A method and system for controlling the composite molding of car trunk carpets

Abstract

This invention discloses a composite molding control method and system for automotive trunk carpets, belonging to the field of intelligent manufacturing technology. The method includes: acquiring multi-layer carpet material to obtain a composite material to be molded; collecting multi-layer hot-pressing related characteristic information, generating a pressure-temperature field control spectrum and dynamic surface contact state of key surfaces that vary with time and space through digital twin simulation; controlling the hot press to perform molding operations using the pressure-temperature field control spectrum, activating a flexible thin-film sensor array to monitor the pressure distribution, material contact state, and micro-vibration of key surfaces in real time, and generating synchronous surface contact state data; comparing the synchronous surface contact state data with the dynamic surface contact state, calculating the state difference vector field, and performing dynamic compensation. This application solves the technical problem in the prior art where the hot-pressing molding parameters for automotive trunk carpets are fixed, making it difficult to accurately control key surfaces, resulting in numerous defects such as wrinkles and cracks, and poor batch consistency.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, specifically to a composite molding control method and system for automotive trunk carpets. Background Technology

[0002] As the automotive industry upgrades towards lightweight and high-quality designs, the molding precision and structural stability of car trunk carpets, as a core interior component, directly affect the overall vehicle's feel. Composite molding technology, which balances material lightweighting and structural strength, has become the mainstream manufacturing method for trunk carpets, leading to an increasingly urgent need for precise control over the molding process within the industry.

[0003] Currently, automotive trunk carpet composite molding mostly adopts traditional hot pressing processes, relying on experience to set pressure-temperature parameters, lacking dynamic adaptation to the characteristics of multi-layer materials. During the molding process of key surfaces, it is impossible to monitor the contact state and deviation in real time, making it difficult to achieve precise compensation. This leads to defects such as wrinkles and poor adhesion in the product, hindering the improvement of molding quality. Summary of the Invention

[0004] This application provides a composite molding control method and system for automotive trunk carpets, aiming to solve the technical problems in the prior art where the hot pressing molding parameters of automotive trunk carpets are fixed, making it difficult to accurately control key surfaces, resulting in numerous defects such as wrinkles and cracks, and poor batch consistency of products.

[0005] In view of the above problems, this application provides a composite molding control method and system for automotive trunk carpets.

[0006] The first aspect disclosed in this application provides a composite molding control method for automotive trunk carpets. The method includes: acquiring multi-layer carpet material for composite molding; placing the stacked multi-layer carpet material between the upper and lower molds of a hot press to obtain a composite material to be molded; collecting multi-layer hot-press-related characteristic information of the composite material to be molded; generating a pressure-temperature field control spectrum and dynamic surface contact state of key surfaces that vary with time and space through digital twin simulation; controlling the hot press to perform molding operations using the pressure-temperature field control spectrum; synchronously activating a flexible thin-film sensor array to monitor the pressure distribution, material contact state, and micro-vibration of key surfaces in real time, generating synchronous surface contact state data; comparing the synchronous surface contact state data with the dynamic surface contact state, calculating the state difference vector field, and driving a digital hydraulic unit embedded in the upper and lower mold surfaces to perform dynamic compensation.

[0007] Another aspect of this application discloses a composite molding control system for automotive trunk carpets. The system includes: a composite material acquisition module for acquiring multi-layer carpet material for composite molding, placing the stacked multi-layer carpet material between the upper and lower molds of a hot press to obtain the composite material to be molded; a digital twin simulation module for collecting multi-layer hot-press related characteristic information of the composite material to be molded, generating a pressure-temperature field control spectrum that varies with time and space and a dynamic surface contact state of key surfaces through digital twin simulation; a state data generation module for controlling the hot press to perform molding operations using the pressure-temperature field control spectrum, synchronously activating a flexible thin-film sensor array to monitor the pressure distribution, material contact state, and micro-vibration of key surfaces in real time, and generating synchronous surface contact state data; and a dynamic compensation module for comparing the synchronous surface contact state data with the dynamic surface contact state, calculating the state difference vector field, and driving a digital hydraulic unit embedded in the upper and lower mold surfaces to perform dynamic compensation.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: They adopt a closed-loop technical solution that uses digital twin simulation to generate pressure-temperature field control maps, flexible thin-film sensor arrays to monitor the key surface status in real time, and digital hydraulic units to dynamically compensate for deviations. This solves the technical problems of fixed hot pressing molding parameters and low precision in key surface control of existing automotive trunk carpets, which leads to frequent defects such as wrinkles and cracks and poor batch consistency. It achieves the technical effects of precise control of molding status, significant reduction of defect rate, and improvement of surface fit and batch stability.

[0009] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0010] Figure 1 This application provides a schematic flowchart of a composite molding control method for automotive trunk carpets. Figure 2 This application provides a flowchart illustrating the dynamic compensation process in a composite molding control method for automotive trunk carpets, as described in an embodiment of the present application. Figure 3 This application provides a schematic diagram of the structure of a composite molding control system for an automotive trunk carpet.

[0011] Figure labeling: Composite material acquisition module 11, digital twin simulation module 12, state data generation module 13, dynamic compensation module 14. Detailed Implementation

[0012] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0013] The overall concept of the technical solution provided in this application is as follows: This application provides a composite molding control method and system for automotive trunk carpets. It integrates digital twin simulation, flexible sensing monitoring, and digital hydraulic compensation technology. First, a pressure-temperature field control spectrum is generated through simulation. Then, the state of key surfaces is monitored in real time. The differences between simulation and measured data are calculated, and the actuator is driven to dynamically compensate, achieving precise molding of the automotive trunk carpet.

[0014] After introducing the basic principles of this application, various non-limiting embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0015] Example 1, as Figure 1 As shown in the figure, this application provides a method for controlling the composite molding of automotive trunk carpets, the method comprising: Step S100: Obtain multi-layer carpet material for composite molding, place the stacked multi-layer carpet material between the upper and lower molds of a hot press to obtain the composite material to be molded.

[0016] Specifically, multi-layered carpet materials refer to a combination of multiple substrates with different functions adapted to the composite molding requirements of automotive trunk carpets. Each layer of material plays a specific role, such as load-bearing, cushioning, and sound insulation, and must meet the bonding compatibility requirements during the hot pressing process. For example, a common combination for mid-to-high-end cars is: PET needle-punched cotton (cushioning and sound insulation layer) + PP honeycomb core (structural load-bearing layer) + thermally rolled nonwoven fabric (surface molding layer) + EVA film (interlayer bonding layer); a combination for economy cars is: recycled cotton (sound insulation layer) + corrugated paper core (lightweight load-bearing layer) + PE coated nonwoven fabric (waterproof surface layer). The upper and lower molds of the hot press refer to the core mold components in the hot press that realize hot pressing molding. The upper mold is usually a punch, and the lower mold is a die. The cavity contour is perfectly matched with the surface of the target automotive trunk carpet, and the internal heating unit and subsequent dynamic compensation digital hydraulic unit mounting position are integrated.

[0017] Specifically, suitable multi-layer carpet materials are sourced through supply chain procurement or customization. For example, a four-layer combination of "thermal rolled nonwoven fabric + EVA film + PP honeycomb core + PET needle-punched cotton" is used for trunk carpets in mid-to-high-end SUVs. After acquisition, each layer of material is inspected using a thickness gauge and a weight balance. According to the functional sequence of the product design drawings, an automated stacking machine is used for precise stacking. Taking "thermal rolled nonwoven fabric (upper layer) → EVA film → PP honeycomb core → PET needle-punched cotton (lower layer)" as an example, the interlayer alignment is corrected in real time through a visual positioning system during the stacking process. For small-batch trial production, manual stacking aids such as positioning fixtures with carpet outlines can be used to ensure alignment. The stacked multi-layered material is then transferred to the upper and lower molds of a single-station CNC hot press using a robotic arm or manual transfer device. The upper and lower molds are custom-designed arc-shaped molds based on the shape of the luggage carpet (the convex surface of the upper mold matches the inner surface of the carpet, and the concave surface of the lower mold matches the outer surface). During placement, ensure that the multi-layered material completely covers the mold cavity and that the edges are aligned with the positioning pins. This results in a composite material with a complete structure and precise positioning.

[0018] This step, as a preparatory stage for composite molding, provides a consistent and controllable input foundation for subsequent digital twin simulation and hot pressing through standardized multi-layer material screening and inspection, precise stacking and positioning, and standardized mold adaptation. It effectively improves the batch consistency of the composite materials to be molded, laying the foundation for the accuracy of subsequent real-time monitoring and dynamic compensation, ultimately helping to improve the molding pass rate and surface accuracy stability of automotive trunk carpets.

[0019] Step S200: Collect multilayer hot-press related characteristic information of the composite material to be molded, and generate a pressure-temperature field control spectrum and dynamic surface contact state of key surfaces that vary with time and space through digital twin simulation.

[0020] Specifically, multilayer hot-press related characteristic information refers to the set of physicochemical properties and structural parameters of multilayer materials directly related to the hot-press bonding and molding deformation of the composite material to be molded, and serves as the input data for digital twin simulation. Digital twin simulation is a technology that constructs a virtual simulation model that maps 1:1 to the actual hot-pressing molding scenario based on the physical properties of the composite material to be molded and the hardware parameters of the hot press. It achieves process parameter prediction and state forecasting by simulating the entire hot-pressing process. The pressure-temperature field control map refers to the distribution map of optimal pressure and temperature parameters in various regions of the mold surface as hot-pressing time progresses and the mold's spatial position changes, containing both time series and spatial coordinate dimensions.

[0021] Specifically, detection equipment is used to collect hot-pressing related feature information of each layer of the composite material to be molded. The collected multi-dimensional data is normalized to generate a multi-layer hot-pressing related feature information dataset. This dataset is imported into a pre-constructed hot-pressing digital twin model to simulate the entire process from mold closing to pressure holding. The optimal pressure and temperature range of each simulation grid node on the mold surface at different time points is calculated. Combined with the control domain division of the heating / pressurizing unit of the hot press, the parameters of adjacent grid nodes are aggregated and optimized to generate a pressure-temperature field control spectrum that varies with time and space. In the digital twin model, key surfaces, such as the curved corners and edge protrusions of a luggage carpet, are focused on. Parameters such as the surface pressure distribution, material-mold contact gap, and contact area ratio at different time points during the simulation are extracted to form dynamic surface contact state data of key surfaces.

[0022] This step provides high-precision, standardized input data for digital twin simulation by accurately collecting hot-pressing-related characteristic information of multi-layer materials, avoiding simulation distortion caused by missing or incorrect material parameters. The pressure-temperature field control map generated by digital twin technology overcomes the limitations of traditional hot-pressing processes that rely on empirical parameters, achieving precise spatiotemporal control of process parameters. Simultaneously, the advance prediction of the dynamic contact state of key surfaces can identify potential problems such as poor contact and uneven pressure before actual hot-pressing, providing a clear reference benchmark for subsequent real-time monitoring and dynamic compensation. This effectively reduces the incidence of defects such as wrinkles and cracks on key surfaces during the molding process, improving the molding accuracy and batch consistency of automotive trunk carpets.

[0023] Step S300: Control the hot press to perform molding operation using the pressure-temperature field control spectrum, simultaneously activate the flexible thin film sensor array, monitor the pressure distribution, material contact state and micro-vibration of key surfaces in real time, and generate synchronous surface contact state data.

[0024] Furthermore, the flexible thin-film sensing array is located on the inner wall of the mold surface of the upper and lower molds, and includes a pressure sensing unit, a capacitive proximity sensing unit, and a piezoelectric vibration sensing unit.

[0025] Specifically, the flexible thin-film sensor array is an integrated sensing component made of flexible substrate materials, such as polyimide film, that can be fitted to the inner wall of a mold surface and has the ability to synchronously acquire multi-dimensional data. The pressure sensing unit is one of the core modules of the flexible thin-film sensor array. Based on the piezoresistive effect, it realizes real-time acquisition of pressure distribution on the mold surface and outputs pressure values ​​and distribution gradient data. The capacitive proximity sensing unit is a module that determines the contact state between the material and the mold surface by detecting the capacitance change between the sensor and the composite material surface, and can acquire parameters such as contact gap and contact area ratio. The piezoelectric vibration sensing unit is a module that acquires micro-vibration signals of the mold surface during hot pressing based on the piezoelectric effect, and can capture the minute vibration frequencies and amplitudes generated during material melting and bonding. Synchronous surface contact state data refers to a real-time dataset synchronously acquired by the flexible thin-film sensor array, containing three types of data: key surface pressure distribution, material contact state, and micro-vibration, with dual identification of timestamps and spatial coordinates.

[0026] Specifically, the pressure-temperature field control graph is imported into the PLC control system of the hot press. Based on the time series and spatial partition parameters of the graph, the PLC sends control commands to the heating unit and hydraulic pressurization unit of the hot press, driving the hot press to perform molding operations such as mold closing, heating, and pressure holding. For example, during the 0-10s stage of mold closing, the hydraulic unit in the edge area is controlled to output 1.0MPa. The pressure and heating unit raise the temperature to 65°C. Simultaneously, the flexible thin-film sensor array embedded in the inner wall of the upper and lower model surfaces is activated via a data acquisition card. This array integrates a pressure sensing unit, a capacitive proximity sensing unit, and a piezoelectric vibration sensing unit, which can simultaneously monitor the pressure distribution, material contact gap and contact area ratio, and micro-vibration frequency and amplitude of key surfaces. For example, the pressure sensing unit provides real-time feedback on pressure fluctuations in the corner area, the capacitive proximity sensing unit identifies changes in the contact state between the material and the mold, and the piezoelectric vibration sensing unit captures the characteristic vibration signals generated by the melting of the adhesive film. Finally, the data processing module timestamps and matches the spatial coordinates of the three types of monitoring data to generate synchronous surface contact state data containing multi-dimensional information on pressure, contact state, and micro-vibration.

[0027] This step enables the simultaneous execution of hot pressing molding and multi-dimensional state monitoring. Through the precise acquisition of data from a flexible thin-film sensor array, real-time state data of key surfaces is obtained. This not only provides high-precision data support for subsequent comparison of digital twins to predict states and calculate state difference vector fields, but also captures abnormal states in the molding process in real time, laying the foundation for timely triggering of dynamic compensation. This effectively improves the process controllability and final product quality of automotive trunk carpet composite molding.

[0028] Step S400: Compare the synchronous surface contact state data with the dynamic surface contact state, calculate the state difference vector field, and drive the digital hydraulic unit embedded in the upper and lower model surfaces to perform dynamic compensation.

[0029] Furthermore, the digital hydraulic unit includes a micro piezoelectric actuator and a magnetostrictive actuator, wherein the micro piezoelectric actuator corresponds to a damping compensation mode and the magnetostrictive actuator corresponds to a displacement compensation mode.

[0030] Specifically, the state difference vector field refers to the spatial distribution and temporal evolution of parameter differences calculated by comparing synchronous surface contact state data with dynamic surface contact state data according to timestamps and spatial grid nodes. Each grid node corresponds to a three-dimensional vector containing pressure difference, contact state difference, and micro-vibration difference. The digital hydraulic unit is a miniature hydraulic control component pre-embedded in key areas of the upper and lower model surfaces. It integrates two core actuators: a micro-piezoelectric actuator and a magnetostrictive actuator. It can perform precise compensation actions based on the commands output by the state difference vector field. The micro-piezoelectric actuator is an actuator that operates based on the piezoelectric effect, generating minute damping forces through electrical signals. Corresponding to a damping compensation mode, it is used to correct surface contact instability caused by excessive micro-vibration during the molding process. The magnetostrictive actuator is an actuator that operates based on the magnetostrictive effect, generating nanometer-level displacement through magnetic field changes. Corresponding to a displacement compensation mode, it is used to correct poor fit caused by excessive surface contact gap or insufficient pressure during the molding process. Damping compensation mode is a method for compensating for micro-vibration deviations on critical surfaces. It uses a miniature piezoelectric actuator to adjust the damping force, suppressing abnormal vibrations between the material and the mold, and stabilizing the contact state of the surfaces. Displacement compensation mode is a method for compensating for pressure and contact gap deviations on critical surfaces. It uses a magnetostrictive actuator to make minute displacement adjustments, changing the local pressure distance on the mold surface, improving the uniformity of pressure distribution and the proportion of contact area.

[0031] Specifically, synchronous surface contact state data and dynamic surface contact state data are matched according to timestamps and spatial grid nodes to ensure consistent comparison dimensions between the two types of data. A global state difference vector field is constructed using pressure difference, contact state difference, and micro-vibration difference as components for each grid node, and key surface regions corresponding to abnormal vectors are selected through preset difference thresholds. The corresponding compensation mode is matched according to the type of abnormal vector: the damping compensation mode is matched for regions dominated by micro-vibration deviation, and the displacement compensation mode is matched for regions dominated by pressure or contact gap deviation. The PLC control system sends instructions to the digital hydraulic unit embedded in the abnormal region to drive the corresponding actuator to perform compensation actions, activate the micro piezoelectric actuator to output damping force to suppress vibration, and activate the magnetostrictive actuator to extend and adjust the surface displacement, thereby completing the dynamic compensation of the key surface.

[0032] This step constructs a state difference vector field to realize the quantitative analysis of the forming state deviation of the key surface. Combined with the dual-mode compensation mechanism of the digital hydraulic unit, it breaks through the technical bottleneck of traditional hot pressing forming of "fixed parameters and inability to adjust in real time" and can adaptively correct dynamic deviations in the forming process.

[0033] Furthermore, the multilayer hot-pressing related feature information of the composite material to be molded is collected, including: constructing a hot-pressing related attribute set, including several material properties related to the hot-pressing process; collecting the multilayer material property values ​​corresponding to the composite material to be molded based on the hot-pressing related attribute set and performing normalization processing to generate the multilayer hot-pressing related feature information.

[0034] Specifically, the hot-pressing-related attribute set refers to the collection of physicochemical properties of materials directly related to the hot-pressing composite molding process of multi-layer carpet materials. It needs to cover the key dimensions affecting material adhesion, deformation, and curing during the hot-pressing process. For example, for the four-layer material commonly used in automotive trunk carpets—"thermal rolled nonwoven fabric + EVA film + PP honeycomb core + PET needle-punched cotton"—the constructed hot-pressing-related attribute set includes three main categories: thermal properties (melting point, thermal conductivity, coefficient of thermal expansion), mechanical properties (compressive strength, elastic modulus, Poisson's ratio), and adhesive properties (melt viscosity, adhesive strength threshold). Multi-layer material attribute values ​​refer to the specific parameter values ​​of the hot-pressing-related attribute set corresponding to each layer of the composite material to be molded. The attribute values ​​of different layers differ significantly. Normalization refers to mapping multi-layer material attribute values ​​with different dimensions and numerical ranges to a unified numerical range, such as a standardized data processing method for the 0-1 range, eliminating the interference of dimensional differences on subsequent digital twin simulations. Multi-layer hot-pressing-related feature information: refers to the standardized dataset of multi-layer material hot-pressing attributes formed after normalization processing, which is the core input parameter for digital twin simulations. Example: A feature information table generated for a four-layer carpet material includes the normalized thermal, mechanical, and adhesive properties of each layer, as well as auxiliary parameters such as the initial gap between layers and the thickness ratio.

[0035] Specifically, based on the hot-pressing requirements of multi-layer composite materials for automotive trunk carpets, a set of hot-pressing-related attributes covering thermal, mechanical, and adhesive dimensions is constructed by combining material properties. For example, for a four-layer material consisting of "thermal-rolled nonwoven fabric + EVA film + PP honeycomb core + PET needle-punched cotton," the attribute set is defined to include multiple core attributes such as melting point, thermal conductivity, compressive strength, and melt viscosity. For each layer of the composite material to be molded, corresponding testing equipment is used to collect attribute values. For example, a differential scanning calorimeter is used to determine the melting point of the EVA film, an electronic universal testing machine is used to determine the compressive strength of the PP honeycomb core, and a rotational viscometer is used to determine the melt viscosity of the EVA film. At the same time, structural parameters such as the thickness and interlayer gap of each layer are recorded. Then, a linear normalization method is used to map attribute values ​​of different dimensions to the 0-1 range, eliminating dimensional interference between different attributes. The normalized attribute values ​​of each layer are integrated with the structural parameters to generate standardized multi-layer hot-pressing-related feature information, providing unified and accurate input data for subsequent digital twin simulation.

[0036] This step, by constructing a targeted set of hot-pressing-related attributes, achieves comprehensive coverage of the factors influencing the hot pressing of multi-layer carpet materials, avoiding simulation deviations caused by the lack of key attributes. Relying on precise attribute values ​​collected by professional testing equipment, combined with normalization processing to eliminate dimensional differences, the generated multi-layer hot-pressing-related feature information has standardized and quantifiable characteristics, which can directly drive high-precision simulation of the digital twin model. At the same time, the layered acquisition and normalization processing method can accurately match the characteristic differences of different layers of materials, laying a data foundation for the subsequent generation of spatiotemporally differentiated pressure-temperature field control maps, effectively improving the fit between the digital twin simulation results and the actual molding process.

[0037] Furthermore, through digital twin simulation, a pressure-temperature field control map and dynamic surface contact state of key surfaces that vary with time and space are generated. This includes: importing the multi-layer hot-press related feature information into a pre-constructed hot-press digital twin model, running the simulation, and calculating the optimal temperature setting range and optimal pressure setting range at each simulation grid node on the mold surface during the entire time series from mold closing to pressure holding, thus forming a pressure-temperature field control range distribution; collecting the distribution of the heating unit and hydraulic pressurization unit of the hot press, and combining the pressure-temperature field control range distribution with the execution of control partitioning and pressure-temperature field aggregation to generate the pressure-temperature field control map; inputting the pressure-temperature field control map into the hot-press digital twin model for key surface simulation to obtain the dynamic surface contact state, including surface pressure distribution, surface material contact state, and surface micro-vibration.

[0038] Furthermore, the key surface is a local region in the mold corresponding to the upper and lower molds where the geometry has abrupt changes in depth, curvature, or cross-sectional area in three-dimensional space.

[0039] Specifically, the hot-press digital twin model is a 1:1 virtual mapping model built based on a real hot-press molding scenario. It integrates elements such as mold structure, material properties, and hot press hardware parameters, enabling dynamic simulation of the entire hot-pressing process. Specifically, the hot-press digital twin model's structure includes four core layers: a physical entity layer, a virtual mapping layer, a data interaction layer, and a simulation analysis layer. Its construction method involves: collecting the hardware parameters of the hot press, including: heating unit power / distribution location, hydraulic pressurization unit pressure range / response speed; the three-dimensional structural data of the mold, including upper and lower model cavity dimensions, key surface curvature / depth abrupt change parameters, pre-embedded positions of digital hydraulic units, and multi-layered physicochemical properties of the composite material to be molded; completing a 1:1 geometric model of the hot press and mold using SolidWorks; defining the thermo-structural coupling constitutive equation of the composite material using ANSYS Material Designer; building the geometric model, physical property model, and molding behavior model of the virtual mapping layer on the ANSYS Workbench platform; and then developing a model based on OPC. The UA protocol's data interaction layer interface enables bidirectional transmission of real-time data such as pressure, temperature, and vibration from the flexible thin-film sensor array to the virtual model. Simultaneously, the simulation analysis layer embeds control partitioning algorithms, pressure-temperature field interval intersection extraction algorithms, and key surface dynamic contact state discrimination algorithms. By inputting multi-layer hot-pressing related feature information, the simulation runs, outputting pressure-temperature field control maps and dynamic contact state prediction data. Multiple sets of gradient hot-pressing experiments are conducted, comparing the real molding data collected from physical experiments with the virtual simulation data. The model's mesh node accuracy and material property thresholds, such as the EVA film melt viscosity correction coefficient and kinetic equation parameters, are iteratively corrected. Ultimately, a hot-pressing digital twin model with full element mapping to the real hot-pressing molding scenario and high-precision prediction capabilities is constructed.

[0040] Simulation mesh nodes refer to the smallest simulation units that divide the mold surface according to preset dimensions. Each node corresponds to a unique spatial coordinate and is used to calculate local pressure-temperature parameters. For example, dividing the surface of a luggage carpet mold into 5mm×5mm mesh nodes generates a total of 12,000 simulation nodes for the entire mold. Pressure-temperature field control interval distribution refers to the set of optimal pressure and temperature parameter intervals corresponding to each simulation mesh node throughout the hot-pressing process time series, covering the stages of mold closing, heating, and pressure holding. Control zones are defined by dividing the mold surface into several independent control areas based on the physical distribution of heating and hydraulic pressurization units. Each area corresponds to a set of heating / hydraulic units. For example, dividing the mold into three control zones—edge, center, and corner—based on the heater layout, each zone is equipped with four sets of independent heating / hydraulic units. The pressure-temperature field refers to the distribution set of pressure and temperature parameters that dynamically change with hot-pressing time and spatial location, covering the entire mold surface and the entire composite material to be molded. It includes two related subfields: pressure field and temperature field. Critical surfaces refer to local areas in the upper and lower model surfaces where the geometry exhibits abrupt changes in depth, curvature, or cross-sectional area in three-dimensional space; these are high-risk areas for molding defects. Dynamic surface contact state refers to the real-time interaction state between critical surfaces and the composite material during hot pressing, encompassing three parameters: surface pressure distribution, material contact state (i.e., contact area ratio, contact gap, and surface micro-vibration).

[0041] Specifically, the multi-layer hot-pressing related characteristic information of the composite material to be molded is imported into a pre-constructed hot-pressing digital twin model. Preset quality targets are set, such as carpet-like surface fit ≥98% and no wrinkles or cracks. Simulations are run, and within the time series from mold closing to the end of pressure holding, the optimal pressure and temperature ranges are calculated for each simulation mesh node of the mold surface, forming a pressure-temperature field control range distribution. That is, multi-layer hot-pressing related characteristic information is imported into the hot-pressing digital twin model, and the thermal deformation and bonding / curing constitutive equations of the composite material are defined based on the thermo-structural coupling theory. The mold surface is then divided into simulation meshes of preset sizes. True mesh nodes are established, and preset quality target constraints are input. For each mesh node, multiple combinations of temperature and pressure variables are set, and a full-process hot pressing simulation iteration is carried out to simulate the melting, bonding, and deformation processes of materials under different parameters. The molding effect corresponding to each parameter combination is quantified by constructing an evaluation function, and the effective parameter combination that meets the quality target is selected. For each mesh node, the effective parameter combination at each time stage of mold closing, heating, and pressure holding is extracted into intervals, and the upper and lower thresholds of the parameters are determined. Thus, the optimal temperature setting interval and the optimal pressure setting interval of each simulation mesh node as a function of time can be obtained. The evaluation function is as follows: ; Constraints: ; Parameter meaning explanation: The i-th and j-th mesh nodes are the overall forming quality scores at the k-th process time stage; i and j are the two-dimensional spatial coordinate indices of the simulated mesh nodes; instance values ​​are i=1,2,...,M; j=1,2,...,N (M×N is the total number of meshes); Hot pressing process time stages (mold closing / heating / pressure holding); example values Mold closing stage; The warming phase; Pressure holding stage , , Weighting coefficients (satisfying) ); The surface fit function has a value range of [0,1]. The molding defect rate function (including wrinkles, cracks, and poor bonding) has a value range of [0,1]. Interlayer bond strength function; unit of measurement: MPa; Temperature constraint upper and lower limits of the m-th layer material; Upper and lower limits of pressure constraint on the m-th layer of material; Overall quality score threshold; (Parameter combinations below this value are discarded), where, and It is a dimensionless parameter, with values ​​ranging from [0, 1]; while It is a dimensional parameter; therefore, it is necessary to consider dimensional parameters. Perform normalization processing and map it to the same level as... , For a consistent dimensionless interval of [0, 1], the commonly used linear normalization formula is: ; in, This represents the minimum acceptable value for bond strength. This represents the maximum bond strength that the material can achieve.

[0042] Right now, .

[0043] A laser positioning scanner was used to collect the spatial distribution and coverage of the embedded ceramic heating unit and the micro servo hydraulic press unit of the hot press. Based on the hardware distribution, the mold surface was divided into three control zones: the edge pressurization zone, the central constant temperature zone, and the key surface precision control zone. The pressure-temperature range of all grid nodes in each zone was combined to extract the intersection and complete the pressure-temperature field aggregation, generating a visualized pressure-temperature field control map. This control map was then input into the hot press digital twin model, focusing on the key surface to carry out targeted simulation. Parameters such as the surface pressure distribution, material contact area ratio, and micro-vibration amplitude at different time points were extracted. Finally, the dynamic surface contact state, including the surface pressure distribution, surface material contact state, and surface micro-vibration, was obtained.

[0044] This step achieves accurate prediction of hot pressing process parameters through digital twin simulation, overcoming the limitations of traditional processes that rely on empirical parameters. Through the refined division of simulation mesh nodes and the aggregation and optimization of control zones, the generated pressure-temperature field control map is both accurate and executable, and can directly guide the precise control of the hot press hardware. At the same time, the special simulation for the critical surface of geometric change obtains dynamic contact state data in advance, providing a clear reference benchmark for subsequent real-time monitoring and dynamic compensation, effectively avoiding defects such as wrinkles and cracks caused by uneven pressure and temperature on the critical surface, and improving the molding accuracy and batch consistency of automotive trunk carpets.

[0045] Furthermore, the distribution of the heating units and hydraulic pressurizing units of the hot press is collected, and control zoning and pressure-temperature field aggregation are performed in conjunction with the distribution of the pressure-temperature field control intervals. This includes: constructing a heating control domain for each heating unit and a pressurizing control domain for each hydraulic pressurizing unit based on the distribution of the heating units and hydraulic pressurizing units; matching the distribution of the pressure-temperature field control intervals based on the heating control domains and the pressurizing control domains to generate the temperature field interval distribution within the domain of each heating unit and the pressure field interval distribution within the domain of each hydraulic pressurizing unit; extracting the intersection of each temperature interval in the temperature field interval distribution within the domain, and extracting the intersection of each pressure interval in the pressure field interval distribution within the domain to generate the pressure-temperature field control map.

[0046] Specifically, the heating control domain refers to the spatial range of the mold surface that a single heating unit can effectively regulate; its boundary is determined by the power of the heating unit, the thermal radiation radius, and the thermal conductivity characteristics of the mold. The pressurization control domain refers to the spatial range of the mold surface that a single hydraulic pressurization unit can stably apply pressure; its boundary is determined by the piston stroke of the hydraulic unit, the pressure transmission angle, and the mold stiffness characteristics. The temperature field interval distribution within the domain refers to the set of optimal temperature setting intervals for all simulated mesh nodes within a single heating control domain after spatially matching the pressure-temperature field control interval distribution with the heating control domain. The pressure field interval distribution within the domain refers to the set of optimal pressure setting intervals for all simulated mesh nodes within a single pressurization control domain after spatially matching the pressure-temperature field control interval distribution with the pressurization control domain.

[0047] Specifically, a laser positioning scanner is used to collect the spatial coordinates of the heating and pressurizing units of the hot press. Combined with parameters such as the power of the heating unit and the piston size of the pressurizing unit, CAD 3D modeling software is used to construct the heating control domain for each group of heating units and the pressurizing control domain for each group of hydraulic pressurizing units. The pressure-temperature field control interval distribution is imported into the control domain, and the temperature and pressure intervals of each simulation mesh node are mapped to the corresponding heating and pressurizing control domains. This generates the temperature field interval distribution within the domain for each heating unit and the pressure field interval distribution within the domain for each hydraulic pressurizing unit, such as the edge heating control domain. The temperature ranges of the 200 grid nodes within the domain are matched to form a unified intra-domain temperature field range distribution. The pressure ranges of the 150 grid nodes within the critical surface pressurization control domain are matched to form a unified intra-domain pressure field range distribution. Finally, the intersection of the intra-domain temperature field range distributions of each heating control domain is extracted to obtain the optimal execution temperature range for that region. At the same time, the intersection of the intra-domain pressure field range distributions of each pressurization control domain is extracted to obtain the optimal execution pressure range for that region. The temperature and pressure execution ranges of all control domains are integrated according to time series and spatial location to generate a pressure-temperature field control map.

[0048] This step, by constructing heating and pressurization control domains, achieves precise matching between the pressure-temperature field control range distribution and the hot press hardware, avoiding parameter redundancy caused by insufficient hardware control capabilities. It effectively improves the accuracy of mold surface zoning control, reduces defects such as wrinkles and cracks on key surfaces caused by pressure-temperature mismatch, and ensures the stability of the molding quality of automotive trunk carpets.

[0049] Furthermore, such as Figure 2As shown, by comparing the synchronous surface contact state data with the dynamic surface contact state, a state difference vector field is calculated, and a digital hydraulic unit embedded in the upper and lower model surfaces is driven to perform dynamic compensation. This includes: identifying abnormal state difference vectors that do not meet a preset difference threshold and the corresponding abnormal surface positions based on the state difference vector field; performing compensation mode matching and compensation parameter matching based on the abnormal state difference vectors, wherein the compensation mode includes a displacement compensation mode and / or a damping compensation mode; matching the corresponding digital hydraulic component in the digital hydraulic unit according to the matched compensation mode, and controlling it based on the matched compensation parameters.

[0050] Specifically, the abnormal state difference vector refers to a three-dimensional vector in the state difference vector field where at least one of the indicators—pressure difference, contact state difference, or micro-vibration difference—exceeds a preset difference threshold. It is the core basis for determining whether compensation is needed in the molding process. The preset difference threshold is a pre-defined allowable range of state differences based on ideal states from digital twin simulation and historical molding data, used to distinguish between normal fluctuations and abnormal deviations requiring compensation. The abnormal surface location refers to the location of the mold surface mesh node corresponding to the abnormal state difference vector, typically concentrated on critical surfaces. For example, the 35th mesh node at the curved corner of the mold corresponding to the abnormal state difference vector is the abnormal surface location. Compensation mode matching is the process of matching the corresponding compensation method based on the dominant deviation type of the abnormal state difference vector; the dominant deviation determines the compensation priority. Examples: anomalies dominated by pressure difference and contact area difference are matched with displacement compensation modes; anomalies dominated by micro-vibration difference are matched with damping compensation modes. Digital hydraulic components refer to the actuators in the digital hydraulic unit that correspond one-to-one with the compensation modes; magnetostrictive actuators correspond to displacement compensation modes, and micro-piezoelectric actuators correspond to damping compensation modes.

[0051] Specifically, the system imports state difference vector field data and preset difference thresholds. By comparing the thresholds, it identifies abnormal state difference vectors that do not meet the threshold requirements and locates the corresponding abnormal surface positions, classifying them as abnormal locations. Based on a compensation matching model trained using digital twin simulation, it matches compensation modes and compensation parameters according to the dominant deviation type of the abnormal vector. For example, if the dominant deviation is a pressure difference of -0.1MPa and a contact area difference of -10%, a displacement compensation mode is matched, and the elongation parameter of the magnetostrictive actuator is matched to 0.05mm based on the deviation value. If the dominant deviation is a micro-vibration difference of +10μm, a damping compensation mode is matched, and the damping force parameter of the micro-piezoelectric actuator is matched to 5N. Finally, through the PLC control system, it sends instructions to the digital hydraulic unit pre-embedded in the abnormal surface position to match the corresponding digital hydraulic components. That is, the displacement compensation mode activates the magnetostrictive actuator, and the damping compensation mode activates the micro-piezoelectric actuator, executing precise compensation actions according to the matched compensation parameters.

[0052] This step achieves accurate identification of abnormal deviations through threshold comparison, and realizes "differentiated and precise compensation" based on the compensation mode and parameter matching of the deviation type. Combined with the dual-mode execution mechanism of the digital hydraulic components, it can simultaneously cope with pressure contact deviation and micro-vibration deviation, forming a closed-loop adaptive control of the molding process, which effectively reduces the defect rate of wrinkles, cracks and other defects on key surfaces.

[0053] Furthermore, compensation mode matching is performed based on the abnormal state difference vector, wherein the compensation mode includes a displacement compensation mode and / or a damping compensation mode, including: adding a simulation module about the digital hydraulic unit to the hot-press digital twin model, performing compensation simulation training by pre-setting an abnormal state difference vector sample set, and establishing a compensation simulation model; using the compensation simulation model to perform compensation simulation on the abnormal state difference vector, and outputting the matched compensation mode and compensation parameters.

[0054] Specifically, the simulation module of the digital hydraulic unit is integrated within the thermo-pressurized digital twin model. This sub-module, used to simulate the working characteristics of micro piezoelectric actuators and magnetostrictive actuators, includes the mapping relationship between the actuator's displacement output, damping force output, and deviation correction effect. The intelligent matching model, obtained after compensation simulation training, has the ability to output the optimal compensation mode and corresponding compensation parameters based on any abnormal state difference vector input.

[0055] Specifically, a simulation module for a digital hydraulic unit is added to the pre-constructed hot-pressing digital twin model. Based on the physical characteristics of micro piezoelectric actuators and magnetostrictive actuators, a simulation logic of "input parameters - execution action - correction effect" is established. Historical deviation data of hot-pressing of car trunk carpets is collected, and combined with the extreme deviation scenarios simulated in the simulation, a sample set of abnormal state difference vectors containing multiple sets of different deviation combinations is constructed. Using MATLAB machine learning tools, the sample set is input into the digital hydraulic unit simulation module to carry out compensation simulation training. With the goal of achieving a conformity of ≥98% between the contact state of the profile after deviation correction and the ideal state, the matching rules of the compensation mode and compensation parameters are iteratively optimized to establish a compensation simulation model. The abnormal state difference vectors identified in the actual molding process are input into this model, and the matching compensation mode and compensation parameters are output through simulation calculations.

[0056] The compensation simulation model adopts a two-layer integrated structure of "digital hydraulic unit simulation submodule + intelligent matching submodule". The digital hydraulic unit simulation submodule embeds a thermo-pressurized digital twin model, which is the core of the simulation at the physical execution level. It includes two parallel actuator response models: one is the "voltage-damping force" mapping model of the micro piezoelectric actuator, which incorporates the inverse piezoelectric effect coefficient of the piezoelectric material and can accurately output the damping force and response delay corresponding to different input voltages; the other is the "current-displacement" mapping model of the magnetostrictive actuator, which inputs the magnetostriction coefficient of the Terfenol-D magnetostrictive material to achieve a linear mapping between input current and output displacement, while integrating a mold stiffness compensation coefficient to correct the actual displacement output deviation. The intelligent matching submodule employs a lightweight supervised learning neural network structure, using a 3-layer fully connected network: the input layer has 3 nodes, corresponding to the pressure difference, contact area difference, and micro-vibration difference in the abnormal state difference vector; the hidden layer has 16 nodes, using the ReLU activation function to enhance feature extraction capabilities; the output layer has 3 nodes, corresponding to the compensation mode, displacement compensation amount, and damping force output value, respectively. The network uses a Dropout layer (dropout rate of 0.2) to avoid overfitting. The two submodules communicate in real time through the OPC UA data interface. The digital hydraulic unit simulation submodule provides feedback on the correction effect to the intelligent matching submodule, while the intelligent matching submodule outputs precise control commands to the former.

[0057] The construction process revolves around the core logic of "physical property mapping – data foundation construction – module integration and adaptation," and consists of three steps: First, a digital hydraulic unit simulation submodule is built. In the Transient Structural module of ANSYS Workbench, the 3D geometric models of the micro piezoelectric actuator and magnetostrictive actuator are imported, material parameters are entered, and the "input-output" response curve is fitted through static structural simulation. For example, a gradient current load of 0-2A is applied to the magnetostrictive actuator, output displacement data is extracted, and a linear mapping equation is obtained using the least squares method. Simultaneously, the surface stiffness parameters of the hot press mold are integrated to construct the simulation logic of "actuator action – mold deformation – deviation correction." Second, an intelligent matching submodule is built. Using MATLAB's Neural Network Toolbox, a three-layer fully connected neural network was created. The input layer data normalization range was set to [-1, 1], the hidden layer used the ReLU activation function, and the output layer used the Sigmoid function to map displacement and damping force parameters to the actual execution range. Historical deviation data from hot pressing and simulation limit deviation scenario data were imported to construct an abnormal state difference vector sample set containing multiple sets of samples. Each sample includes five dimensions: pressure difference, contact area difference, micro-vibration difference, optimal compensation mode, and optimal compensation parameter. These were divided into training, validation, and test sets in a 7:2:1 ratio. The third step was to implement module integration and interface adaptation. A bidirectional data communication link between the digital hydraulic unit simulation submodule and the intelligent matching submodule was established through the MATLAB Co-Simulation interface of ANSYS Workbench. A data interaction protocol was defined: the compensation command output by the intelligent matching submodule serves as the input load of the simulation submodule, and the deviation correction output by the simulation submodule serves as the feedback data of the intelligent matching submodule. At the same time, a data preprocessing module was developed to realize the timestamp alignment, spatial coordinate matching, and dimension normalization of the anomaly vectors, ensuring the consistency of data interaction between the two submodules.

[0058] The training process employs an "iterative training - verification and optimization - solidified output" strategy to ensure model matching accuracy and generalization ability: First, the training set samples are imported into the intelligent matching submodule, and training parameters are set as follows: learning rate 0.001, iteration count 500, batch size 32. The core evaluation metric is "corrected surface fit ≥ 98%", and the mean squared error (MSE) is used as the loss function to optimize network weights and biases. During training, every 50 iterations, the compensation scheme output by the current network is input into the digital hydraulic unit simulation submodule to simulate the correction effect and obtain actual fit data. If the fit does not meet the standard, the network parameters are adjusted in reverse. The second step involves inputting the validation set into the trained model and statistically analyzing the compensation pattern matching accuracy and parameter prediction error. If the pattern matching accuracy is below 95% or the parameter prediction error is below 95%, 50 sets of high-bias scene samples are added to the training set, and the model is retrained using a learning rate decay strategy (halving the learning rate every 100 iterations). Simultaneously, the network structure is optimized by adjusting the number of hidden layer nodes (16→20) and changing the activation function to Leaky ReLU to reduce the risk of gradient vanishing. The third step involves inputting the test set into the optimized model to verify its generalization ability. The required pattern matching accuracy is ≥98%, parameter prediction error is ≤0.003mm (displacement) and ≤0.3N (damping force), and the corrected surface fit is ≥98%. After achieving these targets, the model parameters (weight matrix, bias vector) are solidified, generating a directly callable model file, which is then integrated into the hot-pressing PLC control system. Simultaneously, a model call interface program is written to implement real-time compensation scheme output after actual abnormal vector input.

[0059] This step, by building and training a compensation simulation model in a digital twin model, achieves intelligent matching of compensation modes and parameters, overcoming the limitations of traditional compensation that relies on manual experience to set parameters, and effectively improving the correction accuracy of key surface deviations.

[0060] In summary, the composite molding control method for automotive trunk carpets provided in this application has the following technical effects: 1. By integrating closed-loop methods such as digital twin simulation to generate pressure-temperature field maps, real-time monitoring by flexible sensor arrays, and dynamic compensation by digital hydraulic units, the limitations of traditional hot pressing fixed parameters are overcome, the forming state of key surfaces is precisely controlled, defects such as wrinkles and cracks are effectively solved, the surface fit of car trunk carpets is improved, and the batch consistency and forming qualification rate of products are greatly enhanced.

[0061] 2. By using digital twin simulation, mesh node parameter calculation, and critical surface-specific simulation, a spatiotemporal pressure-temperature field control map and dynamic contact state data are generated, realizing the transformation from global experience-based control to local precise control. This provides clear parameter benchmarks for hot pressing operations, allows for early prediction of critical surface forming risks, and reduces the probability of defects caused by parameter mismatch.

[0062] 3. By identifying abnormal vectors and locations through threshold comparison, matching compensation modes and parameters, and driving digital hydraulic components, differentiated and precise compensation is achieved, solving the drawbacks of traditional "one-size-fits-all" compensation. Pressure / contact deviation and vibration deviation are controlled separately to quickly correct molding abnormalities, improve the accuracy of key surface deviation correction, and ensure stable molding quality.

[0063] Example 2, based on the same inventive concept as the composite molding control method for a car trunk carpet in the foregoing examples, such as... Figure 3 As shown in the figure, this application provides a composite molding control system for automotive trunk carpets. The system includes: a composite material acquisition module 11, used to acquire multi-layer carpet material for composite molding, and place the stacked multi-layer carpet material between the upper and lower molds of a hot press to obtain the composite material to be molded; a digital twin simulation module 12, used to collect multi-layer hot pressing related feature information of the composite material to be molded, and generate a pressure-temperature field control spectrum and dynamic surface contact state of key surfaces that vary with time and space through digital twin simulation; a state data generation module 13, used to control the hot press to perform molding operations with the pressure-temperature field control spectrum, synchronously activate a flexible thin film sensor array, monitor the pressure distribution, material contact state and micro-vibration of key surfaces in real time, and generate synchronous surface contact state data; and a dynamic compensation module 14, used to compare the synchronous surface contact state data with the dynamic surface contact state, calculate the state difference vector field, and drive the digital hydraulic unit embedded in the upper and lower mold surfaces to perform dynamic compensation.

[0064] Furthermore, the digital twin simulation module 12 is also used to perform the following steps: constructing a hot-pressing related attribute set, including several material properties related to the hot-pressing process; collecting the multi-layer material property values ​​corresponding to the composite material to be molded based on the hot-pressing related attribute set and performing normalization processing to generate the multi-layer hot-pressing related feature information.

[0065] Furthermore, the digital twin simulation module 12 is also used to perform the following steps: importing the multi-layer hot-press related feature information into the pre-constructed hot-press digital twin model, running the simulation, and calculating the optimal temperature setting range and optimal pressure setting range at each simulation grid node on the mold surface in the entire time series from the start of mold closing to the end of pressure holding, forming a pressure-temperature field control range distribution; collecting the distribution of the heating unit and hydraulic pressurization unit of the hot press, and combining the pressure-temperature field control range distribution to perform control partitioning and pressure-temperature field aggregation, generating the pressure-temperature field control map; inputting the pressure-temperature field control map into the hot-press digital twin model to perform key surface simulation, and obtaining the dynamic surface contact state, including surface pressure distribution, surface material contact state, and surface micro-vibration.

[0066] Furthermore, the digital twin simulation module 12 is also used to perform the following steps: constructing a heating control domain for each heating unit and a pressurization control domain for each hydraulic pressurization unit based on the distribution of the heating unit and the hydraulic pressurization unit; matching the pressure-temperature field control interval distribution based on the heating control domain and the pressurization control domain to generate an intra-domain temperature field interval distribution for each heating unit and an intra-domain pressure field interval distribution for each hydraulic pressurization unit; extracting the intersection of each temperature interval in the intra-domain temperature field interval distribution, and extracting the intersection of each pressure interval in the intra-domain pressure field interval distribution to generate the pressure-temperature field control map.

[0067] Furthermore, the flexible thin-film sensing array is located on the inner wall of the mold surface of the upper and lower molds, and includes a pressure sensing unit, a capacitive proximity sensing unit, and a piezoelectric vibration sensing unit.

[0068] Furthermore, the key surface is a local region in the mold corresponding to the upper and lower molds where the geometry has abrupt changes in depth, curvature, or cross-sectional area in three-dimensional space.

[0069] Furthermore, the dynamic compensation module 14 is also used to perform the following steps: identifying abnormal state difference vectors that do not meet the preset difference threshold and the corresponding abnormal surface positions based on the state difference vector field; performing compensation mode matching and compensation parameter matching based on the abnormal state difference vectors, wherein the compensation mode includes displacement compensation mode and / or damping compensation mode; matching the corresponding digital hydraulic components in the digital hydraulic unit according to the matched compensation mode, and performing control based on the matched compensation parameters.

[0070] Furthermore, the dynamic compensation module 14 is also used to perform the following steps: adding a simulation module for the digital hydraulic unit to the hot-press digital twin model, performing compensation simulation training by pre-setting a sample set of abnormal state difference vectors, and establishing a compensation simulation model; using the compensation simulation model to perform compensation simulation on the abnormal state difference vectors, and outputting a matching compensation mode and compensation parameters.

[0071] Furthermore, the digital hydraulic unit includes a micro piezoelectric actuator and a magnetostrictive actuator, wherein the micro piezoelectric actuator corresponds to a damping compensation mode and the magnetostrictive actuator corresponds to a displacement compensation mode.

[0072] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall fall within the scope of the present invention.

Claims

1. A method for controlling the composite molding of automotive trunk carpets, characterized in that, include: Obtain multi-layer carpet material for composite molding, place the stacked multi-layer carpet material between the upper and lower molds of a hot press to obtain the composite material to be molded; Collect multilayer hot-press related characteristic information of the composite material to be formed, and generate pressure-temperature field control spectrum and dynamic surface contact state of key surfaces that vary with time and space through digital twin simulation. The hot press is controlled to perform molding operations using the pressure-temperature field control spectrum, and the flexible thin film sensor array is activated simultaneously to monitor the pressure distribution, material contact state and micro-vibration of key surfaces in real time, and generate synchronous surface contact state data. By comparing the synchronous surface contact state data with the dynamic surface contact state, the state difference vector field is calculated, and the digital hydraulic unit embedded in the upper and lower model surfaces is driven to perform dynamic compensation.

2. The composite molding control method for automotive trunk carpet as described in claim 1, characterized in that, Collect multilayer hot-press related characteristic information of the composite material to be molded, including: Construct a set of hot-pressing related properties, including several material properties related to the hot-pressing process; Based on the hot-pressing related attribute set, the multi-layer material attribute values ​​corresponding to the composite material to be formed are collected and normalized to generate the multi-layer hot-pressing related feature information.

3. The composite molding control method for automotive trunk carpet as described in claim 1, characterized in that, Through digital twin simulation, pressure-temperature field control maps and dynamic surface contact states of key surfaces that vary with time and space are generated, including: The multi-layer hot pressing related feature information is imported into the pre-constructed hot pressing digital twin model. The simulation is run, and the preset quality target is calculated to obtain the optimal temperature setting range and optimal pressure setting range at each simulation grid node on the mold surface in the whole process time series from the start of mold closing to the end of pressure holding, which constitutes the pressure-temperature field control range distribution. The distribution of the heating unit and hydraulic pressurization unit of the hot press is collected, and the pressure-temperature field control interval distribution is combined with the pressure-temperature field aggregation to generate the pressure-temperature field control map. The pressure-temperature field control spectrum is input into the hot-pressing digital twin model for key surface simulation to obtain the dynamic surface contact state, including surface pressure distribution, surface material contact state, and surface micro-vibration.

4. The composite molding control method for automotive trunk carpet as described in claim 3, characterized in that, The distribution of the heating unit and hydraulic pressurization unit of the hot press is collected, and the control zoning and pressure-temperature field aggregation are executed in conjunction with the distribution of the pressure-temperature field control interval, including: Based on the distribution of heating units and hydraulic pressurization units, a heating control domain for each heating unit and a pressurization control domain for each hydraulic pressurization unit are constructed. The pressure-temperature field control interval distribution is matched based on the heating control domain and the pressurization control domain to generate the temperature field interval distribution within the domain of each heating unit and the pressure field interval distribution within the domain of each hydraulic pressurization unit. The intersection of each temperature range within the temperature field range is extracted, and the intersection of each pressure range within the pressure field range is extracted to generate the pressure-temperature field control map.

5. The composite molding control method for automotive trunk carpet as described in claim 1, characterized in that, The flexible thin-film sensing array is located on the inner wall of the mold surface of the upper and lower molds, and includes a pressure sensing unit, a capacitive proximity sensing unit, and a piezoelectric vibration sensing unit.

6. The composite molding control method for automotive trunk carpet as described in claim 1, characterized in that, The key surface is a local region in the mold corresponding to the upper and lower molds where the geometry has abrupt changes in depth, curvature, or cross-sectional area in three-dimensional space.

7. The composite molding control method for automotive trunk carpet as described in claim 1, characterized in that, By comparing the synchronous surface contact state data with the dynamic surface contact state, a state difference vector field is calculated, and the digital hydraulic units embedded in the upper and lower model surfaces are driven to perform dynamic compensation, including: Based on the state difference vector field, identify abnormal state difference vectors that do not meet the preset difference threshold, and the corresponding abnormal surface positions; Compensation mode matching and compensation parameter matching are performed based on the abnormal state difference vector, wherein the compensation mode includes displacement compensation mode and / or damping compensation mode. According to the matching compensation mode, the corresponding digital hydraulic component is matched in the digital hydraulic unit, and control is performed based on the matching compensation parameters.

8. The composite molding control method for automotive trunk carpet as described in claim 7, characterized in that, Compensation mode matching is performed based on the abnormal state difference vector, wherein the compensation mode includes a displacement compensation mode and / or a damping compensation mode, including: A simulation module for the digital hydraulic unit is added to the hot-press digital twin model. An abnormal state difference vector sample set is preset for compensation simulation training to establish a compensation simulation model. The compensation simulation model is used to perform compensation simulation on the abnormal state difference vector, and the matching compensation mode and compensation parameters are output.

9. The composite molding control method for automotive trunk carpet as described in claim 8, characterized in that, The digital hydraulic unit includes a miniature piezoelectric actuator and a magnetostrictive actuator. The miniature piezoelectric actuator corresponds to a damping compensation mode, and the magnetostrictive actuator corresponds to a displacement compensation mode.

10. A composite molding control system for automotive trunk carpets, characterized in that, The system is used for performing the composite molding control method for automotive trunk carpets according to any one of claims 1 to 9, the system comprising: The composite material obtaining module is used to obtain multi-layer carpet material for composite molding. The stacked multi-layer carpet material is placed between the upper and lower molds of a hot press to obtain the composite material to be molded. The digital twin simulation module is used to collect multi-layer hot-press related characteristic information of the composite material to be formed. Through digital twin simulation, it generates a pressure-temperature field control spectrum that varies with time and space and a dynamic surface contact state of key surfaces. The state data generation module is used to control the hot press to perform molding operations based on the pressure-temperature field control spectrum, and simultaneously activate the flexible thin film sensor array to monitor the pressure distribution, material contact state and micro-vibration of key surfaces in real time, and generate synchronous surface contact state data. The dynamic compensation module is used to compare the synchronous surface contact state data with the dynamic surface contact state, calculate the state difference vector field, and drive the digital hydraulic unit embedded in the upper and lower model surfaces to perform dynamic compensation.

Images