Plate production control system with heating function

Abstract

The invention relates to the technical field of control systems, in particular to a plate production control system with a heating function, which comprises a cross-domain data fusion module for acquiring deformation data, energy transfer data, electrical signal data, acoustic emission signals and stress distribution images through a distributed sensor network; generating a characteristic matrix containing a mapping relation of thermal resistance distribution, stress gradient and current density based on a physical field coupling model; the multi-parameter coupling control module synchronously outputs a temperature gradient adjustment signal, a pressure distribution control signal and a current density modulation signal according to the characteristic matrix; the heat flow coordinated regulation and control module combines conductive layer power data and temperature field characteristics; the digital twinning optimization module constructs a thermodynamic electric coupling virtual model, optimizes a weight coefficient through a genetic algorithm and returns the weight coefficient to the control module. Through cross-domain data association and multi-target cooperative control, heat conduction path stability and interface stress dynamic digestion are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of control systems, and in particular to a plate production control system with a heating function. Background Art

[0002] The production of panels with heating functions mainly achieves heat conduction and distribution through material composites and structural design. During the production process, the base material is selected from polymer or metal composite materials that take into account both thermal conductivity and strength. A uniformly distributed conductive heating layer or heat conduction medium or composite conductive heating layer is embedded inside, and the functional layer and protective layer are tightly combined through a lamination process. To prevent excessive heat concentration, a partitioned circuit design or gradient material arrangement is adopted to allow heat to diffuse along a preset path. The surface treatment process optimizes thermal radiation efficiency while improving wear resistance, decorativeness, and aging resistance, while the packaging technology isolates internal components from the external environment, taking into account electrical safety and long-term stability. This type of panel needs to undergo thermal distribution simulation and durability testing to verify its adaptability in the building heating system, and ultimately form an invisible heating unit that can be integrated with walls, floors, and ceilings. The thermal effect is triggered by electrical energy or hot water circulation, balancing the temperature of the space and reducing the space occupied by existing heating equipment, thereby improving heating comfort.

[0003] The lack of a dynamic coordination mechanism for multi-level production parameters in existing technologies leads to unstable heat generation and conduction efficiency. Specifically, during the lamination process, when the precision of the functional layer's fit with the substrate is affected by both temperature gradients and pressure fluctuations, there's no real-time feedback system to coordinately compensate for the current density of the conductive heating layer and the flow rate of the heat transfer medium. This causes the internal thermal resistance distribution of the board to deviate from the preset model, affecting the accuracy of the heating unit's fit within the building structure. This flaw stems from the existing segmented control system's failure to establish a cross-domain correlation model between material deformation, energy transfer, and electrical signals, which can easily lead to thermal field distortion and interfacial stress accumulation in continuous production scenarios. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the present invention provides a plate production control system with a heating function. The present invention solves the problem of unstable heat conduction efficiency caused by the lack of a dynamic coordination mechanism for multi-level production parameters in the existing segmented control system.

[0005] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0006] The present invention provides a plate production control system with a heating function, comprising:

[0007] A cross-domain data fusion module is configured to collect deformation data, energy transfer data, electrical signal data, acoustic emission signals, and stress distribution images through a distributed sensor network, pre-process the deformation data, energy transfer data, electrical signal data, acoustic emission signals, and stress distribution images based on a preset physical field coupling model, and generate a feature matrix including the mapping relationship between thermal resistance distribution, stress gradient, and current density;

[0008] a multi-parameter coupling control module configured to receive a characteristic matrix, generate a temperature gradient adjustment signal according to a thermal resistance distribution, generate a pressure distribution control signal according to a stress gradient, generate a current density modulation signal according to a current density, and synchronously output the temperature gradient adjustment signal, the pressure distribution control signal, and the current density modulation signal;

[0009] A heat flow collaborative control module is configured to receive the temperature gradient adjustment signal and the current density modulation signal, combine the conductive layer power data extracted from the electrical signal data in the cross-domain data fusion module with the substrate temperature field characteristics generated by the energy transfer data in the characteristic matrix, and generate the heat transfer medium flow rate control parameters through a preset Bayesian optimization algorithm;

[0010] An interface stress elimination module is configured to receive pressure distribution control signals and heat transfer medium flow rate control parameters, analyze acoustic emission signals and stress distribution images in the cross-domain data fusion module, generate asymmetric compensation instructions, and drive the distributed actuator array to adjust the lamination roller pressure;

[0011] The digital twin optimization module is configured to receive the characteristic matrix of the cross-domain data fusion module as physical entity data, generate virtual simulation data corresponding to the thermal resistance distribution, stress gradient and current density through the built-in thermodynamic simulation engine, and optimize the weight coefficients of the thermal resistance distribution, stress gradient and current density through the preset multi-objective optimization algorithm, and return the optimized weight coefficients to the multi-parameter coupling control module.

[0012] Furthermore, the plate production control system with heating function of the present invention also includes: the output end of the cross-domain data fusion module transmits the characteristic matrix to the multi-parameter coupling control module, the input end of the heat flow collaborative control module receives the temperature gradient adjustment signal and the current density modulation signal of the multi-parameter coupling control module, the input end of the interface stress elimination module receives the heat conduction medium flow rate control parameter of the heat flow collaborative control module, and the output end of the digital twin optimization module feeds back the optimized weight coefficient to the multi-parameter coupling control module to update the temperature gradient adjustment signal, the pressure distribution control signal and the current density modulation signal to form a closed-loop control link;

[0013] In the closed-loop control link, the cross-domain data fusion module transmits the thermal resistance distribution, stress gradient and current density parameters in the characteristic matrix to the multi-parameter coupling control module; the optimized weight coefficient fed back by the digital twin optimization module is used to update the temperature control threshold in the temperature gradient adjustment signal, the pressure set value in the pressure distribution control signal and the duty cycle parameter in the current density modulation signal.

[0014] Furthermore, the cross-domain data fusion module of the plate production control system with heating function of the present invention includes:

[0015] a deformation data acquisition unit configured to acquire deformation data of the substrate using a high-precision strain sensor and a laser profilometer;

[0016] an energy data acquisition unit configured to acquire temperature field distribution data through an infrared thermal imager and a heat flow sensor;

[0017] an electrical data acquisition unit configured to monitor current-carrying parameters of the conductive layer via a current transformer and a voltage probe;

[0018] The data processing unit is configured to perform timestamp alignment and cubic spline interpolation processing on the data output by the deformation data acquisition unit, the energy data acquisition unit and the electrical data acquisition unit, and extract the thermal resistance distribution, stress gradient and current density as extraction features through principal component analysis, and generate a thermo-electric synergy feature matrix through the extracted features obtained by the principal component analysis.

[0019] Furthermore, the plate production control system with heating function of the present invention includes a multi-parameter coupling control module:

[0020] a temperature gradient adjustment unit configured to generate a heating power compensation instruction for each temperature control zone according to the thermal resistance distribution parameter in the characteristic matrix;

[0021] a pressure distribution control unit configured to adjust the pressure output of the hydraulic servo system based on the stress gradient parameters in the characteristic matrix and the real-time feedback signal from the deformation data acquisition unit;

[0022] The current density modulation unit is configured to dynamically adjust the pulse width modulation signal according to the thermal resistance distribution map.

[0023] Furthermore, the heat flow coordinated control module of the plate production control system with heating function of the present invention includes:

[0024] a Bayesian optimization unit configured to calculate a flow rate equilibrium threshold of the heat transfer medium based on the conductive layer power data received by the heat flow coordinated control module and energy transfer data in the characteristic matrix, wherein the energy transfer data includes a surface temperature field distribution of the substrate collected by the infrared thermal imager;

[0025] The compound control execution unit is configured to perform feedforward and feedback compound control on the medium flow rate through the proportional valve and the variable frequency pump group, wherein the feedforward module predicts the flow rate demand based on the flow rate balance threshold, and the feedback module corrects the deviation according to the thermal flow sensor data.

[0026] Furthermore, the panel production control system with heating function of the present invention includes an interface stress relief module comprising:

[0027] A stress detection unit is configured to analyze the acoustic emission signal and stress distribution image in the cross-domain data fusion module, wherein the acoustic emission signal is collected by the acoustic emission sensor to generate the interlayer crack signal, and the stress distribution image is generated by analyzing the substrate surface displacement field using a preset digital image analysis model;

[0028] The dynamic compensation unit is configured to drive the piezoelectric ceramic actuator and the pneumatic fine-tuning mechanism based on the stress gradient direction and deformation data.

[0029] Furthermore, the digital twin optimization module of the plate production control system with heating function of the present invention includes:

[0030] a thermodynamic simulation engine configured to construct a coupled model of the virtual production environment based on the thermal resistance distribution, stress gradient, and current density parameters in the characteristic matrix and to predict the deviation trend of the thermal resistance distribution relative to a preset diffusion model;

[0031] The multi-objective optimization unit is configured to collaboratively optimize the thermal resistance uniformity, stress threshold and energy efficiency of the control weight coefficient through a genetic algorithm.

[0032] Furthermore, the plate production control system with heating function of the present invention further includes:

[0033] a data synchronization unit configured to eliminate timing deviations of multi-source data through timestamp alignment and cubic spline interpolation algorithm;

[0034] The protocol conversion unit is configured to convert the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image synchronization unit processed deformation data, energy transfer data and electrical signal data into the OPC-UA protocol data format supported by the cross-domain data fusion module, and transmit the data to the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image processing unit via industrial Ethernet. The output end of the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image synchronization unit is connected to the input end of the protocol conversion unit, and the output end of the protocol conversion unit is communicatively connected to the input end of the cross-domain data fusion module.

[0035] Beneficial effects of the present invention:

[0036] The present invention integrates deformation data, energy transfer data, electrical signal data and acoustic emission signals through a cross-domain data fusion module, and generates a characteristic matrix of thermal resistance distribution, stress gradient and current density mapping relationships based on a physical field coupling model, providing a unified data benchmark for multi-parameter dynamic coordination; the multi-parameter coupling control module synchronously outputs temperature gradient adjustment signals, pressure distribution control signals and current density modulation signals based on the characteristic matrix, realizing the coordinated control of heat conduction path regulation, lamination pressure balance and conductive layer power matching; the heat flow coordinated control module generates medium flow rate control parameters through a Bayesian optimization algorithm, and combines the acoustic emission signal analysis and dynamic compensation mechanism of the interface stress elimination module to suppress thermal field distortion and residual stress accumulation; the digital twin optimization module constructs a thermo-mechanical-electric coupling virtual model, and uses a genetic algorithm to perform multi-objective optimization of thermal resistance uniformity, stress threshold and energy efficiency, forming a closed-loop iterative feedback mechanism. The above technical solution breaks through the parameter island limitation of traditional segmented control, improves the stability of heat conduction efficiency and the adaptability of the production system through cross-domain data association and intelligent algorithm nesting, and enhances the adaptation accuracy of heating panels and building structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0038] Figure 1 This is a system architecture diagram of a plate production control system with heating function provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. The technical solutions provided by each embodiment of the present invention are described in detail below in conjunction with the drawings. In order to better understand the purpose of the present invention, the present invention is further described in detail below.

[0040] See also Figure 1 The present invention provides a plate production control system with a heating function, comprising:

[0041] The cross-domain data fusion module is configured to collect deformation data, energy transfer data, electrical signal data, acoustic emission signals and stress distribution images through a distributed sensor network, preprocess the deformation data, energy transfer data, electrical signal data, acoustic emission signals and stress distribution images based on a preset physical field coupling model, and generate a feature matrix including thermal resistance distribution, stress gradient and current density mapping relationship;

[0042] The multi-parameter coupling control module is configured to receive the feature matrix, generate a temperature gradient adjustment signal according to the thermal resistance distribution, generate a pressure distribution control signal according to the stress gradient, generate a current density modulation signal according to the current density, and synchronously output the temperature gradient adjustment signal, the pressure distribution control signal and the current density modulation signal.

[0043] The heat flow collaborative regulation module is configured to receive the temperature gradient adjustment signal and the current density modulation signal, combine the conductive layer power data extracted from the electrical signal data in the cross-domain data fusion module and the substrate temperature field characteristics generated from the energy transfer data in the feature matrix, and generate heat conduction medium flow rate regulation parameters through a preset Bayesian optimization algorithm.

[0044] The interface stress resolution module is configured to receive the pressure distribution control signal and the heat conduction medium flow rate regulation parameters, analyze the acoustic emission signals and stress distribution images in the cross-domain data fusion module, generate an asymmetric compensation instruction and drive a distributed actuator array to adjust the laminating roller pressure.

[0045] The digital twin optimization module is configured to receive the feature matrix of the cross-domain data fusion module as physical entity data, generate virtual simulation data corresponding to the thermal resistance distribution, stress gradient and current density through a built-in thermodynamic simulation engine, and optimize the weight coefficients of the thermal resistance distribution, stress gradient and current density through a preset multi-objective optimization algorithm, and return the optimized weight coefficients to the multi-parameter coupling control module.

[0046] The present application provides a kind of plate production control system with heating function, deformation data, energy transfer data, electrical signal data, acoustic emission signals and stress distribution images are collected by cross-domain data fusion module, and multiple-source data is preprocessed using preset physical field coupling model.The preprocessing process includes time-space alignment, noise reduction processing and principal component analysis, and thermal resistance distribution, stress gradient and current density are extracted as key features to generate thermal-electricity collaborative action feature matrix.The feature matrix is transmitted to multi-parameter coupling control module through industrial Ethernet, as the input reference of subsequent control instruction.

[0047] After receiving the characteristic matrix, the multi-parameter coupling control module generates heating power compensation instructions for each temperature-controlled zone based on the thermal resistance distribution parameters. The zone temperature control module adjusts the heating power to suppress local overheating. At the same time, based on the stress gradient parameters in the characteristic matrix and the real-time feedback signal from the deformation data acquisition unit, the pressure output of the hydraulic servo system is adjusted to achieve micron-level precision control of the laminating roller pressure. The current density modulation unit dynamically adjusts the duty cycle of the pulse width modulation signal based on the thermal resistance distribution map to match the heating power of the conductive layer with the thermal conductivity of the substrate. The temperature gradient adjustment signal, pressure distribution control signal, and current density modulation signal are synchronously output to the downstream module to ensure the timing coordination of the control instructions.

[0048] After receiving the temperature gradient adjustment signal and current density modulation signal, the heat flow coordinated control module uses a Bayesian optimization algorithm to calculate the flow velocity equilibrium threshold for the heat transfer medium, combining the conductive layer power data extracted from the electrical signal data in the cross-domain data fusion module with the substrate surface temperature field distribution data collected by the infrared thermal imager in the characteristic matrix. The feedforward module predicts the flow velocity demand based on this threshold, and the feedback module corrects the flow velocity deviation based on the measured data from the heat flow sensor. The proportional valve and variable frequency pump group perform a composite control of the medium flow velocity to maintain the spatial consistency of the heat transfer path with the preset diffusion model.

[0049] The interface stress reduction module receives pressure distribution control signals and heat transfer medium flow rate control parameters, and analyzes the acoustic emission signals from the cross-domain data fusion module and the stress distribution image generated using digital image correlation technology. The acoustic emission sensor captures interlaminar crack signals, and the stress distribution image is generated by analyzing the substrate surface displacement field. Based on the stress gradient direction and deformation data, the dynamic compensation unit drives the piezoelectric ceramic actuator and pneumatic fine-tuning mechanism to asymmetricly compensate for the lamination roller pressure, eliminating residual stress accumulation caused by thermal field distortion.

[0050] The digital twin optimization module receives the feature matrix from the cross-domain data fusion module as physical entity data. Using its built-in thermodynamic simulation engine, it constructs a thermo-electric coupling virtual model to predict the deviation trend of thermal resistance distribution relative to the preset diffusion model. The multi-objective optimization unit uses a genetic algorithm to collaboratively optimize thermal resistance uniformity, stress threshold, and energy efficiency, generating weight coefficients that are fed back to the multi-parameter coupling control module. These optimized weight coefficients dynamically update the temperature control threshold, pressure setpoint, and duty cycle parameters, forming a cross-level closed-loop iterative learning mechanism.

[0051] The system carries out timestamp alignment and cubic spline interpolation processing on multi-source data through a data synchronization unit, and eliminates time sequence deviation.A protocol conversion unit converts deformation data, energy transfer data and electrical signal data into an OPC-UA protocol format, and transmits the data to a data processing unit through an industrial Ethernet.The data flow between modules passes through a priority channel determined by network division, ensuring low-delay transmission of critical control instructions, and realizing a full-link closed loop from data acquisition, feature extraction, dynamic control to model optimization.

[0052] Specifically, the plate production control system with heating function further comprises: the output end of the cross-domain data fusion module transmits the feature matrix to the multi-parameter coupling control module; the input end of the heat flow collaborative control module receives the temperature gradient adjustment signal and the current density modulation signal of the multi-parameter coupling control module; the input end of the interface stress resolution module receives the heat transfer medium flow rate control parameter of the heat flow collaborative control module; the output end of the digital twin optimization module feeds back the optimized weight coefficient to the multi-parameter coupling control module to update the temperature gradient adjustment signal, the pressure distribution control signal and the current density modulation signal, forming a closed-loop control link.

[0053] In the closed-loop control link, the cross-domain data fusion module transmits the thermal resistance distribution, stress gradient and current density parameters in the feature matrix to the multi-parameter coupling control module; and the optimized weight coefficient fed back by the digital twin optimization module is used to update the temperature control threshold in the temperature gradient adjustment signal, the pressure set value in the pressure distribution control signal and the duty cycle parameter in the current density modulation signal.

[0054] In the plate production control system with heating function, the cross-domain data fusion module collects deformation data, energy transfer data, electrical signal data, acoustic emission signals and stress distribution images through a distributed sensor network, and generates a feature matrix containing thermal resistance distribution, stress gradient and current density mapping relationship through a physical field coupling model preprocessing. The feature matrix is transmitted to the multi-parameter coupling control module through an industrial Ethernet, and is used as the basis for generating the temperature gradient adjustment signal, the pressure distribution control signal and the current density modulation signal. The thermal resistance distribution parameter in the feature matrix is used to divide the temperature control area and calculate the heating power compensation amount, the stress gradient parameter is combined with the real-time deformation feedback signal to adjust the pressure output of the hydraulic servo system, and the current density parameter dynamically matches the duty cycle of the pulse width modulation signal.

[0055] The heat flow coordinated control module receives the temperature gradient adjustment signal and current density modulation signal output by the multi-parameter coupling control module. Combined with the real-time power value of the conductive layer extracted from the electrical signal data in the cross-domain data fusion module and the surface temperature distribution data of the substrate collected by the infrared thermal imager within the feature matrix, it calculates the equilibrium threshold of the heat transfer medium flow rate using a Bayesian optimization algorithm. The feedforward control module predicts the flow rate demand based on the threshold, while the feedback control module corrects flow rate deviations based on real-time monitoring data from the heat flow sensor. Through coordinated adjustment of the proportional valve and variable frequency pump group, the spatial consistency of the heat transfer path with the preset diffusion model is maintained.

[0056] The interface stress reduction module receives the heat transfer medium flow rate control parameters output by the heat flow collaborative control module and analyzes the interlaminar crack signals collected by the acoustic emission sensor in the cross-domain data fusion module and the stress distribution image generated by digital image correlation technology. Spectral analysis of the acoustic emission signal identifies crack propagation characteristics, and the stress distribution image quantifies the spatial coordinates of stress concentration areas using a displacement field analysis algorithm. The dynamic compensation unit, based on the vector relationship between the stress gradient direction and deformation data, drives the piezoelectric ceramic actuator and pneumatic fine-tuning mechanism to apply asymmetric pressure compensation to the lamination rollers, eliminating residual stress at the interface.

[0057] The digital twin optimization module receives the feature matrix from the cross-domain data fusion module as physical entity data input. Using its built-in thermodynamic simulation engine, it constructs a thermodynamic-electric multi-physics coupling model to simulate the evolution of thermal resistance distribution during sheet metal production. The multi-objective optimization unit uses a genetic algorithm to collaboratively optimize thermal resistance uniformity, interface stress thresholds, and system energy efficiency to generate control weight coefficients. These optimized weight coefficients are transmitted back to the multi-parameter coupling control module via the OPC-UA protocol, dynamically adjusting the partitioned temperature control thresholds in the temperature gradient adjustment signal, the pressure setting gradient in the pressure distribution control signal, and the duty cycle range in the current density modulation signal.

[0058] The data synchronization unit deployed in the system performs timestamp alignment and cubic spline interpolation on multi-source sensor data to eliminate timing deviations caused by differences in sampling frequency. The protocol conversion unit converts the synchronized deformation data, energy transfer data, and electrical signal data into the OPC-UA protocol format and transmits it to the data processing unit of the cross-domain data fusion module via industrial Ethernet. The control instructions output by the multi-parameter coupling control module are transmitted to the execution terminal via a real-time data bus. The feedback data from the digital twin optimization module updates the control parameters via a low-latency RDMA network, forming a full-link closed-loop control from data acquisition, feature extraction, dynamic regulation, to model iteration.

[0059] Specifically, the cross-domain data fusion module of the plate production control system with heating function of the present invention includes:

[0060] a deformation data acquisition unit configured to acquire deformation data of the substrate through a high-precision strain sensor and a laser profilometer;

[0061] an energy data acquisition unit configured to acquire temperature field distribution data through an infrared thermal imager and a heat flow sensor;

[0062] an electrical data acquisition unit configured to monitor current-carrying parameters of the conductive layer via a current transformer and a voltage probe;

[0063] The data processing unit is configured to perform timestamp alignment and cubic spline interpolation processing on the data output by the deformation data acquisition unit, the energy data acquisition unit and the electrical data acquisition unit, and extract the thermal resistance distribution, stress gradient and current density as extraction features through principal component analysis, and generate a thermo-electric synergy feature matrix through the extracted features obtained by the principal component analysis.

[0064] In the plate production control system with heating function of the present invention, the deformation data acquisition unit of the cross-domain data fusion module monitors the microscopic deformation of the substrate in real time through a high-precision strain sensor, and at the same time uses a laser profiler to scan the surface profile fluctuation data of the substrate. The resistance change signal of the strain sensor and the optical measurement data of the laser profiler are converted from analog to digital and then transmitted to the data processing unit via industrial Ethernet. The energy data acquisition unit captures the surface temperature field distribution image of the substrate through an infrared thermal imager, and the heat flux sensor synchronously measures the heat flux density data. The frame data of the infrared thermal imager and the analog signal of the heat flux sensor are converted into digital signals through a signal conditioning circuit. The electrical data acquisition unit deploys a current transformer and a voltage probe to respectively collect the real-time current value and voltage fluctuation parameters of the current-carrying conductor of the conductive layer. The current signal is converted by the Hall effect, and the voltage signal is processed by a differential amplifier circuit.

[0065] After the data processing unit receives the deformation, energy and electrical data, it first performs a timestamp alignment operation and uses the global clock signal to normalize the time stamps of the data collected by each sensor to eliminate the timing offset caused by differences in the sampling cycles of the equipment. For non-uniform sampling data, the cubic spline interpolation algorithm is used to reconstruct the continuous timing curve to fill the data gaps caused by communication delays or packet loss. The pre-processed multi-source data is input into the principal component analysis model, and the eigenvector is calculated through the covariance matrix to extract the thermal resistance distribution, stress gradient and current density as key feature dimensions. The thermal resistance distribution parameter reflects the impedance change of the heat conduction path inside the substrate, the stress gradient parameter characterizes the mechanical state of the laminate interface, and the current density parameter describes the charge distribution characteristics of the current-carrying cross section of the conductive layer. The above-mentioned feature dimensions are constructed through linear combination to construct the thermodynamic and electrical synergistic effect feature matrix, which is stored in the distributed time series database after data compression to provide structured input for the downstream control module.

[0066] The strain sensor and laser profilometer in the deformation data acquisition unit form a complementary measurement mechanism. Strain data provides local deformation details, while laser profiling data depicts the overall geometric deformation trend. The two are combined to generate a three-dimensional deformation field for the substrate. The infrared thermal imager and heat flow sensor in the energy data acquisition unit form a thermal field monitoring system. Infrared images analyze surface temperature distribution, while heat flow data quantifies energy transfer efficiency. Combined with the electrical parameters of the conductive layer, an energy-current correlation model is constructed. The principal component analysis process in the data processing unit eliminates redundant noise and preserves the physical field coupling relationships between cross-domain data. The generated feature matrix is ​​pushed to the multi-parameter coupling control module via the OPC-UA protocol, driving temperature gradient adjustment, pressure distribution control, and the generation of current density modulation signals. Data from each acquisition unit is unified into a standardized format via a protocol conversion gateway. Transmission over a deterministic network ensures the real-time performance of control instructions, forming a complete technical chain from data acquisition and feature extraction to control decision-making.

[0067] Specifically, the plate production control system with heating function of the present invention includes a multi-parameter coupling control module:

[0068] a temperature gradient adjustment unit configured to generate a heating power compensation instruction for each temperature control zone according to the thermal resistance distribution parameter in the characteristic matrix;

[0069] a pressure distribution control unit configured to adjust the pressure output of the hydraulic servo system based on the stress gradient parameters in the characteristic matrix and the real-time feedback signal from the deformation data acquisition unit;

[0070] The current density modulation unit is configured to dynamically adjust the pulse width modulation signal according to the thermal resistance distribution map.

[0071] In the plate production control system with heating function of the present invention, the temperature gradient adjustment unit of the multi-parameter coupling control module receives the characteristic matrix generated by the cross-domain data fusion module and analyzes the thermal resistance distribution parameters therein. The thermal resistance distribution parameters are divided into temperature control areas through discretization processing, and each area corresponds to an independent heating power compensation instruction. The heating power compensation instruction is dynamically adjusted according to the deviation between the thermal resistance value and the preset threshold value. The fuzzy control algorithm is used to calculate the compensation coefficient, and the resistance wire or electromagnetic heating unit of the partition temperature control module is driven to adjust the output power to suppress material deformation caused by local overheating. The compensated temperature gradient data is fed back to the characteristic matrix update process through the real-time data bus to form a dynamic correction mechanism for the thermal resistance parameters.

[0072] The pressure distribution control unit generates pressure regulation instructions of the hydraulic servo system based on stress gradient parameters in the feature matrix and in combination with real-time feedback signals of the high-precision strain sensor of the deformation data acquisition unit. The stress gradient parameters determine the pressure distribution weight of the laminated interface through vector decomposition, and the deformation feedback signals are subjected to Kalman filtering to eliminate noise interference, and then are subjected to convolution operation with the weight coefficient to generate the pressure set value of the hydraulic cylinder. The servo valve adjusts the oil way flow according to the pressure set value to drive the laminated roller to apply balanced pressure to the base material, and meanwhile, the displacement sensor is used to monitor the displacement of the laminated roller to form a closed loop control circuit of pressure output.

[0073] The current density modulation unit receives the thermal resistance distribution map in the feature matrix to establish the mapping relationship between the thermal resistance value and the current-carrying cross section of the conductive layer. The duty cycle of the pulse width modulation signal is dynamically adjusted according to the gradient change of the thermal resistance distribution map, and a feedforward and feedback compound control strategy is adopted: the feedforward channel predicts the current density demand according to the thermal resistance gradient, the feedback channel monitors the actual current value of the conductive layer in real time through the current transformer, calculates the deviation and corrects the duty cycle parameter. The modulated PWM signal drives the IGBT power module to control the heating power of the conductive layer to match the heat conduction rate of the base material, so as to avoid the thermal resistance mutation caused by current overload.

[0074] The output instructions of the temperature gradient adjustment unit are transmitted to the partition temperature control module through the industrial Ethernet, the pressure set value of the pressure distribution control unit is issued to the hydraulic servo system through the CAN bus, and the PWM signal of the current density modulation unit is transmitted to the power drive circuit through the optical fiber interface. The instruction data of the multi-parameter coupled control module and the weight coefficient fed back by the digital twin optimization module are fused and operated to update the temperature control threshold, the pressure gradient and the duty cycle range, so as to form the cross-domain collaborative control of thermal and electric parameters. The control instructions and sensor feedback data of each unit are stored in association through the time sequence database, supporting abnormal traceability and model iterative optimization in the production process.

[0075] Specifically, the plate production control system with a heating function provided by the application comprises:

[0076] The Bayesian optimization unit is configured to calculate the flow velocity equalization threshold of the heat conduction medium based on the conductive layer power data received by the heat flow collaborative control module and the energy transfer data in the feature matrix, wherein the energy transfer data comprises the surface temperature field distribution of the base material collected by the infrared thermal imager.

[0077] The composite control execution unit is configured to perform feedforward and feedback compound control on the medium flow velocity through the proportional valve and the frequency conversion pump group, wherein the feedforward module predicts the flow velocity demand based on the flow velocity equalization threshold, and the feedback module corrects the deviation according to the heat flow sensor data.

[0078] In the plate production control system with heating function of the present invention, the Bayesian optimization unit of the heat flow collaborative control module receives the conductive layer power data output by the cross-domain data fusion module and the energy transfer data in the characteristic matrix. The conductive layer power data is obtained by calculating the real-time current-carrying parameters collected by the current transformer and the voltage probe, and the energy transfer data includes the surface temperature field distribution of the substrate captured by the infrared thermal imager and the heat flux density value monitored by the heat flow sensor. The Bayesian optimization unit inputs the conductive layer power data and the temperature field distribution data into the optimization model, constructs the correlation function between the flow velocity of the heat conduction medium and the heat flow path, and solves the equilibrium threshold of the flow velocity of the heat conduction medium through the Markov chain Monte Carlo sampling method. This threshold reflects the optimal solution of the medium flow velocity under the current thermal resistance distribution state, and is used to balance the thermal conductivity efficiency of the substrate and the accumulation of thermal stress at the interface.

[0079] The composite control execution unit receives the flow rate equilibrium threshold generated by the Bayesian optimization unit and implements feedforward and feedback composite control of the medium flow rate through the proportional valve and variable frequency pump group. The feedforward control module predicts the target flow rate demand based on the equilibrium threshold and generates a proportional valve opening command and a variable frequency pump group speed reference value. The feedback control module uses a heat flux sensor to collect the actual heat flux density data of the heat conduction path in real time, calculates the deviation from the theoretical value of the preset diffusion model, and uses a proportional-integral control algorithm to correct the flow regulation coefficient of the proportional valve and the dynamic response parameters of the variable frequency pump group. The feedforward command and feedback correction amount are weighted and fused at the edge computing node, and a composite control signal is output to the execution terminal.

[0080] The flow rate control parameters of the heat transfer medium are transmitted to the medium injection node at the front end of the lamination process via industrial Ethernet. The proportional valve adjusts the flow cross-sectional area of ​​the medium pipeline according to the control signal, and the variable frequency pump group adjusts the output frequency of the drive motor according to the speed reference value, collaboratively controlling the instantaneous flow rate and flow rate distribution of the medium. The regulated medium flow rate data is transmitted back to the digital twin optimization module via the OPC-UA protocol for real-time calibration of the virtual model. The correction results monitored by the heat flow sensor synchronously update the energy transfer data in the characteristic matrix, forming a dynamic tuning closed loop for the heat conduction path. The Bayesian optimization unit periodically recalculates the equilibrium threshold based on the updated characteristic matrix to adapt to the dynamic changes in the thermal resistance distribution during the production process.

[0081] Specifically, the panel production control system with heating function of the present invention, the interface stress relief module includes:

[0082] A stress detection unit is configured to analyze the acoustic emission signal and stress distribution image in the cross-domain data fusion module, wherein the acoustic emission signal is collected by the acoustic emission sensor to generate the interlayer crack signal, and the stress distribution image is generated by analyzing the substrate surface displacement field using a preset digital image analysis model;

[0083] The dynamic compensation unit is configured to drive the piezoelectric ceramic actuator and the pneumatic fine-tuning mechanism based on the stress gradient direction and deformation data.

[0084] In the panel production control system with heating function of the present invention, the stress detection unit of the interface stress dissipation module receives the acoustic emission signal and stress distribution image data transmitted by the cross-domain data fusion module. The acoustic emission signal is collected by an acoustic emission sensor arranged at the laminate interface. The characteristic frequency components of interlaminar crack initiation are extracted using a wavelet transform algorithm, and the spatiotemporal distribution characteristics of crack propagation are identified. The stress distribution image is captured by a high-speed industrial camera to capture the displacement field changes of the speckle marks on the substrate surface. The strain distribution is calculated using digital image correlation technology to generate a quantitative map containing the coordinates of the stress concentration area and the gradient direction. After data fusion of the acoustic emission signal and the stress distribution image, a vector field model of the stress state of the laminate interface is output.

[0085] The dynamic compensation unit receives the vector field model generated by the stress detection unit and the real-time deformation data from the cross-domain data fusion module, analyzing the spatial correspondence between the stress gradient direction and the substrate deformation. The piezoelectric ceramic actuator generates a high-frequency micro-displacement compensation signal based on the amplitude and direction of the stress gradient vector, applying nanometer-scale pressure adjustment to the local area of ​​the laminating roller through a rigid connection mechanism. The pneumatic fine-tuning mechanism drives the cylinder group to adjust the overall pressure distribution of the laminating roller based on the macro-profile fluctuations in the deformation data, compensating for the interfacial stress offset caused by thermal-mechanical coupling. The rapid response characteristics of the piezoelectric ceramic complement the macro-adjustment capabilities of the pneumatic mechanism, achieving multi-scale coordinated control of stress relief.

[0086] Stress relief instructions are transmitted to the lamination roller drive system via a real-time data bus. The control signal for the piezoelectric ceramic actuator uses PWM modulation technology to drive the piezoelectric stack. The pressure setpoint of the pneumatic fine-tuning mechanism adjusts the compressed air flow via a proportional valve. Compensated lamination pressure data is collected via a fiber optic sensor and fed back to the cross-domain data fusion module to update the stress gradient parameters in the feature matrix. The digital twin optimization module synchronously receives the stress relief data and verifies the effectiveness of the compensation strategy using a thermodynamic simulation engine. The optimized parameters are then transmitted back to the multi-parameter coupling control module via a closed-loop control link, dynamically adjusting the output setpoint of the pressure distribution control unit. This completes the control loop from stress detection, dynamic compensation, to model iteration.

[0087] Specifically, the digital twin optimization module of the plate production control system with heating function of the present invention includes:

[0088] a thermodynamic simulation engine configured to construct a coupled model of the virtual production environment based on the thermal resistance distribution, stress gradient, and current density parameters in the characteristic matrix and to predict the deviation trend of the thermal resistance distribution relative to a preset diffusion model;

[0089] The multi-objective optimization unit is configured to collaboratively optimize the thermal resistance uniformity, stress threshold and energy efficiency of the control weight coefficient through a genetic algorithm.

[0090] In the plate production control system with heating function of the present invention, the thermodynamic simulation engine of the digital twin optimization module receives the characteristic matrix generated by the cross-domain data fusion module, and analyzes the thermal resistance distribution, stress gradient and current density parameters therein. The thermal resistance distribution parameters are mapped to the geometric model of the virtual production environment through discrete grid division, the stress gradient parameters drive the mechanical simulation unit to calculate the interface contact stress distribution, and the current density parameters are input into the electromagnetic field simulation module to reconstruct the charge density field of the current-carrying cross section of the conductive layer. The thermodynamic simulation engine constructs a virtual model of thermodynamic and electrical synergy based on the multi-physics field coupling algorithm to simulate the dynamic evolution of the heat conduction path during the plate production process. The simulation results are compared and analyzed with the theoretical values ​​of the preset diffusion model to predict the spatial coordinates and amplitude characteristics of the thermal resistance distribution deviation trend, and generate a deviation correction recommendation data package.

[0091] The multi-objective optimization unit receives the deviation correction recommendation data package output by the thermodynamic simulation engine and uses a genetic algorithm to perform multi-objective collaborative optimization of the control weight coefficients. The fitness function integrates the thermal resistance uniformity index, the interface stress threshold constraint, and the system energy efficiency parameter, and iteratively updates the candidate solution set of the weight coefficients through crossover and mutation operators. The thermal resistance uniformity index is calculated based on the standard deviation of the thermal resistance distribution in the virtual model. The interface stress threshold constraint extracts the maximum Mises stress value of the mechanical simulation unit. The energy efficiency parameter links the current density modulation signal with the actual energy consumption data of the power module. The optimized weight coefficients are converted into executable parameters for temperature control threshold, pressure gradient, and duty cycle range through a dynamic encoding mechanism and transmitted back to the multi-parameter coupling control module via the OPC-UA protocol.

[0092] The optimized weight coefficients are fused with the real-time command data from the multi-parameter coupling control module to update the partition compensation coefficients of the temperature gradient adjustment signal, the servo valve opening reference value of the pressure distribution control signal, and the duty cycle parameters of the current density modulation signal. The updated control parameters are sent to the execution terminal via industrial Ethernet to drive the coordinated action of the partition heating unit, hydraulic servo system, and IGBT power module. The digital twin optimization module synchronously receives the control result data fed back by the physical entity sensor, dynamically adjusts the interface contact coefficient of the thermodynamic simulation engine through an incremental learning algorithm, and iteratively optimizes the prediction accuracy of the virtual model. The predicted data of the thermal resistance distribution deviation trend and the update log of the real-time control parameters are stored in a time series database, supporting production anomaly backtracking and adaptive tuning of process parameters, forming a closed-loop iterative link from virtual simulation, multi-objective optimization to physical control.

[0093] Specifically, the plate production control system with heating function of the present invention also includes:

[0094] a data synchronization unit configured to eliminate timing deviations of multi-source data through timestamp alignment and cubic spline interpolation algorithm;

[0095] The protocol conversion unit is configured to convert the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image synchronization unit processed deformation data, energy transfer data and electrical signal data into the OPC-UA protocol data format supported by the cross-domain data fusion module, and transmit the data to the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image processing unit via industrial Ethernet. The output end of the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image synchronization unit is connected to the input end of the protocol conversion unit, and the output end of the protocol conversion unit is communicatively connected to the input end of the cross-domain data fusion module.

[0096] In the plate production control system with heating function of the present invention, the data synchronization unit receives deformation data, energy transfer data and electrical signal data from the distributed sensor network. The deformation data is collected by high-precision strain sensors and laser profilers, the energy transfer data is obtained by infrared thermal imagers and heat flow sensors, and the electrical signal data is monitored by current transformers and voltage probes. The data synchronization unit aligns the timestamps of each sensor data based on the global clock signal to eliminate the timing offset caused by the difference in sampling frequency. For non-uniformly sampled data streams, the cubic spline interpolation algorithm is used to reconstruct the continuous timing curve, fill the gaps caused by communication delays or data packet loss, and generate a multi-source data sequence with consistent time dimensions.

[0097] The protocol conversion unit receives the aligned data output by the data synchronization unit and converts the strain signals and contour fluctuation data in the deformation data, the temperature field distribution and heat flux density values ​​in the energy transfer data, and the current and voltage parameters in the electrical signal data into the structured data format specified by the OPC-UA protocol. The protocol conversion process is implemented through an embedded gateway. The deformation data is mapped into a three-dimensional coordinate point set in the OPC-UA object model, the energy data is converted into a temperature-heat flux correlation matrix, and the electrical data is encapsulated into current-voltage time-series data packets. The converted data is transmitted via Industrial Ethernet to the data processing unit of the cross-domain data fusion module. The transmission process uses deterministic network technology to divide data priority channels to ensure the real-time performance of critical control data.

[0098] The output end of the data synchronization unit is connected to the input end of the protocol conversion unit through a high-speed serial bus, and the output end of the protocol conversion unit establishes a communication link with the input end of the cross-domain data fusion module through an industrial Ethernet switch. After receiving the OPC-UA protocol data, the data processing unit of the cross-domain data fusion module extracts the strain gradient parameters in the deformation data, the temperature field distribution characteristics in the energy data, and the current-carrying cross-section parameters in the electrical data, and inputs them into the physical field coupling model for feature fusion. The synchronized multi-source data is processed by principal component analysis and dimensionality reduction to generate a thermal resistance distribution, stress gradient, and current density mapping relationship matrix, providing standardized input for the downstream control module. The data format conversion log and synchronization timing information of the protocol conversion unit are stored in the edge node cache area, supporting data traceability and abnormal diagnosis in the production process, forming a full-link data connection from data acquisition, synchronization processing to protocol conversion.

[0099] Explanation of the technical features of the present invention:

[0100] Cross-domain data fusion module:

[0101] Deformation data: collected through high-precision strain sensors and laser profilometers, reflecting the microscopic deformation (such as local stretching / compression) and macroscopic profile fluctuations (such as surface flatness deviation) of the substrate during the lamination process.

[0102] Energy transfer data: obtained by an infrared thermal imager (capturing the surface temperature distribution of the substrate) and a heat flux sensor (measuring the heat flux density value) to quantify the thermal energy conduction efficiency and distribution state.

[0103] Electrical signal data: collected through current transformers (monitoring the current intensity of the conductive layer) and voltage probes (measuring voltage fluctuations) to characterize the current-carrying characteristics and impedance changes of the conductive layer.

[0104] Physical field coupling model: A mathematical model that correlates multiple physical field data such as deformation, energy, and electrical, extracts key features such as thermal resistance distribution (thermal path impedance), stress gradient (interface mechanical state), and current density (charge distribution) through principal component analysis, and generates a cross-domain correlation feature matrix.

[0105] Multi-parameter coupling control module:

[0106] Temperature gradient adjustment signal: Based on the thermal resistance distribution parameters in the characteristic matrix, the temperature control area is divided and a heating power compensation instruction (such as adjusting the power of the resistance wire) is generated to suppress uneven material expansion caused by local overheating.

[0107] Pressure distribution control signal: Combines stress gradient parameters (reflecting interface stress distribution) with real-time deformation feedback signals to dynamically adjust the pressure output of the hydraulic servo system to ensure that the laminating roller applies balanced pressure to the substrate.

[0108] Current density modulation signal: Based on the thermal resistance distribution map (which shows the change in the impedance of the heat conduction path), the duty cycle of the pulse width modulation (PWM) signal is adjusted to control the heating power of the conductive layer to match the thermal conductivity of the substrate.

[0109] Heat flow coordinated control module:

[0110] Bayesian optimization unit: Based on the conductive layer power data (current × voltage) and temperature field distribution data (collected by an infrared thermal imager), the equilibrium threshold of the heat conduction medium flow rate is calculated through a probabilistic model to balance the heat conduction efficiency and interface thermal stress.

[0111] Composite control execution unit: adopts feedforward and feedback composite control strategy. The feedforward module predicts the proportional valve opening and variable frequency pump speed according to the flow rate threshold. The feedback module corrects the flow rate deviation through the measured data of the thermal flow sensor to achieve dynamic adjustment of the medium flow rate.

[0112] Interface stress relief module:

[0113] Acoustic emission signal: The acoustic emission sensor collects high-frequency elastic wave signals (such as crack propagation sound waves) during the lamination process, and identifies the location and propagation trend of interface cracks through spectrum analysis.

[0114] Stress distribution images: Digital image correlation (DIC) is used to analyze the displacement field of speckle marks on the substrate surface and generate 2D / 3D distribution maps of stress concentration areas.

[0115] Dynamic compensation unit: According to the stress gradient direction (vector) and deformation data (strain sensor feedback), it drives the piezoelectric ceramic actuator (nanoscale fine-tuning) and the pneumatic fine-tuning mechanism (macro pressure compensation) to eliminate residual stress accumulation.

[0116] Digital Twin Optimization Module:

[0117] Thermodynamics simulation engine: Based on the thermal resistance, stress gradient, and current density parameters in the characteristic matrix, a multi-physics field coupling model of the virtual production environment is constructed to simulate the evolution of the heat conduction path and predict the thermal resistance deviation trend.

[0118] Multi-objective optimization unit: A genetic algorithm (simulating biological evolutionary mechanisms) is used to iteratively optimize the control weight coefficients to collaboratively improve thermal resistance uniformity (minimizing standard deviation), stress threshold (interface strength constraint) and energy efficiency (power / heat output ratio).

[0119] Data synchronization and protocol conversion unit:

[0120] Timestamp alignment: Normalizes the acquisition timestamps of multi-source sensor data through a global clock signal to eliminate timing misalignment caused by sampling frequency differences.

[0121] Cubic spline interpolation: reconstruct continuous curve for non-uniform sampling data, fill in data gaps caused by communication delay or packet loss.

[0122] OPC-UA protocol conversion: convert deformation, energy, and electrical data into standardized structured data formats (such as three-dimensional coordinate point set, temperature-heat flow matrix), and realize real-time interaction of cross-module data through industrial Ethernet.

[0123] The physical field coupling model establishes the mapping relationship between thermal resistance distribution, stress gradient, and current density by associating deformation data (strain sensors and laser profilometer acquisition), energy transfer data (infrared thermal imager and heat flow sensor acquisition), and electrical signal data (current transformer and voltage probe monitoring). This model uses principal component analysis to reduce the dimensionality of multi-source heterogeneous data, extracts key correlation parameters across physical fields, and generates a characteristic matrix representing the synergistic action of heat, power, and electricity, providing unified data input for subsequent control modules.

[0124] The Bayesian optimization algorithm is applied to the heat flow collaborative control module, based on the power data of the conductive layer (current x voltage) and the temperature field distribution data in the feature matrix (infrared thermal imager acquisition), a probability model of heat conduction medium flow rate and heat flow path is constructed, and the optimal flow rate threshold is solved by Markov chain Monte Carlo sampling, balancing heat conduction efficiency and interface thermal stress accumulation.

[0125] Genetic algorithm realizes multi-objective collaborative optimization in digital twin optimization module, simulates biological evolution mechanism to iteratively select control weight coefficients, and fitness function integrates thermal resistance uniformity (standard deviation minimization), stress threshold (Mises stress constraint), and energy efficiency (power / heat output ratio). Finally, the optimized weight coefficients are returned to the control module to dynamically adjust the temperature control threshold, pressure gradient, and duty cycle parameters.

[0126] Digital image correlation technology (DIC) is used in the interface stress dissipation module to analyze the displacement field changes of the substrate surface speckle markers, generate stress distribution images, and quantify the coordinates and gradient directions of stress concentration areas. Combined with acoustic emission signals (crack propagation characteristic frequency), an interface stress vector field model is constructed to provide data support for dynamic compensation. The above models realize cross-domain collaborative control of production parameters through data correlation and algorithm nesting, solving the technical problem of unstable heat conduction efficiency in traditional segmented control.

[0127] The specific implementation of the panel production control system with heating function of the present invention is as follows: the cross-domain data fusion module collects deformation data, energy transfer data, electrical signal data, acoustic emission signals and stress distribution images in real time through a distributed sensor network. The deformation data is synchronously acquired by a high-precision strain sensor and a laser profiler, wherein the strain sensor monitors the microscopic deformation of the substrate and the laser profiler scans the macroscopic contour fluctuation of the surface; the energy transfer data is captured by an infrared thermal imager to capture the surface temperature field distribution of the substrate, and the heat flux sensor measures the heat flux density value; the electrical signal data is collected by a current transformer and a voltage probe to collect the current carrying parameters of the conductive layer. The above-mentioned multi-source data is timestamp aligned and processed by cubic spline interpolation through a data synchronization unit, and input into the data processing unit after eliminating the timing deviation. Based on the physical field coupling model, the data processing unit extracts thermal resistance distribution, stress gradient and current density as key features through principal component analysis, generates a thermo-electric synergistic effect characteristic matrix, and stores it in a distributed time series database.

[0128] After receiving the characteristic matrix, the multi-parameter coupling control module divides the temperature control zone according to the thermal resistance distribution parameters, generates heating power compensation instructions for each zone, and adjusts the resistance wire power through the zone temperature control module. The pressure distribution control unit calculates the pressure setpoint of the hydraulic servo system based on real-time feedback signals from the stress gradient parameters and deformation data, driving the laminating roller to apply balanced pressure. The current density modulation unit analyzes the thermal resistance distribution map and dynamically adjusts the duty cycle of the pulse width modulation signal to match the heating power of the conductive layer with the thermal conductivity of the substrate. Control instructions are simultaneously transmitted to the execution terminal via industrial Ethernet, achieving coordinated regulation of temperature, pressure, and current parameters.

[0129] The heat flow coordinated control module combines the power data of the conductive layer with the temperature field distribution in the characteristic matrix and calculates the flow velocity equilibrium threshold of the heat transfer medium using a Bayesian optimization algorithm. The feedforward control module predicts the flow velocity demand based on the threshold, while the feedback control module corrects for deviations using measured data from the heat flow sensor. The proportional valve and variable frequency pump unit coordinately regulate the medium flow. The interface stress reduction module analyzes acoustic emission signals and stress distribution images. The acoustic emission sensor identifies interlaminar crack characteristics, and digital image correlation technology quantifies the coordinates of stress concentration areas. The dynamic compensation unit drives the piezoelectric ceramic actuator and pneumatic mechanism to asymmetricly compensate for the lamination roller pressure. The digital twin optimization module constructs a thermo-mechanical-electric coupled virtual model based on the characteristic matrix. Using a genetic algorithm, it optimizes the weighting coefficients for thermal resistance uniformity, stress threshold, and energy efficiency. This is fed back to the multi-parameter coupled control module to update the temperature control threshold, pressure gradient, and duty cycle parameters, forming a closed-loop iterative control system. The data synchronization unit and protocol conversion unit convert multi-source data into the OPC-UA protocol format and transmit it via Industrial Ethernet, ensuring real-time data exchange across modules, ultimately achieving dynamic coordination between heat transfer path stability and production parameters.

[0130] The application collects deformation data, energy transmission data and electrical signal data in real time through a cross-domain data fusion module, performs time-space alignment and noise reduction processing on multi-source heterogeneous data based on a physical field coupling model, and generates a feature matrix containing thermal resistance distribution, stress gradient and current density mapping relationship. The matrix extracts key physical field correlation parameters through principal component analysis, provides a unified data benchmark for dynamic coordination of multi-level production parameters, and solves the problem of missing parameter correlation caused by data island in traditional segmented control. The output end of the cross-domain data fusion module directly communicates with the input end of the multi-parameter coupling control module to realize synchronous analysis and instruction generation of thermal power parameters.

[0131] After receiving the feature matrix, the multi-parameter coupling control module generates a partition temperature control compensation instruction based on the thermal resistance distribution to suppress local overheating, adjusts the lamination roller pressure output based on the stress gradient parameter and real-time deformation feedback signal, and dynamically adjusts the duty cycle of the pulse width modulation signal according to the current density mapping relationship. The temperature gradient adjustment signal, pressure control signal and current density modulation signal are synchronously issued to the execution terminal through industrial Ethernet to form a cooperative control link of heat conduction path regulation, interface stress dissipation and conductive layer power matching, eliminating the thermal field distortion caused by parameter response lag in segmented control.

[0132] The digital twin optimization module constructs a thermal power coupling virtual model, performs multi-objective optimization on thermal resistance uniformity, stress threshold and energy efficiency through a genetic algorithm, generates control weight coefficients and returns them to the multi-parameter coupling control module. The optimized weight coefficients dynamically update the temperature control threshold, pressure gradient and duty cycle parameters, combined with the real-time compensation data of the interface stress dissipation module, form a closed-loop feedback mechanism from physical entity perception, multi-parameter cooperative control to virtual model iterative optimization, realizing the dynamic balance of heat conduction path and structural stability.

Claims

1. A plate production control system with heating function, characterized in that: include: A cross-domain data fusion module is configured to collect deformation data, energy transfer data, electrical signal data, acoustic emission signals, and stress distribution images, and pre-process the deformation data, energy transfer data, electrical signal data, acoustic emission signals, and stress distribution images based on a preset physical field coupling model to generate a feature matrix including thermal resistance distribution, stress gradient, and current density mapping relationships; a multi-parameter coupling control module configured to receive a characteristic matrix, generate a temperature gradient adjustment signal according to a thermal resistance distribution, generate a pressure distribution control signal according to a stress gradient, generate a current density modulation signal according to a current density, and synchronously output the temperature gradient adjustment signal, the pressure distribution control signal, and the current density modulation signal; A heat flow collaborative control module is configured to receive the temperature gradient adjustment signal and the current density modulation signal, combine the conductive layer power data extracted from the electrical signal data in the cross-domain data fusion module with the substrate temperature field characteristics generated by the energy transfer data in the characteristic matrix, and generate the heat transfer medium flow rate control parameters through a preset Bayesian optimization algorithm; An interface stress elimination module is configured to receive pressure distribution control signals and heat transfer medium flow rate control parameters, analyze acoustic emission signals and stress distribution images in the cross-domain data fusion module, generate asymmetric compensation instructions, and drive the distributed actuator array to adjust the lamination roller pressure; The digital twin optimization module is configured to receive the characteristic matrix of the cross-domain data fusion module as physical entity data, generate virtual simulation data corresponding to the thermal resistance distribution, stress gradient and current density through the built-in thermodynamic simulation engine, and optimize the weight coefficients of the thermal resistance distribution, stress gradient and current density through the preset multi-objective optimization algorithm, and return the optimized weight coefficients to the multi-parameter coupling control module.

2. The plate production control system with heating function according to claim 1 is characterized in that: Also includes: The output end of the cross-domain data fusion module transmits the characteristic matrix to the multi-parameter coupling control module. The input end of the heat flow collaborative control module receives the temperature gradient adjustment signal and current density modulation signal of the multi-parameter coupling control module. The input end of the interface stress elimination module receives the heat transfer medium flow rate control parameter of the heat flow collaborative control module. The output end of the digital twin optimization module feeds back the optimized weight coefficient to the multi-parameter coupling control module to update the temperature gradient adjustment signal, pressure distribution control signal and current density modulation signal, forming a closed-loop control link. In the closed-loop control link, the cross-domain data fusion module transmits the thermal resistance distribution, stress gradient and current density parameters in the characteristic matrix to the multi-parameter coupling control module; the optimized weight coefficient fed back by the digital twin optimization module is used to update the temperature control threshold in the temperature gradient adjustment signal, the pressure set value in the pressure distribution control signal and the duty cycle parameter in the current density modulation signal.

3. The plate production control system with heating function according to claim 2 is characterized in that: The cross-domain data fusion module includes: a deformation data acquisition unit configured to acquire deformation data of the substrate through a high-precision strain sensor and a laser profilometer; an energy data acquisition unit configured to acquire temperature field distribution data through an infrared thermal imager and a heat flow sensor; an electrical data acquisition unit configured to monitor current-carrying parameters of the conductive layer via a current transformer and a voltage probe; The data processing unit is configured to perform timestamp alignment and cubic spline interpolation processing on the data output by the deformation data acquisition unit, the energy data acquisition unit and the electrical data acquisition unit, and extract the thermal resistance distribution, stress gradient and current density as extraction features through principal component analysis, and generate a thermo-electric synergy feature matrix through the extracted features obtained by the principal component analysis.

4. The plate production control system with heating function according to claim 3 is characterized in that: The multi-parameter coupling control module includes: a temperature gradient adjustment unit configured to generate a heating power compensation instruction for each temperature control zone according to the thermal resistance distribution parameter in the characteristic matrix; a pressure distribution control unit configured to adjust the pressure output of the hydraulic servo system based on the stress gradient parameters in the characteristic matrix and the real-time feedback signal from the deformation data acquisition unit; The current density modulation unit is configured to dynamically adjust the pulse width modulation signal according to the thermal resistance distribution map.

5. The plate production control system with heating function according to claim 4, characterized in that: The heat flow coordinated control module includes: a Bayesian optimization unit configured to calculate a flow rate equilibrium threshold of the heat transfer medium based on the conductive layer power data received by the heat flow coordinated control module and energy transfer data in the characteristic matrix, wherein the energy transfer data includes a surface temperature field distribution of the substrate collected by the infrared thermal imager; The compound control execution unit is configured to perform feedforward and feedback compound control on the medium flow rate through the proportional valve and the variable frequency pump group, wherein the feedforward module predicts the flow rate demand based on the flow rate balance threshold, and the feedback module corrects the deviation according to the thermal flow sensor data.

6. The plate production control system with heating function according to claim 5, characterized in that: The interface stress relief module includes: A stress detection unit is configured to analyze the acoustic emission signal and stress distribution image in the cross-domain data fusion module, wherein the acoustic emission signal is collected by the acoustic emission sensor to generate the interlayer crack signal, and the stress distribution image is generated by analyzing the substrate surface displacement field using a preset digital image analysis model; The dynamic compensation unit is configured to drive the piezoelectric ceramic actuator and the pneumatic fine-tuning mechanism based on the stress gradient direction and deformation data.

7. The plate production control system with heating function according to claim 6, characterized in that: The digital twin optimization module includes: a thermodynamic simulation engine configured to construct a coupled model of the virtual production environment based on the thermal resistance distribution, stress gradient, and current density parameters in the characteristic matrix and to predict the deviation trend of the thermal resistance distribution relative to a preset diffusion model; The multi-objective optimization unit is configured to collaboratively optimize the thermal resistance uniformity, stress threshold and energy efficiency of the control weight coefficient through a genetic algorithm.

8. The plate production control system with heating function according to claim 7, characterized in that: Also includes: a data synchronization unit configured to eliminate timing deviations of multi-source data through timestamp alignment and cubic spline interpolation algorithm; The protocol conversion unit is configured to convert the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image synchronization unit processed deformation data, energy transfer data and electrical signal data into the OPC-UA protocol data format supported by the cross-domain data fusion module, and transmit the data to the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image processing unit via industrial Ethernet. The output end of the deformation data, energy transfer data, electrical signal data, acoustic emission signal and stress distribution image synchronization unit is connected to the input end of the protocol conversion unit, and the output end of the protocol conversion unit is communicatively connected to the input end of the cross-domain data fusion module.

Images