MES-based iron core production full-process intelligent digital management system

Abstract

The invention discloses an MES-based iron core production full-process intelligent digital management system, and relates to the technical field of iron core production digital management, and the system comprises a data collection unit, an MES integration unit, a production scheduling unit, a quality control unit, a digital twinning unit, a tracing unit, an operation and maintenance unit, and a power supply guarantee unit. A data interaction link is established by each unit through an industrial Ethernet, a Profinet protocol is adopted by a communication link, a frequency band is adaptive to electromagnetic environment characteristics of an iron core production workshop through a frequency spectrum adaptation algorithm, and production process beat characteristics of a signal transmission time sequence are anchored by a process beat synchronizer; the data acquisition unit adopts a multi-sensor fusion and edge computing cooperation mechanism, high-fidelity data acquisition is realized in combination with a Kalman filtering algorithm, the problems of traditional data fragmentation and large noise interference are solved, accurate data support is provided for management decision, and each unit forms linkage with a hardware signal link through a user-defined data interaction protocol.

Description

Technical Field

[0001] This invention relates to the field of digital management technology for iron core production, and in particular to an intelligent digital management system for the entire iron core production process based on MES. Background Technology

[0002] As a core component of power equipment and new energy equipment, the production process of iron cores involves many closely related processes, such as raw material inspection, stamping, lamination, curing, and finished product testing. Moreover, the process parameters are sensitive, and the requirements for precise control of the production process are extremely high.

[0003] Current iron core production management largely relies on traditional MES systems for task assignment and data recording. However, this management method suffers from numerous technical shortcomings due to a lack of deep integration with the entire production process:

[0004] First, data acquisition is fragmented, often relying on single sensors to collect local parameters without achieving multi-sensor fusion and edge computing synergy. This results in poor data noise reduction and feature extraction, failing to provide high-fidelity data support for management decisions. Second, the linkage between production units is loose. Data interaction between the MES system and production and testing equipment is limited to format adaptation, lacking cross-unit parameter linkage, verification, and iterative correction mechanisms, leading to insufficient adaptability of scheduling instructions to actual working conditions. Third, quality control relies on manually set thresholds, lacking a dual threshold library and dynamic mapping relationship between industry standards and enterprise internal control standards. This results in highly subjective judgments, making it difficult to accurately match the quality requirements of iron core production. Fourth, digital twin models are mostly static geometric mappings, lacking dynamic calibration links with real-time production data. This causes a disconnect between virtual simulation and actual production, making it difficult to support process optimization and fault prediction. Fifth, the traceability system only stores basic flow data, failing to adopt a hybrid architecture of blockchain and cloud storage. This results in easily tampered data and unclear traceability links, making it impossible to achieve accurate traceability throughout the entire process. Summary of the Invention

[0005] To address the technical problems existing in the prior art, this invention provides an intelligent digital management system for the entire iron core production process based on MES. The technical solution is as follows:

[0006] It provides an intelligent digital management system for the entire iron core production process based on MES, including a data acquisition unit, an MES integration unit, a production scheduling unit, a quality control unit, a digital twin unit, a traceability unit, an operation and maintenance unit, and a power supply guarantee unit.

[0007] Each unit establishes a data interaction link through industrial Ethernet. The communication link adopts the Profinet protocol. The frequency band is adapted to the electromagnetic environment characteristics of the iron core production workshop through a spectrum adaptation algorithm. The signal transmission timing is anchored to the production process timing characteristics by the process timing synchronizer.

[0008] The data acquisition unit integrates a multi-sensor fusion module, an edge computing node, and a data preprocessing module. The edge computing node and the multi-sensor fusion module construct a collaborative link for noise reduction and feature extraction through real-time data streams. The preprocessed data is synchronized to the MES integration unit and the digital twin unit.

[0009] The MES integration unit is configured with a heterogeneous data adaptation interface, a real-time interaction engine, and a data verification module. It builds a bidirectional data link with the existing MES system through the interface adaptation protocol. The data interaction format and the MES system interface specification form a one-to-one mapping. The verification module verifies the data integrity through the CRC32 algorithm.

[0010] The production scheduling unit has a built-in task decomposition module, a hybrid optimization algorithm engine, and a working condition feedback module. The hybrid optimization algorithm engine adopts a hybrid strategy that combines genetic algorithms and rule-based reasoning. The working condition feedback module transmits the equipment operating status in real time to dynamically adjust the scheduling parameters.

[0011] The quality control unit is equipped with a multi-dimensional detection module, a data processing chip, and a standard threshold storage module. The detection parameters are mapped to the quantitative indicators of the iron core production industry standards and the company's internal control standards.

[0012] The digital twin unit constructs a high-fidelity virtual model containing workshop layout, equipment structure, and process logic. The model forms a real-time calibration link with the data acquisition unit through a parameter iteration algorithm.

[0013] Each unit forms a linkage architecture through hardware signal links and custom data interaction protocols. The architecture embeds technical mechanisms for cross-unit parameter linkage, verification, and iterative correction.

[0014] Preferably, the raw material characteristic acquisition component of the data acquisition unit is equipped with material analysis equipment, size measurement equipment, and magnetic property detection equipment, and the monitoring dimensions cover the material composition, geometric dimensions, magnetic permeability, and iron loss value of the raw materials for iron core production; the sensing elements of the process parameter acquisition module are respectively deployed on the key equipment of the stamping, lamination, and curing processes. The sensing elements are adapted to the process parameter change law of each process after signal amplification and filtering circuit. The output signal is processed by the Kalman filtering algorithm of the edge computing node and then transmitted to the data preprocessing module.

[0015] The output of the finished product testing and acquisition module establishes a physical connection with the input interface of the quality control unit through a data calibration link. The test data format is compatible with the quality judgment quantification logic through a data format mapping table.

[0016] All acquisition components and modules use the Modbus communication protocol, and the sampling frequency is anchored to the production process rhythm by the process cycle synchronizer.

[0017] Preferably, the interface adapter component of the MES integration unit adopts a dual interface design of RS485 and Ethernet, and adapts to the existing MES system hardware interface type through interface pin definitions;

[0018] The data conversion module has a built-in heterogeneous data parsing engine, format mapping rule base and data verification algorithm, which converts the equipment data, test data and operating condition data collected by each unit into a standardized format compatible with the MES system. The conversion logic follows the data exchange standard of the iron core production industry.

[0019] The interactive collaboration module receives production task instructions, process adjustment parameters, and quality standard update information issued by the MES system. After verification by the data verification algorithm, it transmits the data to the data conversion module and synchronously outputs production progress statistics, quality inspection results, and other data to the MES system.

[0020] Preferably, the task parsing component of the production scheduling unit decomposes the production tasks according to the sequence and dependency of the iron core production process, forming a hierarchical process task list;

[0021] The operating condition assessment module collects data such as production equipment load rate, raw material supply status, process completion progress, and inspection qualification status, and generates operating condition assessment results through a weighted summation algorithm.

[0022] The scheduling instruction generation module has a built-in rule base for priority rules of iron core production processes and an equipment load balancing model. The rule base includes core rules such as process connection intervals, equipment capacity limits, and raw material consumption thresholds. The model optimizes equipment load distribution through linear programming algorithms and outputs instructions for equipment start-up and shutdown, process adjustment, capacity allocation, and raw material scheduling. The instruction parameters are adapted to the operating characteristics and process requirements of the production equipment through an equipment operation parameter mapping table.

[0023] Preferably, the multi-dimensional detection components of the quality control unit include a high-precision magnetic property detection device, a laser dimensional accuracy measurement device, and a machine vision appearance defect recognition device, and the detection parameters cover the magnetic permeability, iron loss value, geometric dimensional tolerance, and surface defect density of the finished iron core.

[0024] The data processing module uses the Kalman filter algorithm for noise reduction and extracts key information through a feature extraction model. The processed data is mapped by a data mapping algorithm and corresponds to the quantization threshold of the standard threshold storage module. The quality judgment module has a built-in dual threshold library of iron core production industry standards and enterprise internal control standards. It generates the judgment results of qualified, rework, and scrap by calculating the difference between the test data and the standard threshold. The judgment results are transmitted to the MES system and traceability unit through the data synchronization protocol.

[0025] Preferably, the full-process model building component of the digital twin unit adopts a multi-software collaborative modeling architecture. Through the parameter association algorithm of the three-dimensional geometric modeling module and the dynamic simulation module, a one-to-one correspondence mapping between the workshop layout, equipment structure, process flow logic and physical production system is realized. The mapping relationship is defined by the parameter association matrix.

[0026] The real-time calibration module interacts in real time with the raw material data, process parameters, and finished product test data of the data acquisition unit. It iteratively corrects the model parameters using the least squares method, and the corrected parameters are written into the virtual model through the model update interface.

[0027] The virtual simulation testing component builds a multi-scenario simulation platform. By modeling physical laws and embedding process constraints into algorithms, it adapts to the physical laws and process requirements of actual production, supporting test scenarios such as production load fluctuations, equipment failure simulation, process parameter adjustment, and changes in raw material characteristics.

[0028] Preferably, the distributed data storage component of the traceability unit adopts a hybrid architecture combining blockchain and cloud storage. The blockchain module uses a hash encryption algorithm to ensure that the data is tamper-proof, and the cloud storage module uses a multi-node backup mechanism to achieve off-site backup. The stored data covers raw material batch information, production equipment number, operator identity information, process parameters of each process, quality inspection data, and finished product delivery information.

[0029] The traceability link construction module establishes associated links according to the actual flow logic of raw material warehousing, process processing, finished product inspection, and outbound delivery. The link association is realized through hash mapping of unique batch identifiers. The association rules are adapted to the actual execution characteristics of the production process by the link node association algorithm. The link nodes correspond to the key data collection points of each production process through data collection point identifiers.

[0030] The multi-dimensional query output module connects to the read interface of the distributed data storage component through a dedicated query interface, and the query results are output to the MES system and operation and maintenance unit in a standardized report format.

[0031] Preferably, the equipment status monitoring component of the operation and maintenance unit is equipped with vibration sensing devices, temperature sensing devices, current sensing devices, and energy consumption monitoring devices to collect data such as vibration frequency, operating temperature, operating current, and real-time energy consumption of the production equipment. The monitoring parameters are adapted to the equipment operation safety characteristics and performance degradation law by the equipment characteristic database.

[0032] The intelligent fault diagnosis module has a built-in equipment fault feature library and machine learning diagnostic model. The fault feature library contains core features of typical equipment faults, such as vibration frequency threshold, temperature change range, and current fluctuation range. The machine learning diagnostic model compares real-time monitoring data with historical fault data through fault feature extraction algorithms, covering diagnostic types such as mechanical faults, electrical faults, and control system faults.

[0033] The maintenance plan generation module generates targeted maintenance plans based on diagnostic results, equipment maintenance cycles, and spare parts inventory status. The plans include information such as maintenance locations, maintenance processes, tool lists, spare parts requirements, and maintenance time limits, and are pushed to the operation and maintenance terminal through a dedicated communication protocol.

[0034] Preferably, the power supply guarantee unit adopts a dual-mode power supply with mains power as the primary source and energy storage as backup. It has a built-in power management chip, voltage stabilization module and load detection module. The load detection module collects production load data in real time and dynamically adjusts the output voltage and power through a power regulation algorithm. Under high production load conditions, the power management chip automatically increases the output power to prioritize the power supply stability of production equipment, data acquisition unit, MES integration unit and digital twin unit. The output voltage fluctuation is controlled within the safety threshold range of the workshop electrical system by the PID regulation algorithm of the voltage stabilization module.

[0035] During low-load or shutdown conditions, the power supply unit reduces the power supply to non-core modules while charging the energy storage module composed of lithium battery packs. The unit has overvoltage, overcurrent, and overheat protection functions. The protection thresholds are adapted to the electrical safety requirements of the workshop electrical system by the electrical safety standard database. After the protection is triggered, the non-core circuit is automatically cut off through the protection circuit and a warning signal is sent.

[0036] Preferably, the system further includes a closed-loop feedback correction unit; the input end of the closed-loop feedback correction unit establishes a data connection with the process parameter output end of the data acquisition unit, the judgment result output end of the quality control unit, and the simulation data output end of the digital twin unit to collect actual production operation data, quality judgment results, and virtual simulation comparison data;

[0037] The deviation value between the collected data and the production target data issued by the MES system is calculated by the deviation quantification algorithm. Based on the deviation value, the scheduling parameters of the production scheduling unit, the judgment threshold of the quality control unit, and the model coefficients of the digital twin unit are corrected in reverse. The correction process is implemented by the PID algorithm. The correction parameters are adapted to the dynamic balance requirements of iron core production efficiency and product quality through parameter mapping rules, and a technical closed loop is formed by relying on the cross-unit data linkage mechanism.

[0038] Beneficial effects

[0039] The benefits brought about by deep multi-unit collaboration and closed-loop technology design are significant:

[0040] Firstly, the data acquisition unit adopts a multi-sensor fusion and edge computing collaborative mechanism, combined with the Kalman filter algorithm to achieve high-fidelity data acquisition, which solves the problems of traditional data fragmentation and large noise interference, providing accurate data support for management decisions. At the same time, each unit forms a linkage with the hardware signal link through a custom data interaction protocol, breaking the limitation of loose linkage between production units and greatly improving the adaptability of production scheduling and quality control.

[0041] Secondly, the quality control unit has a built-in dual threshold library of industry standards and enterprise internal control standards. Through data mapping and difference calculation, it achieves accurate quality judgment, avoids the subjectivity of manually setting thresholds, and significantly improves the product qualification rate.

[0042] Third, the digital twin unit achieves a high degree of consistency between virtual simulation and actual production through parameter iteration algorithms and real-time data dynamic calibration, supports multi-scenario process optimization and fault prediction, and reduces trial and error costs and production risks.

[0043] Fourth, the traceability unit adopts a hybrid architecture of blockchain and cloud storage, ensuring data immutability through hash encryption and multi-node backup, and building a full-process traceability link based on a unique batch identifier to achieve accurate traceability of raw materials, processes, finished products and delivery in all dimensions.

[0044] Fifth, the closed-loop feedback correction unit combines the PID algorithm to dynamically correct scheduling parameters, judgment thresholds and model coefficients. With the intelligent fault diagnosis and precise maintenance solution of the operation and maintenance unit, it effectively reduces equipment downtime, continuously optimizes the dynamic balance between production efficiency and product quality, and improves the overall intelligence, precision and traceability of iron core production management, providing reliable technical support for the large-scale and refined operation of iron core production enterprises. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 A flowchart of the intelligent digital management system for the entire iron core production process based on MES provided in this application embodiment. Detailed Implementation

[0047] The technical solution provided in this application will now be described with reference to the accompanying drawings.

[0048] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0049] First, in this application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character generally indicates an "or" relationship between the preceding and following related objects, but does not exclude the possibility of indicating an "and" relationship. The specific meaning can be understood in context. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, "at least one of a, b, or c" can represent: a, b, c; a and b; a and c; b and c; or a and b and c. Here, a, b, and c can be single or multiple.

[0050] Second, in this application, the use of prefixes such as "first" and "second" is merely for the purpose of distinguishing and describing different things belonging to the same name category, and does not constrain the order, size, or quantity of things. For example, "first message" and "second message" are simply different messages, and there is no temporal sequence, size, or priority relationship between them.

[0051] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0052] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0053] like Figure 1 The diagram shown is a structural schematic of the intelligent digital management system for the entire iron core production process based on MES provided in this application embodiment. The system includes a data acquisition unit, an MES integration unit, a production scheduling unit, a quality control unit, a digital twin unit, a traceability unit, an operation and maintenance unit, a power supply guarantee unit, and a closed-loop feedback correction unit. Each unit establishes a data interaction link through an industrial Ethernet network. The communication link adopts the Profinet protocol, and the frequency band is adapted to the electromagnetic environment characteristics of the iron core production workshop through a spectrum adaptation algorithm. The signal transmission timing is anchored to the production process timing characteristics by a process cycle synchronizer to ensure the real-time performance and reliability of data transmission.

[0054] The data acquisition unit integrates a multi-sensor fusion module, an edge computing node, and a data preprocessing module. The edge computing node and the multi-sensor fusion module construct a collaborative link for noise reduction and feature extraction through real-time data streams. The preprocessed data is synchronized to the MES integration unit and the digital twin unit, providing high-fidelity data input for subsequent analysis and control. The edge computing node uses an NVIDIA Jetson Xavier NX edge computing board, and the data preprocessing module uses an STM32H750VBT6 processor.

[0055] The components are equipped with material analysis equipment, dimensional measurement equipment, and magnetic property testing equipment, and the monitoring dimensions cover the material composition, geometric dimensions, magnetic permeability, and iron loss value of the raw materials used in iron core production. The material analysis equipment used is the Oxford Instruments X, MET8000, which employs X-ray fluorescence spectroscopy to detect elements covering core components such as Fe, Si, and Mn. The detection accuracy is ≤0.01%, and the detection time is ≤3 seconds. It can accurately identify whether the raw material meets the requirements for core-specific materials such as Q345 steel and silicon steel sheets. The dimensional measurement equipment used is the Keyence LJ, V7000, which employs line laser scanning technology with a scanning frequency of 1000Hz, a measurement range of 0 and 500mm, and an accuracy of ±0.01mm. It can collect geometric parameters such as the length, width, and thickness of the raw material. The magnetic property testing equipment used is the Magnet, Physik MPG200, which employs the Epstein square method. The testing frequency is 50Hz and 1kHz, and the magnetic field strength is 0 and 2T. It can measure magnetic properties such as the permeability (range 1000 and 10000 μH / m) and iron loss (range 0.1 and 5W / kg).

[0056] The module's sensing elements are deployed in key equipment of the stamping, lamination, and curing processes. These sensing elements are amplified and filtered by circuits to adapt to the changing process parameters of each process. In the stamping process, HBM1 and C2A / 100N pressure sensors and Micro and EpsilonoptoNCDT1420 displacement sensors are deployed to collect parameters such as stamping pressure (range 0, 100kN), punch displacement (range 0, 50mm), and stamping frequency (range 10, 50 times / minute). In the lamination process, pressure sensors and KeyenceGT2 and H12K thickness sensors are deployed to collect parameters such as lamination pressure (range 0, 50kN), lamination thickness (range 5, 50mm), and lamination alignment. In the curing process, PT100 temperature sensors, SHT30 humidity sensors, and time sensors are deployed to collect parameters such as curing temperature (range 100, 200℃), curing humidity (range 30%, 60%RH), and curing time (range 1, 4 hours). The output signal of the sensing element is processed by the Kalman filter algorithm of the edge computing node. The process noise variance Q=0.01 and the observation noise variance R=0.1, which effectively suppresses the signal noise caused by electromagnetic interference and equipment vibration. The processed signal is then transmitted to the data preprocessing module for normalization.

[0057] The module output establishes a physical connection with the quality control unit input interface via a data calibration link. The data calibration link uses shielded twisted-pair cable, and the detection data format is compatible with the quality judgment quantification logic via a data format mapping table. The module is equipped with a machine vision appearance defect recognition system consisting of a Magnet, a Physik MPG300 high-precision magnetic property testing device, a Keyence IM, a 6500 laser dimensional accuracy measuring device, a Halcon 18.11 + Baslerac A2500 industrial camera, and a 14uc sensor. The magnetic property testing device measures the magnetic permeability and iron loss value of the finished product, while the dimensional measurement device measures the geometric dimensional tolerances of the finished product, such as length, width, thickness, and aperture. The machine vision equipment uses image segmentation and defect feature matching algorithms to identify surface defects such as scratches, dents, and cracks, with a defect recognition accuracy ≤0.1mm. All acquisition components and modules use the Modbus communication protocol. The sampling frequency is anchored to the production process rhythm by the process cycle synchronizer: 50Hz for stamping, 30Hz for lamination, 10Hz for curing, and 20Hz for finished product inspection.

[0058] The MES integration unit is configured with a heterogeneous data adaptation interface, a real-time interaction engine, and a data verification module. It establishes a bidirectional data link with the existing MES system through an interface adaptation protocol, ensuring a one-to-one mapping between data interaction formats and MES system interface specifications. The real-time interaction engine uses a distributed interaction framework developed in Java, and the data verification module verifies data integrity using the CRC32 algorithm.

[0059] The system employs a dual-interface design with RS485 and Ethernet, adapting to existing MES system hardware interface types through interface pin definitions. The RS485 interface pins are defined as GND (pin 1), A+ (pin 2), B (pin 3), and VCC (pin 4), with configurable communication rates of 9600bps and 115200bps. The Ethernet interface uses an RJ45 connector, supports the TCP / IP protocol, and offers adaptive communication rates of 100Mbps / 1Gbps, ensuring hardware compatibility with different types of MES systems.

[0060] The system incorporates a heterogeneous data parsing engine, a format mapping rule base, and a data verification algorithm. It converts equipment data, testing data, and operating condition data collected from each unit into a standardized format compatible with the MES system. This standardized format is XML. The format mapping rule base includes data field mapping tables; for example, it maps stamping pressure to the PROD_PRESSURE field of the MES system and magnetic permeability to the MAG_PERMEABILITY field.

[0061] The system receives production task instructions, process adjustment parameters, and quality standard update information from the MES system. Production task instructions include product model, production batch, and delivery deadline. Process adjustment parameters include stamping pressure threshold and stacking thickness requirements. After verifying data integrity using a data verification algorithm, the received data is transmitted to the data conversion module. Simultaneously, it outputs production progress statistics, quality inspection results, equipment operating status, and raw material consumption data to the MES system. Production progress statistics include completed output, output to be completed, and process pass rate. Quality inspection results include qualified batches, reworked batches, and scrapped batches. Equipment operating status includes runtime, number of failures, and load rate, enabling two-way data interaction and collaboration.

[0062] The production scheduling unit has a built-in task decomposition module, a hybrid optimization algorithm engine, and a working condition feedback module. The hybrid optimization algorithm engine adopts a hybrid strategy that combines genetic algorithms and rule-based reasoning. The working condition feedback module transmits the equipment operating status in real time to dynamically adjust the scheduling parameters. The output end establishes a physical connection with the production equipment control module through the IO interface. The IO interface uses the DI / DO module of Siemens S7 1500 PLC.

[0063] Production tasks are broken down according to the sequence and dependencies of the iron core production process, forming a hierarchical process task list. The first-level task is the total production task of the product, such as producing 1,000 iron cores of model XC and 100; the second-level task is the process task of each workshop, such as the stamping workshop producing 1,000 sets of XC and 100 iron core laminations and the stacking workshop completing the stacking of 1,000 sets of laminations; the third-level task is the task of a single piece of equipment, such as stamping equipment 1 producing 500 sets of XC and 100 iron core laminations and stamping equipment 2 producing 500 sets of XC and 100 iron core laminations.

[0064] Data such as production equipment load rate, raw material supply status, process completion progress, and inspection pass rate are collected. A weighted summation algorithm is used to generate a working condition evaluation result, with evaluation values ​​ranging from 0 to 100 points. The weight allocation is as follows: equipment load rate 30%, raw material supply status 20%, process completion progress 30%, and inspection pass rate 20%. For example, an equipment load rate of 80% scores 24 points, sufficient raw material supply scores 20 points, process completion progress is 60% scored 18 points, and inspection pass rate is 95% scored 19 points, resulting in a working condition evaluation score of 81 points.

[0065] The system incorporates a built-in rule base for priority rules in core production processes and an equipment load balancing model. The rule base includes core rules such as process connection intervals, equipment capacity limits, and raw material consumption thresholds. The process connection interval requirement is that the interval between stamping and laminating be ≤30 minutes. The equipment capacity limit is set at a maximum of 50 sets / hour for a single stamping machine. The raw material consumption threshold stipulates that the raw material consumption per set of XC or 100 cores is ≤2kg. The equipment load balancing model optimizes equipment load allocation using a linear programming algorithm. The objective function is to minimize the equipment load variance, with constraints including equipment capacity limits and process connection requirements. Output instructions include equipment start / stop, process adjustment, capacity allocation, and raw material scheduling. Instruction parameters are adapted to the operating characteristics and process requirements of the production equipment through an equipment operating parameter mapping table. For example, the stamping pressure instruction for stamping equipment 1 is set to 80kN and the stamping frequency to 30 times / minute, while the laminating pressure instruction for the laminating equipment is set to 40kN.

[0066] The quality control unit is equipped with a multi-dimensional detection module, a data processing chip, and a standard threshold storage module. The detection parameters are mapped to quantitative indicators of the iron core production industry standards and the company's internal control standards. The data processing chip is a TITMS320C6748, and the standard threshold storage module uses an SD card for storage.

[0067] The equipment includes high-precision magnetic property testing equipment, laser dimensional accuracy measurement equipment, and machine vision appearance defect recognition equipment. Testing parameters cover the magnetic permeability, iron loss value, geometric dimensional tolerances, and surface defect density of the finished iron core. The acceptable ranges for magnetic permeability are: industry standard 2000, 8000 μH / m; company internal control standard 3000, 7000 μH / m. The acceptable ranges for iron loss value are: industry standard ≤3 W / kg; company internal control standard ≤2.5 W / kg. The acceptable ranges for geometric dimensional tolerance are: industry standard ±0.05 mm; company internal control standard ±0.03 mm. The acceptable ranges for surface defect density are: industry standard ≤1 defect / 100 cm²; company internal control standard ≤0.5 defects / 100 cm².

[0068] Noise reduction is achieved using a Kalman filter algorithm, and key information is extracted through a feature extraction model based on a backpropagation (BP) neural network. The BP neural network's input layer contains detection data, including permeability, iron loss, dimensional parameters, and defect quantity. The hidden layer consists of 10 neurons, and the output layer is a feature vector, such as magnetic performance level, dimensional accuracy level, and appearance quality level. The processed data is then mapped to quantization thresholds in a standard threshold storage module using a data mapping algorithm. This algorithm compares the feature vectors with the standard thresholds to generate quantization comparison results; for example, a permeability of 3500 μH / m corresponds to compliance with internal control standards.

[0069] The system incorporates a dual threshold library of industry standards and internal control standards for iron core production. It generates a pass / fail (qualified, rework, or scrap) determination by calculating the difference between the tested data and the standard thresholds. If all tested parameters meet the internal control standards, the system is deemed qualified; if tested parameters meet industry standards but not the internal control standards, it is deemed reworkable (e.g., a magnetic permeability of 2500 μH / m meets industry standards but is below the lower limit of the internal control standard); if tested parameters do not meet industry standards, it is deemed scrapable (e.g., an iron loss value of 3.5 W / kg exceeds the upper limit of the industry standard). The determination results are transmitted to the MES system and traceability unit via a data synchronization protocol using TCP / IP to ensure data consistency.

[0070] The digital twin unit constructs a high-fidelity virtual model containing workshop layout, equipment structure, and process logic. The model forms a real-time calibration link with the data acquisition unit through a parameter iteration algorithm.

[0071] A multi-software collaborative modeling architecture is adopted. Through parameter association algorithms between the 3D geometric modeling module and the dynamic simulation module, a one-to-one mapping between workshop layout, equipment structure, process flow logic, and physical production system is achieved. The mapping relationship is defined by a parameter association matrix with dimensions of 100×50, which includes the association coefficients of equipment parameters, process parameters, and timing parameters. Geometric modeling recreates the production workshop layout (50m long × 30m wide × 8m high) and the 3D structure of equipment at a 1:1 scale, including stamping equipment, laminating equipment, curing ovens, and testing equipment. Equipment details include external dimensions, interface positions, and moving parts. Dynamic modeling uses Simulink to build equipment dynamic models, such as the punch motion model of the stamping equipment and the pressure control model of the laminating equipment, as well as process flow models, such as the time sequence model of raw material to finished product flow.

[0072] The system interacts in real time with raw material data, process parameters, and finished product inspection data from the data acquisition unit, and iteratively corrects model parameters using the least squares method. Physical system data is acquired every 500 milliseconds, and the deviation between the virtual model output data and the physical data is calculated. For example, if the stamping pressure in the virtual model is 78 kN and the physical data is 80 kN, the deviation is 2 kN. The least squares method is used to iteratively correct model parameters, such as adjusting the friction coefficient and stiffness coefficient in the model, until the deviation is less than a threshold (≤0.5 kN). The corrected parameters are then written into the virtual model via the model update interface to maintain consistency between the model and the physical system state.

[0073] A multi-scenario simulation platform is built, adapting to the physical laws and process requirements of actual production through physical law modeling and process constraint embedding algorithms. It supports scenarios with fluctuating production loads, such as a 50% increase in production batch size or a 90% increase in equipment load rate; equipment failure simulation scenarios, such as pressure sensor failure in stamping equipment or hydraulic system leakage in stacking equipment; process parameter adjustment scenarios, such as a 10°C increase in curing temperature or a 10-times / minute increase in stamping frequency; and raw material characteristic change scenarios, such as a ±10% fluctuation in raw material magnetic permeability. Simulation results output data such as equipment operating status prediction, product quality prediction, and process bottleneck analysis, providing support for production scheduling optimization and process improvement.

[0074] The traceability unit includes a distributed data storage component, a traceability link construction module, and a multi-dimensional query output module, enabling traceable management of data throughout the entire iron core production process.

[0075] The system employs a hybrid architecture combining blockchain and cloud storage. The blockchain module is based on the Hyperledger Fabric consortium blockchain, using SHA and 256 hash encryption algorithms to ensure data immutability. The cloud storage module utilizes Alibaba Cloud OSS, employing a multi-node backup mechanism (3 backup nodes) for off-site backup. Stored data covers raw material batch information, production equipment serial numbers, operator identification information, process parameters for each step, quality inspection data, and finished product delivery information. Raw material batch information includes batch number, supplier, material testing report, and warehousing time; production equipment serial numbers include equipment ID, model, and maintenance records; operator identification information includes employee ID, name, and position; process parameters for each step include stamping pressure, stacking thickness, and curing temperature; quality inspection data includes magnetic properties, dimensions, and appearance defect judgment results; and finished product delivery information includes delivery order number, customer name, and delivery time. The blockchain block structure includes a block header and a block body. The block header contains a version number, the hash value of the previous block, a timestamp, and a Merkle root hash; the block body contains traceability data and transaction information. The hash value of each block is associated with the previous block, ensuring data immutability.

[0076] A traceability chain is established based on the actual flow logic of raw material warehousing, process processing, finished product inspection, and outbound delivery. The chain association is achieved through a hash mapping of a unique batch identifier, which consists of an 8-digit date code and a 6-digit serial number, such as 20250610000001. The association rules are adapted to the actual execution characteristics of the production process using a chain node association algorithm. Chain nodes correspond to key data collection points in each production process through data collection point identifiers; for example, the raw material warehousing node corresponds to the raw material characteristic collection component, and the stamping process node corresponds to the stamping parameter collection point. Taking the iron core with batch number 20250610000001 as an example, its traceability chain is: Raw material warehousing (batch A001) → Stamping equipment 1 (parameters: pressure 80kN, frequency 30 times / minute) → Lamination equipment 2 (parameters: pressure 40kN, thickness 20mm) → Curing oven 3 (parameters: temperature 150℃, duration 2 hours) → Finished product inspection (qualified) → Outbound delivery (customer B).

[0077] The system interfaces with the read interface of the distributed data storage component via a dedicated query interface. This dedicated query interface is a RESTful API that supports queries by batch number, product number, production date, raw material batch, and other dimensions. The query response time is ≤1 second, and the query results are output to the MES system and operations and maintenance unit in standardized report formats, including Excel and PDF. The reports contain complete traceability data, detailed parameters for each stage, quality assessment results, and other information.

[0078] The operation and maintenance unit includes equipment status monitoring components, intelligent fault diagnosis modules, and maintenance plan generation modules, enabling status monitoring, fault diagnosis, and precise maintenance of production equipment.

[0079] The system is equipped with vibration sensors, temperature sensors, current sensors, and energy consumption monitoring equipment to collect data on vibration frequency, operating temperature, operating current, and real-time energy consumption of the production equipment. The vibration sensor used is PCB352C65, the temperature sensor is PT100, the current sensor is LEMLA55 / P, and the energy consumption monitoring equipment is a DL / T645 / 2007 smart meter. The vibration frequency monitoring range is 0-10kHz, the temperature monitoring range is 20℃-200℃, the current monitoring range is 0-50A, and the energy consumption monitoring accuracy is ±0.5%. The monitoring parameters are adapted to the equipment's operational safety characteristics and performance degradation patterns using an equipment characteristic database. This database includes vibration safety thresholds for different equipment (e.g., ≤5mm / s for stamping equipment), temperature safety thresholds (e.g., ≤85℃ for motors), and current fluctuation thresholds (e.g., ±10% of rated current).

[0080] The system incorporates a built-in equipment fault feature library and a machine learning diagnostic model, based on the random forest algorithm. The fault feature library includes core features of typical equipment faults such as vibration frequency thresholds, temperature variation ranges, and current fluctuation intervals. Typical fault features include punch wear in stamping equipment (vibration frequency 3-5kHz, current fluctuation +15%), hydraulic leakage in laminating equipment (pressure drop ≥10%, vibration frequency 1-2kHz), and heating element damage in curing ovens (slow temperature rise, energy consumption increase ≥20%). The machine learning diagnostic model compares real-time monitoring data with historical fault data using a fault feature extraction algorithm, achieving a diagnostic accuracy of ≥95%, covering mechanical faults, electrical faults, and control system faults.

[0081] Based on the diagnostic results, equipment maintenance cycle, and spare parts inventory status, a targeted maintenance plan is generated. The plan includes information such as the maintenance location, maintenance process, tool list, spare parts requirements, and maintenance time limit. Taking a punch wear failure in a stamping machine as an example, the maintenance plan is as follows: maintenance location (punch), maintenance process (disassembly of punch → grinding and repair / replacement → installation and calibration → trial operation), tool list (wrench, calibrator), spare parts requirements (punch spare parts models XC, 100, 01), and maintenance time limit (4 hours). The maintenance plan is pushed to the maintenance terminal via a dedicated communication protocol. The dedicated communication protocol uses the MQTT protocol, and the maintenance terminal includes smartphones and tablets to ensure that maintenance personnel receive and execute the plan promptly.

[0082] The power supply unit adopts a dual-mode power supply with mains power as the primary source and energy storage as backup. It integrates a power management chip, a voltage stabilization module, and a load detection module. The load detection module collects production load data in real time and dynamically adjusts the output voltage and power through a power regulation algorithm. The power management chip used is TIBQ76952, the voltage stabilization module is LM2596S, and the load detection module is ACS712.

[0083] The mains power supply is converted from AC220V to DC24V by a power module. The power module is MeanWellRD 65B, with an output voltage of 24V±0.5V and an output power of 50W and 500W. The backup energy storage module uses a lithium battery pack with a capacity of 200Ah and a voltage of 24V. When the mains power is interrupted, it automatically switches to lithium battery power supply, with a battery life of ≥8 hours and a switching time of ≤10ms to ensure uninterrupted power supply to the system.

[0084] Under high production load conditions, such as when multiple devices are running simultaneously or when digital twin simulation is in operation, the power management chip automatically increases the output power to 300 or 500W to prioritize the power supply stability of production equipment, data acquisition units, MES integration units, and digital twin units. Output voltage fluctuations are controlled within the safe threshold range (±2%) of the workshop electrical system by the PID regulation algorithm of the voltage stabilization module. Under low production load or shutdown conditions, the power supply unit reduces the power supply to non-core modules to 50 or 100W. Non-core modules include the query output module, while simultaneously charging the lithium battery pack.

[0085] It features overvoltage, overcurrent, and overheat protection functions, with protection thresholds adapted to the electrical safety standards database to meet the safety requirements of workshop electrical systems. Overvoltage protection threshold ≥28V, overcurrent protection threshold ≥20A, and overheat protection threshold ≥60℃. Upon triggering protection, it automatically disconnects non-core circuits and sends an early warning signal to the maintenance unit, ensuring equipment and personnel safety.

[0086] The input of the closed-loop feedback correction unit is connected to the process parameter output of the data acquisition unit, the judgment result output of the quality control unit, and the simulation data output of the digital twin unit to collect actual production operation data, quality judgment results, and virtual simulation comparison data.

[0087] The deviation value between the collected data and the production target data issued by the MES system is calculated using a deviation quantification algorithm. The deviation value is calculated as: Deviation value = |Actual data / Target data| / Target data × 100%. For example, if the target stamping pressure is 80kN and the actual data is 75kN, the deviation value is 6.25%; if the target pass rate is 98% and the actual pass rate is 95%, the deviation value is 3.06%.

[0088] Based on the deviation values, the scheduling parameters of the production scheduling unit, the judgment thresholds of the quality control unit, and the model coefficients of the digital twin unit are corrected in reverse. Regarding the correction of production scheduling parameters, if the stamping pressure deviation is ≥5%, the pressure command of the stamping equipment is adjusted, for example, from 75kN to 80kN; if the process completion progress deviation is ≥10%, the equipment capacity allocation is adjusted, for example, increasing the number of stamping equipment in operation. Regarding the correction of quality judgment thresholds, if the pass rate of a batch of iron cores is low (deviation ≥5%) due to fluctuations in raw material characteristics, the company's internal control thresholds are fine-tuned, provided they meet industry standards; for example, the lower limit of internal control for magnetic permeability is adjusted from 3000μH / m to 2800μH / m. Regarding the correction of digital twin model coefficients, if the deviation between the virtual simulation product quality and the actual test results is ≥8%, the correlation coefficients of process parameters in the model are corrected, for example, the influence coefficient of curing temperature on iron loss.

[0089] The correction process is implemented through a PID algorithm with a proportional coefficient Kp=0.8, an integral coefficient Ki=0.05, and a derivative coefficient Kd=0.1. The correction parameters are adapted to the dynamic balance requirements of core production efficiency and product quality through parameter mapping rules. A technical closed loop is formed by relying on the cross-unit data linkage mechanism to ensure continuous optimization of system performance.

[0090] The system's full workflow diagram provided in this application embodiment includes the following specific steps:

[0091] Data acquisition phase: The multi-sensor fusion module of the data acquisition unit collects raw material characteristics, process parameters, and finished product testing data. After noise reduction and feature extraction by the edge computing node, the data is synchronized to the MES integration unit and the digital twin unit.

[0092] MES Interaction Phase: The MES integration unit converts the collected data into a standardized format and transmits it to the existing MES system, while simultaneously receiving production task instructions and process parameters from the MES system.

[0093] Scheduling decision-making phase: The production scheduling unit breaks down production tasks, generates scheduling instructions based on the working condition assessment results, and issues them to the production equipment control module;

[0094] Production execution phase: Production equipment executes production processes according to scheduling instructions, and the quality control unit monitors product quality in real time and generates judgment results;

[0095] Digital twin simulation stage: The digital twin unit calibrates the virtual model based on real-time acquired data, performs multi-scenario simulations, and outputs optimization suggestions;

[0096] Traceability and Operation & Maintenance Phase: The traceability unit stores all process data and establishes a traceability link, while the operation and maintenance unit monitors equipment status, diagnoses faults, and generates maintenance plans.

[0097] Closed-loop correction stage: The closed-loop feedback correction unit calculates the deviation between the actual data and the target data, corrects the parameters of each unit in reverse, and forms a technical closed loop.

[0098] While an overview of the subject matter has been described with reference to specific example embodiments, various modifications and changes can be made to these embodiments without departing from the broader scope of embodiments of this disclosure. Such embodiments of the subject matter are referred to herein, individually or collectively, as inventions, for convenience only, and if more than one disclosure or concept is disclosed in fact, it is not intended to limit the scope of this application to any single disclosure or concept.

[0099] The embodiments described herein have been described in sufficient detail to enable those skilled in the art to practice the disclosed teachings. Other embodiments may be used and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure. Therefore, the detailed description should not be construed as limiting, and the scope of the various embodiments is defined only by the appended claims and the full scope of their equivalents.

Claims

1. A fully intelligent digital management system for iron core production based on MES, characterized in that, It includes a data acquisition unit, a MES integration unit, a production scheduling unit, a quality control unit, a digital twin unit, a traceability unit, an operation and maintenance unit, and a power supply guarantee unit; Each unit establishes a data interaction link through industrial Ethernet. The communication link adopts the Profinet protocol. The frequency band is adapted to the electromagnetic environment characteristics of the iron core production workshop through a spectrum adaptation algorithm. The signal transmission timing is anchored to the production process timing characteristics by the process timing synchronizer. The data acquisition unit integrates a multi-sensor fusion module, an edge computing node, and a data preprocessing module. The edge computing node and the multi-sensor fusion module construct a collaborative link for noise reduction and feature extraction through real-time data streams. The preprocessed data is synchronized to the MES integration unit and the digital twin unit. The MES integration unit is configured with a heterogeneous data adaptation interface, a real-time interaction engine, and a data verification module. It builds a bidirectional data link with the existing MES system through the interface adaptation protocol. The data interaction format and the MES system interface specification form a one-to-one mapping. The verification module verifies the data integrity through the CRC32 algorithm. The production scheduling unit has a built-in task decomposition module, a hybrid optimization algorithm engine, and a working condition feedback module. The hybrid optimization algorithm engine adopts a hybrid strategy that combines genetic algorithms and rule-based reasoning. The working condition feedback module transmits the equipment operating status in real time to dynamically adjust the scheduling parameters. The quality control unit is equipped with a multi-dimensional detection module, a data processing chip, and a standard threshold storage module. The detection parameters are mapped to the quantitative indicators of the iron core production industry standards and the company's internal control standards. The digital twin unit constructs a high-fidelity virtual model containing workshop layout, equipment structure, and process logic. The model forms a real-time calibration link with the data acquisition unit through a parameter iteration algorithm. Each unit forms a linkage architecture through hardware signal links and custom data interaction protocols. The architecture embeds technical mechanisms for cross-unit parameter linkage, verification, and iterative correction.

2. The intelligent digital management system for the entire iron core production process based on MES as described in claim 1, characterized in that, The raw material characteristic acquisition component of the data acquisition unit is equipped with material analysis equipment, size measurement equipment, and magnetic property detection equipment, and the monitoring dimensions cover the material composition, geometric dimensions, magnetic permeability, and iron loss value of the raw materials for iron core production. The sensing elements of the process parameter acquisition module are deployed in the key equipment of the stamping, lamination, and curing processes. The sensing elements are amplified and filtered by the signal amplification and filtering circuit to adapt to the change law of the process parameters of each process. The output signal is processed by the Kalman filtering algorithm of the edge computing node and then transmitted to the data preprocessing module. The output of the finished product testing and acquisition module establishes a physical connection with the input interface of the quality control unit through a data calibration link. The test data format is compatible with the quality judgment quantification logic through a data format mapping table. All acquisition components and modules use the Modbus communication protocol, and the sampling frequency is anchored to the production process rhythm by the process cycle synchronizer.

3. The intelligent digital management system for the entire iron core production process based on MES as described in claim 1, characterized in that, The interface adapter component of the MES integration unit adopts a dual interface design of RS485 and Ethernet, and adapts to the existing MES system hardware interface type through interface pin definition. The data conversion module has a built-in heterogeneous data parsing engine, format mapping rule base and data verification algorithm, which converts the equipment data, test data and operating condition data collected by each unit into a standardized format compatible with the MES system. The conversion logic follows the data exchange standard of the iron core production industry. The interactive collaboration module receives production task instructions, process adjustment parameters, and quality standard update information issued by the MES system. After verification by the data verification algorithm, it transmits the data to the data conversion module and synchronously outputs production progress statistics, quality inspection results, and other data to the MES system.

4. The intelligent digital management system for the entire iron core production process based on MES as described in claim 1, characterized in that, The task parsing component of the production scheduling unit breaks down production tasks according to the sequence and dependency of iron core production processes, forming a hierarchical process task list. The operating condition assessment module collects data such as production equipment load rate, raw material supply status, process completion progress, and inspection qualification status, and generates operating condition assessment results through a weighted summation algorithm. The scheduling instruction generation module has a built-in rule base for priority rules of iron core production processes and an equipment load balancing model. The rule base includes core rules such as process connection intervals, equipment capacity limits, and raw material consumption thresholds. The model optimizes equipment load distribution through linear programming algorithms and outputs instructions for equipment start-up and shutdown, process adjustment, capacity allocation, and raw material scheduling. The instruction parameters are adapted to the operating characteristics and process requirements of the production equipment through an equipment operation parameter mapping table.

5. The intelligent digital management system for the entire iron core production process based on MES as described in claim 2, characterized in that, The multi-dimensional detection components of the quality control unit include high-precision magnetic property detection equipment, laser dimensional accuracy measurement equipment, and machine vision appearance defect recognition equipment. The detection parameters cover the magnetic permeability, iron loss value, geometric dimensional tolerance, and surface defect density of the finished iron core. The data processing module uses the Kalman filter algorithm for noise reduction and extracts key information through a feature extraction model. The processed data is then mapped to the quantization threshold of the standard threshold storage module using a data mapping algorithm. The quality assessment module has a built-in dual threshold library of iron core production industry standards and enterprise internal control standards. It generates a judgment result of qualified, rework, or scrap by calculating the difference between the test data and the standard threshold. The judgment result is transmitted to the MES system and traceability unit via data synchronization protocol.

6. The intelligent digital management system for the entire iron core production process based on MES as described in claim 1, characterized in that, The full-process model building component of the digital twin unit adopts a multi-software collaborative modeling architecture. Through the parameter association algorithm of the three-dimensional geometric modeling module and the dynamic simulation module, it realizes a one-to-one correspondence mapping between workshop layout, equipment structure, process flow logic and physical production system. The mapping relationship is defined by the parameter association matrix. The real-time calibration module interacts in real time with the raw material data, process parameters, and finished product test data of the data acquisition unit. It iteratively corrects the model parameters using the least squares method, and the corrected parameters are written into the virtual model through the model update interface. The virtual simulation testing component builds a multi-scenario simulation platform. By modeling physical laws and embedding process constraints into algorithms, it adapts to the physical laws and process requirements of actual production, supporting test scenarios such as production load fluctuations, equipment failure simulation, process parameter adjustment, and changes in raw material characteristics.

7. The intelligent digital management system for the entire iron core production process based on MES as described in claim 1, characterized in that, The traceability unit's distributed data storage component adopts a hybrid architecture combining blockchain and cloud storage. The blockchain module uses a hash encryption algorithm to ensure data immutability, while the cloud storage module uses a multi-node backup mechanism to achieve off-site backup. The stored data covers raw material batch information, production equipment number, operator identity information, process parameters of each process, quality inspection data, and finished product delivery information. The traceability link construction module establishes associated links according to the actual flow logic of raw material warehousing, process processing, finished product inspection, and outbound delivery. The link association is realized through hash mapping of unique batch identifiers. The association rules are adapted to the actual execution characteristics of the production process by the link node association algorithm. The link nodes correspond to the key data collection points of each production process through data collection point identifiers. The multi-dimensional query output module connects to the read interface of the distributed data storage component through a dedicated query interface, and the query results are output to the MES system and operation and maintenance unit in a standardized report format.

8. The intelligent digital management system for the entire iron core production process based on MES as described in claim 1, characterized in that, The equipment status monitoring component of the operation and maintenance unit is equipped with vibration sensors, temperature sensors, current sensors, and energy consumption monitoring devices. It collects data such as vibration frequency, operating temperature, operating current, and real-time energy consumption of the production equipment. The monitoring parameters are adapted to the equipment operation safety characteristics and performance degradation law by the equipment characteristic database. The intelligent fault diagnosis module has a built-in equipment fault feature library and machine learning diagnostic model. The fault feature library contains core features of typical equipment faults, such as vibration frequency threshold, temperature change range, and current fluctuation range. The machine learning diagnostic model compares real-time monitoring data with historical fault data through fault feature extraction algorithms, covering diagnostic types such as mechanical faults, electrical faults, and control system faults. The maintenance plan generation module generates targeted maintenance plans based on diagnostic results, equipment maintenance cycles, and spare parts inventory status. The plans include information such as maintenance locations, maintenance processes, tool lists, spare parts requirements, and maintenance time limits, and are pushed to the operation and maintenance terminal through a dedicated communication protocol.

9. The intelligent digital management system for the entire iron core production process based on MES as described in claim 1, characterized in that, The power supply guarantee unit adopts a dual-mode power supply with mains power as the primary source and energy storage as backup. It has a built-in power management chip, voltage stabilization module and load detection module. The load detection module collects production load data in real time and dynamically adjusts the output voltage and power through a power regulation algorithm. Under high production load conditions, the power management chip automatically increases the output power to prioritize the power supply stability of production equipment, data acquisition unit, MES integration unit and digital twin unit. The output voltage fluctuation is controlled within the safety threshold range of the workshop electrical system by the PID regulation algorithm of the voltage stabilization module. During low-load or shutdown conditions, the power supply unit reduces the power supply to non-core modules while charging the energy storage module composed of lithium battery packs. The unit has overvoltage, overcurrent, and overheat protection functions. The protection thresholds are adapted to the electrical safety requirements of the workshop electrical system by the electrical safety standard database. After the protection is triggered, the non-core circuit is automatically cut off through the protection circuit and a warning signal is sent.

10. The intelligent digital management system for the entire iron core production process based on MES as described in claim 5, characterized in that, The system also includes a closed-loop feedback correction unit; the input end of the closed-loop feedback correction unit establishes a data connection with the process parameter output end of the data acquisition unit, the judgment result output end of the quality control unit, and the simulation data output end of the digital twin unit to collect actual production operation data, quality judgment results, and virtual simulation comparison data. The deviation value between the collected data and the production target data issued by the MES system is calculated by the deviation quantification algorithm. Based on the deviation value, the scheduling parameters of the production scheduling unit, the judgment threshold of the quality control unit, and the model coefficients of the digital twin unit are corrected in reverse. The correction process is implemented by the PID algorithm. The correction parameters are adapted to the dynamic balance requirements of iron core production efficiency and product quality through parameter mapping rules, and a technical closed loop is formed by relying on the cross-unit data linkage mechanism.

Images