality defect types such as uneven core density, insufficient bonding strength, lamination misalignment, dimensional deviation, and incomplete curing by using an improved Transformer model combined with multiphysics simulation data and measured parameters. The quality tracing module is used to locate the physical location, influence range, and inducing factors of quality anomaly sources based on the temporal-spatial mapping relationship between defect features and twin forming process.

5. The integrated system for hot pressing and quality identification of self-adhesive motor cores according to claim 1, characterized in that, The integrated control unit includes a component life prediction module and a molding-quality coordinated adjustment unit. The component life prediction module is used to predict the remaining service life of key components such as mold cavity, heating element, and pressure actuator based on Weibull distribution and molding process performance degradation data. The molding-quality coordinated adjustment unit is used to dynamically adjust the parameters of hot pressing temperature gradient, pressure loading rate, holding time, and stack alignment correction amount according to quality identification results, life prediction data, and real-time molding accuracy, and simultaneously generate quality defect remediation strategies.

6. The integrated system for hot pressing and quality identification of self-adhesive motor cores according to claim 1, characterized in that, The visualization interaction module includes a linked scene rendering engine, a quality-control correlation display unit, and a multimodal operation module. The linked scene rendering engine is used to render the hot pressing process, the core temperature / stress distribution cloud map, the adhesive curing progress, and the visual markers of quality defects in real time. The quality-control correlation display unit is used to simultaneously display the quality identification results, defect cause analysis, and control parameter adjustment trajectory. The multimodal operation module is used to provide touch control, voice commands, and virtual sectioning functions, allowing for multi-angle observation of the 3D scene, backtracking of the forming process, and viewing of component status details.

7. An integrated method for hot pressing and quality assessment of self-adhesive motor cores, characterized in that... include: Acquire real-time hot pressing parameters, adhesive status data, lamination characteristic parameters, and environmental variables during the hot pressing process of self-bonding motor core; The hot pressing parameters, adhesive state data, stacking characteristic parameters and environmental variables are filtered and normalized to generate a molding-quality correlation dataset. Based on the molding-quality correlation dataset, a full-element digital twin is constructed. According to the molding-quality correlation dataset, the finite element-curing coupling algorithm is integrated to simulate the thermo-mechanical-curing coupling effect in real time. By comparing the deviation between the measured data and the virtual simulation results, the molding quality status is evaluated by combining the improved Transformer model, potential quality defect patterns are identified and anomaly sources are located, and quality assessment results and anomaly location data are obtained. Based on the quality assessment results, combined with the three-dimensional virtual mapping of the dynamic twin and historical molding data, an immersive three-dimensional molding-quality linkage scene is constructed, which renders the hot pressing process, quality status and defect distribution in real time, and generates molding-quality collaborative control strategies in combination with the production plan. Integrated control is performed based on the aforementioned collaborative control strategy, real-time correction of twin physical parameters and curing kinetic model parameters, prediction of key component lifespan using Weibull distribution, and dynamic adjustment of hot pressing molding parameters and quality correction strategy.

8. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the integrated method for hot pressing and quality identification of self-adhesive motor core as claimed in claim 7.

9. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When a computer program or instruction is executed, it implements the integrated method of hot pressing molding and quality identification of self-adhesive motor core as claimed in claim 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, When a computer program or instruction is executed, it implements the integrated method of hot pressing molding and quality identification of self-adhesive motor core as claimed in claim 7.

Images