A method of manufacturing a composite shoe sole and related apparatus

Abstract

The application discloses a composite shoe sole manufacturing method and related device, and relates to the technical field of data processing.The method comprises the following steps: shoe sole mold model identification is carried out based on radio frequency identification technology to obtain shoe sole mold model information; a multi-physical field coupling model is established, and a multi-dimensional variable data set is constructed based on the multi-physical field coupling model; a back propagation neural network is trained based on the multi-dimensional variable data set; process parameter information is generated based on the trained back propagation neural network and the shoe sole mold model information; and the target equipment sprays non-foaming polyurethane on the shoe sole mold, fills and melts the thermoplastic polyurethane foam elastomer based on the process parameter information and the PID feedback regulator, and obtains the composite shoe sole.The application effectively improves the reliability of shoe sole manufacturing, so that the obtained composite shoe sole not only has an attractive appearance and light weight, but also has excellent physical properties such as folding resistance, wear resistance, slip resistance and high resilience.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and apparatus for manufacturing composite material shoe soles. Background Technology

[0002] In the field of shoe sole manufacturing, traditional compression molding and casting processes have long dominated. However, these processes have consistently struggled to overcome the inherent limitations of material properties, resulting in finished products that are generally heavy, lack elasticity, have poor abrasion resistance, and are prone to cracking. Furthermore, current process parameter analysis in shoe sole manufacturing relies heavily on manual experience, which is overly dependent on the professional expertise of personnel, making it difficult to guarantee the accuracy of the analysis. Moreover, with the ever-increasing volume of process parameter data, manual analysis can no longer meet the high efficiency and precision requirements of current shoe sole manufacturing. Additionally, the lack of real-time feedback and adjustment in current shoe sole manufacturing processes leads to significant deviations between the final manufactured soles and the intended results, making it difficult to meet the high-quality requirements for shoe soles. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a method and related apparatus for manufacturing composite material shoe soles, which can effectively improve the precision of shoe sole manufacturing, so that the obtained composite material shoe soles are not only aesthetically pleasing and lightweight, but also have excellent physical properties.

[0004] To address the aforementioned technical problems, the present invention provides a method for manufacturing composite material shoe soles, the method comprising: Shoe sole mold model identification is performed based on radio frequency identification technology to obtain shoe sole mold model information; Establish a multiphysics coupling model and construct a multidimensional variable dataset based on the multiphysics coupling model; The backpropagation neural network is trained based on the multidimensional variable dataset to obtain a trained backpropagation neural network. Based on the trained backpropagation neural network, process parameter information is generated using the shoe sole mold model information; The target equipment uses a proportional-integral-derivative (PID) feedback regulator based on the process parameter information to perform non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole mold to obtain a composite material shoe sole. The PID feedback regulator is used to adjust the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment processes of the shoe sole mold in real time.

[0005] Optionally, the step of identifying the shoe sole mold model based on radio frequency identification technology to obtain shoe sole mold model information includes: Obtain the radio frequency identification (RFID) tag of the shoe sole mold, and identify the shoe sole mold model based on the RFID tag to obtain the shoe sole mold model information.

[0006] Optionally, the step of establishing a multiphysics coupling model and constructing a multidimensional variable dataset based on the multiphysics coupling model includes: Determine the physical field interface, coupling logic, and boundary conditions; Geometric partitioning information is determined based on fluid domain, solid domain, and thermal boundary, and material parameter information is obtained; Determine the meshing strategy based on the geometric partitioning information; Based on the physical field interface, coupling logic, boundary conditions, geometric meshing information, material parameter information, and mesh strategy, the solver and output variables are configured, and a multiphysics coupling model is established based on the solver and output variables. Based on the multiphysics coupling model, the interaction relationship between the temperature field, stress field and flow field of each shoe sole mold model is determined, and a multidimensional variable dataset is constructed based on the interaction relationship.

[0007] Optionally, determining the physical field interface, coupling logic, and boundary conditions includes: Obtain the laminar flow interface, turbulent flow interface, solid mechanics interface, and heat transfer interface, and determine the physical field interface based on the laminar flow interface, turbulent flow interface, solid mechanics interface, and heat transfer interface; Obtain fluid-structure interaction, thermal-fluid interaction, and thermal-structure interaction, and generate coupling logic based on the fluid-structure interaction, thermal-fluid interaction, and thermal-structure interaction; Obtain the fluid inlet, fluid outlet, solid fixed constraint, heat source, and heat sink; determine the coupling boundary based on the arbitrary Lagrange-Euler (ALE) method; and determine the boundary conditions based on the coupling boundary, fluid inlet, fluid outlet, solid fixed constraint, heat source, and heat sink.

[0008] Optionally, training the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network includes: The multidimensional variable dataset is cleaned and labeled to obtain a labeled multidimensional variable dataset. The backpropagation neural network is trained based on the labeled multidimensional variable dataset to obtain a trained backpropagation neural network.

[0009] Optionally, the step of generating process parameter information based on the trained backpropagation neural network using the shoe sole mold model information includes: The shoe sole mold model information is input into a trained backpropagation neural network. The trained backpropagation neural network analyzes process parameters based on the shoe sole mold model information using neuron connection weights, bias values, and activation functions to obtain process parameter information.

[0010] Optionally, the target equipment, based on the process parameter information and combined with a proportional-integral-derivative PID feedback regulator, performs non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the sole mold to obtain a composite material sole, including: The automatic rotary machine aligns the robotic arm with the shoe sole mold at the workstation, and the spray foaming machine delivers non-foamed polyurethane to the spray head of the robotic arm. Based on the process parameter information and combined with the PID feedback regulator, the spraying head of the robotic arm is controlled to perform non-foamed polyurethane spraying on the surface of the shoe sole mold at the workstation to obtain a sprayed shell. Based on the process parameter information, the screw feeder and hot air generator are controlled by a PID feedback regulator to fill the shoe sole mold cavity at the workstation with thermoplastic polyurethane foam elastomer to obtain a filled inner core. Based on the process parameter information and combined with the PID feedback regulator, the automatic opening and closing mechanism of the casting mold is controlled to melt and bond the sprayed outer shell and the filling inner core to obtain a composite material shoe sole.

[0011] Optionally, the real-time adjustment of the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting process of the shoe sole mold includes: Real-time detection values ​​of the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling and melting process of shoe sole mold are obtained based on a combination of sensors. The difference between the real-time detection value and the process parameter information is calculated, and a feedback control quantity is generated based on the difference using a PID feedback regulator. The operating parameters of the target equipment are then adjusted in real time based on the feedback control quantity.

[0012] In addition, the present invention also provides a composite material shoe sole manufacturing apparatus, the apparatus comprising: Shoe sole mold model identification module: used to identify shoe sole mold models based on radio frequency identification technology and obtain shoe sole mold model information; Dataset construction module: used to establish a multiphysics coupling model and construct a multidimensional variable dataset based on the multiphysics coupling model; Network training module: used to train the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network; Parameter generation module: used to generate process parameter information based on the trained backpropagation neural network and the shoe sole mold model information; The sole manufacturing module is used by the target equipment to perform non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling and melting treatment on the sole mold based on the process parameter information and a proportional-integral-derivative PID feedback regulator to obtain a composite material sole. The PID feedback regulator is used to adjust the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling and melting treatment process of the sole mold in real time.

[0013] In addition, the present invention also provides a computer-readable storage medium that stores computer instructions that, when executed on an electronic device, cause the electronic device to perform the above-described composite material shoe sole manufacturing method.

[0014] In this embodiment of the invention, a multiphysics coupling model is established to construct a multidimensional variable dataset. A backpropagation neural network is trained based on this dataset to obtain a trained backpropagation neural network. This trained backpropagation neural network then generates process parameter information using the shoe sole mold model information, effectively improving the accuracy and efficiency of process parameter analysis and significantly reducing reliance on manual labor. The target equipment, based on the process parameter information and combined with a PID feedback regulator, performs non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole mold. This effectively improves the precision of shoe sole manufacturing. Furthermore, by performing non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole, the resulting composite material shoe sole is not only aesthetically pleasing and lightweight but also possesses excellent physical properties such as flexural resistance, abrasion resistance, slip resistance, and high resilience. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic flowchart of the composite material shoe sole manufacturing method in an embodiment of the present invention; Figure 2 This is a flowchart illustrating a method for manufacturing composite material shoe soles according to another embodiment of the present invention; Figure 3 This is a schematic diagram of the structural composition of the composite material shoe sole manufacturing device in an embodiment of the present invention; Figure 4This is a schematic diagram of feedback closed-loop control in an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 Please see Figure 1 , Figure 1 This is a schematic flowchart of a composite material shoe sole manufacturing method according to an embodiment of the present invention, the method comprising: S11: Identify the shoe sole mold model based on radio frequency identification technology to obtain shoe sole mold model information; In the specific implementation of this invention, the radio frequency identification (RFID) electronic tag of the shoe sole mold is obtained, and the shoe sole mold model is identified by the RFID electronic tag. The shoe sole mold model can speed up the analysis of subsequent process parameters and raw material parameters.

[0019] S12: Establish a multiphysics coupling model and construct a multidimensional variable dataset based on the multiphysics coupling model; In the specific implementation of this invention, the physical field interface, coupling logic, and boundary conditions are determined. Geometric partitioning information is determined based on the fluid domain, solid domain, and thermal boundary. Material parameter information is obtained. A meshing strategy is determined based on the geometric partitioning information. The solver and output variables are configured through the physical field interface, coupling logic, boundary conditions, geometric partitioning information, material parameter information, and meshing strategy. A multiphysics coupling model is established based on the solver and output variables. The interaction relationship between the temperature field, stress field, and flow field of each shoe sole mold model is determined based on the multiphysics coupling model. A multidimensional variable dataset is constructed based on the interaction relationship. The multidimensional variable dataset provides sufficiently comprehensive and specific data support for the training of the backpropagation neural network.

[0020] S13: Train the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network; In the specific implementation of this invention, data cleaning and labeling are performed on the multidimensional variable dataset to obtain a labeled multidimensional variable dataset. The backpropagation neural network is trained based on the labeled multidimensional variable dataset, which can effectively improve the performance of the backpropagation neural network.

[0021] S14: Based on the trained backpropagation neural network, process parameter information is generated using the shoe sole mold model information; In the specific implementation of this invention, the shoe sole mold model information is input into the trained backpropagation neural network. The trained backpropagation neural network uses the neuron connection weights, bias values ​​and activation functions based on the shoe sole mold model information to perform process parameter analysis, which can improve the reliability and efficiency of process parameter analysis.

[0022] S15: The target equipment performs non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling and melting treatment on the shoe sole mold based on the process parameter information and combined with the PID feedback regulator to obtain a composite material shoe sole. The PID feedback regulator is used to adjust the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling and melting treatment process of the shoe sole mold in real time.

[0023] In the specific implementation of this invention, the target equipment is a device for spraying non-foamed polyurethane, filling thermoplastic polyurethane foam elastomer, and melting treatment. It includes an automatic rotary machine, a robotic arm, a spray head, a spray foaming machine, an automatic rotary machine, a screw feeder, and a hot air generator. The automatic rotary machine aligns the robotic arm with the shoe sole mold at the workstation. The spray foaming machine delivers non-foamed polyurethane to the spray head of the robotic arm. Based on process parameter information and a proportional-integral-derivative (PID) feedback regulator, the robotic arm's spray head is controlled to spray non-foamed polyurethane onto the surface of the shoe sole mold at the workstation to obtain a sprayed outer shell. Based on process parameter information and a PID feedback regulator, the screw feeder and hot air generator are controlled to fill the cavity of the shoe sole mold at the workstation with thermoplastic polyurethane foam elastomer to obtain a filled inner core. An automatic mold opening and closing mechanism melts and bonds the sprayed outer shell and the filled inner core to obtain a composite material shoe sole. It can effectively improve the precision of shoe sole manufacturing. By mixing non-foamed polyurethane spraying and thermoplastic polyurethane foam elastomer in the shoe sole manufacturing, the resulting composite material shoe sole not only has aesthetics and wearing comfort, but also has high physical properties such as flexural resistance, wear resistance, slip resistance and high elasticity.

[0024] In this embodiment of the invention, a multiphysics coupling model is established to construct a multidimensional variable dataset. A backpropagation neural network is trained based on this dataset to obtain a trained backpropagation neural network. This trained backpropagation neural network then generates process parameter information using the shoe sole mold model information, effectively improving the accuracy and efficiency of process parameter analysis and significantly reducing reliance on manual labor. The target equipment, based on the process parameter information and combined with a PID feedback regulator, performs non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole mold. This effectively improves the precision of shoe sole manufacturing. Furthermore, by performing non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole, the resulting composite material shoe sole is not only aesthetically pleasing and lightweight but also possesses excellent physical properties such as flexural resistance, abrasion resistance, slip resistance, and high resilience.

[0025] Example 2 Please see Figure 2 , Figure 2 This is a schematic flowchart of a composite material shoe sole manufacturing method according to another embodiment of the present invention, the method comprising: S201: Identify shoe sole mold model based on radio frequency identification technology to obtain shoe sole mold model information; In the specific implementation of this invention, the step of identifying the shoe sole mold model based on radio frequency identification technology to obtain shoe sole mold model information includes: acquiring the radio frequency identification electronic tag of the shoe sole mold, and identifying the shoe sole mold model based on the radio frequency identification electronic tag to obtain shoe sole mold model information.

[0026] Specifically, the process involves acquiring the Radio Frequency Identification (RFID) electronic tag of the shoe sole mold. The automatic rotary machine has multiple stations evenly distributed around its circumference, and the station can be automatically switched through a rotating mechanism. Each station has a fixed shoe sole mold, and each shoe sole mold has its corresponding RFID electronic tag. The shoe sole mold model is identified based on the RFID electronic tag. Information is extracted from the RFID electronic tag to obtain the tag information. The shoe sole mold model is matched according to the tag information to obtain the shoe sole mold model information.

[0027] S202: Establish a multiphysics coupling model and construct a multidimensional variable dataset based on the multiphysics coupling model; In the specific implementation of this invention, the establishment of a multiphysics coupling model and the construction of a multidimensional variable dataset based on the multiphysics coupling model include: determining the physical field interface, coupling logic, and boundary conditions; determining geometric partitioning information based on the fluid domain, solid domain, and thermal boundary, and obtaining material parameter information; determining a meshing strategy based on the geometric partitioning information; configuring the solver and output variables based on the physical field interface, coupling logic, boundary conditions, geometric partitioning information, material parameter information, and meshing strategy, and establishing a multiphysics coupling model based on the solver and output variables; determining the interaction relationship between the temperature field, stress field, and flow field of each shoe sole mold model based on the multiphysics coupling model, and constructing a multidimensional variable dataset based on the interaction relationship.

[0028] Specifically, by setting interfaces such as fluid interfaces and solid mechanics interfaces, the configured interfaces are combined to form a physical field interface. The coupling logic is the coupling logic between fluid, solid, and heat transfer, and the boundary conditions are the constraints or boundaries of fluid, solid, and heat transfer. Geometric partitioning information is determined based on the fluid domain, solid domain, and thermal boundaries. Thermal boundaries include high-temperature walls and low-temperature walls. Material parameter information is also obtained, including fluid parameters, solid parameters, and special material parameters. Fluid parameters include density, viscosity, thermal conductivity, and specific heat capacity; solid parameters include elastic modulus, Poisson's ratio, and coefficient of thermal expansion; and special material parameters include anisotropic parameters, such as piezoelectricity. A meshing strategy is determined based on the geometric partitioning information. The fluid domain uses a boundary layer mesh, and the solid domain uses a tetrahedral or hexahedral mesh. Mesh refinement is required in areas with large thermal gradients.

[0029] Based on the physics field interface, coupling logic, boundary conditions, geometric partitioning information, material parameter information, and mesh strategy, the solver and output variables are configured. These parameters are input into the physics field modeling software. The solver configuration can be divided into transient and steady-state problems. Transient problems require small time steps, while steady-state problems can be solved by first solving single physics fields and then gradually coupling them. Output variables include temperature field, velocity field, and stress-strain distribution. A multiphysics coupling model is established based on the solver and output variables. An initial model is generated using the configured solver and output variables. The initial model can be verified for coupling effects in simulation software, such as comparing the stress differences due to thermal expansion. Based on the coupling effect verification results, the initial model is corrected to obtain a fluid-solid-thermal multiphysics coupling model, i.e., a multiphysics coupling model.

[0030] Based on the multiphysics coupling model, the interaction relationships of temperature, stress, and flow fields for each shoe sole mold model are determined. Specifically, the dynamic interaction relationships between these fields are solved using the multiphysics coupling model, and a multidimensional variable dataset is constructed based on these relationships. This dataset includes parameters such as pressure, strain, and heat flux. This multidimensional variable dataset provides sufficiently comprehensive and specific data support for training the backpropagation neural network.

[0031] Furthermore, determining the physical field interface, coupling logic, and boundary conditions includes: acquiring laminar interface, turbulent interface, solid mechanics interface, and heat transfer interface, and determining the physical field interface based on the laminar interface, turbulent interface, solid mechanics interface, and heat transfer interface; acquiring fluid-structure interaction, heat-fluid interaction, and heat-structure interaction, and generating coupling logic based on the fluid-structure interaction, heat-fluid interaction, and heat-structure interaction; acquiring fluid inlet, fluid outlet, solid fixed constraint, heat source, and heat sink, determining the coupling boundary based on the arbitrary Lagrange-Euler (ALE) method, and determining the boundary conditions based on the coupling boundary, fluid inlet, fluid outlet, solid fixed constraint, heat source, and heat sink.

[0032] Specifically, the system acquires laminar flow, turbulent flow, solid mechanics, and heat transfer interfaces. These interfaces are the core interfaces of the multiphysics coupling model, forming the physical field interface. It also acquires fluid-structure interaction, thermal-fluid interaction, and thermal-structure interaction. Fluid-structure interaction enables bidirectional interaction between fluid pressure and structural deformation through the fluid-multibody interaction interface, such as in valve motion simulation. Thermal-fluid interaction refers to using the fluid velocity field as a convection term in the heat transfer interface, and thermal-structure interaction refers to enabling the thermal expansion effect in solid mechanics. The coupling logic consists of fluid-structure interaction, thermal-fluid interaction, and thermal-structure interaction. The fluid inlet, fluid outlet, solid fixed constraints, heat source, and heat sink are obtained. The fluid inlet and fluid outlet can represent velocity and pressure boundaries, respectively. The coupling boundary is determined based on the Arbitrary Lagrange-Euler (ALE) method. The ALE method combines the advantages of the Lagrangian and Eulerian descriptions and is suitable for deformable body modeling in continuum mechanics. Its core idea is to dynamically adjust the mesh deformation in numerical simulation to match the material motion state, thereby improving computational accuracy and stability. By setting the coupling boundary through the ALE method, the deformed mesh can be tracked, and the boundary conditions are determined based on the coupling boundary, fluid inlet, fluid outlet, solid fixed constraints, heat source, and heat sink.

[0033] S203: Train the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network; In the specific implementation of this invention, training the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network includes: cleaning and labeling the multidimensional variable dataset to obtain a labeled multidimensional variable dataset; and training the backpropagation neural network based on the labeled multidimensional variable dataset to obtain a trained backpropagation neural network.

[0034] Specifically, the multidimensional variable dataset undergoes data cleaning and labeling. Data cleaning includes removing abnormal characters or whitespace characters from the dataset, and data labeling includes labeling different types of data in the dataset to obtain a labeled multidimensional variable dataset. A backpropagation neural network is then trained based on this labeled dataset to obtain a trained backpropagation neural network. The backpropagation algorithm is used for training. The backpropagation neural network includes an input layer, an output layer, and hidden layers. The number of neurons in the input layer equals the dimension of the input data, the number of neurons in the output layer equals the dimension of the output control data, and the number of neurons in the hidden layers is 5-10.

[0035] S204: Based on the trained backpropagation neural network, process parameter information is generated using the shoe sole mold model information; In the specific implementation of this invention, the process parameter information generated by using the shoe sole mold model information based on the trained backpropagation neural network includes: inputting the shoe sole mold model information into the trained backpropagation neural network, and the trained backpropagation neural network performing process parameter analysis based on the shoe sole mold model information using neuron connection weights, bias values ​​and activation functions to obtain process parameter information.

[0036] Specifically, the shoe sole mold model information is input into a trained backpropagation neural network. The trained backpropagation neural network analyzes process parameters based on the shoe sole mold model information using neuron connection weights, bias values, and activation functions. The activation function can be a Sigmoid function. The backpropagation neural network automatically calculates the set values ​​of the process parameters according to the neuron connection weights, bias values, and Sigmoid function, thus obtaining the process parameter information. The process parameter information includes raw material ratio, injection weight, injection time, injection pressure, injection speed, heating time, etc.

[0037] S205: The automatic disc machine aligns the robotic arm with the shoe sole mold on the workstation, and the spray foaming machine delivers non-foamed polyurethane to the spray head of the robotic arm. In the specific implementation of this invention, the automatic disc machine adopts a ring layout design with multiple workstations evenly distributed around its circumference. Automatic switching between workstations can be achieved through a rotating mechanism. Each workstation has a fixed shoe sole mold. The automatic disc machine rotates at a preset angle, aligning the robotic arm with the shoe sole mold at the workstation. Based on process parameters, the raw material ratio is determined. The spray foaming machine mixes two polyurethane raw materials according to the ratio to form non-foamed polyurethane, which is then conveyed to the spray head of the robotic arm. The spray head is fixed to the end of the robotic arm and used to spray the non-foamed polyurethane onto the shoe mold.

[0038] S206: Based on the process parameter information and combined with the PID feedback regulator, the spraying head of the robotic arm is controlled to perform non-foamed polyurethane spraying on the surface of the shoe sole mold at the workstation to obtain a sprayed shell. In the specific implementation of this invention, real-time detection values ​​of the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting process of the shoe sole mold are obtained based on a sensor combination; the difference between the real-time detection value and the process parameter information is calculated; a feedback control quantity is generated based on the difference using a PID feedback regulator; and the operating parameters of the target equipment are adjusted in real time based on the feedback control quantity.

[0039] Specifically, based on the process parameter information, the spraying head of the robotic arm is controlled to perform non-foamed polyurethane spraying on the surface of the shoe sole mold at the workstation. This spraying of non-foamed polyurethane forms a high-performance outer shell on the mold surface, thus obtaining the sprayed shell. During the non-foamed polyurethane spraying process, real-time detection values ​​are acquired using a sensor combination, including a flow sensor and an infrared temperature sensor. This sensor combination monitors the injection weight, injection time, and injection speed of the non-foamed polyurethane during the spraying process. The difference between the real-time detection value and the process parameter information is calculated. A PID feedback controller uses this difference to generate a feedback control quantity. The PID feedback controller employs a PID algorithm for state feedback regulation, using proportional, integral, and derivative coefficients combined with the difference to generate the feedback control quantity. Based on this feedback control quantity, the operating parameters of the target equipment are adjusted in real-time. Figure 4 As shown, the feedback control quantity is input into the actuator of the target device to adjust the operating parameters of the device. After the operating parameters of the target device are adjusted, the sensor combination continues to work, forming a control closed loop.

[0040] S207: Based on the process parameter information, the screw feeder and hot air generator are controlled by the PID feedback regulator to fill the shoe sole mold cavity at the workstation with thermoplastic polyurethane foam elastomer to obtain the filled inner core. In the specific implementation of this invention, the screw feeder is controlled according to process parameters to fill the shoe sole mold cavity at the workstation with thermoplastic polyurethane foam elastomer. The mold cover has evenly distributed vent pipes. The filled material continues to be ventilated and heated. Simultaneously, a hot air generator activates steam to heat the shoe sole mold cover and injects hot air into the steam vent holes of the mold studs. The hot air generator generates hot air at approximately 80-150 degrees Celsius to heat the material at the end of the screw conveyed by the screw and the thermoplastic polyurethane foam elastomer filling the shoe mold. After the thermoplastic polyurethane foam elastomer filling is complete, the filled inner core is obtained. During the thermoplastic polyurethane foam elastomer filling process, real-time detection values ​​of the thermoplastic polyurethane foam elastomer filling process of the shoe sole mold are obtained based on a sensor combination. The sensor combination includes a flow sensor, an infrared temperature sensor, etc., that is, the injection weight, injection time, heating time, and injection speed of the thermoplastic polyurethane foam elastomer filling process of the shoe sole mold are detected in real time through the sensor combination. The difference between the real-time detection value and the process parameter information is calculated. Based on the PID feedback regulator, the difference is used to generate a feedback control quantity. The PID feedback regulator uses the PID algorithm for state feedback regulation. It generates the feedback control quantity by combining the proportional coefficient, integral coefficient and derivative coefficient with the difference, and adjusts the operating parameters of the target equipment in real time based on the feedback control quantity.

[0041] S208: Based on the process parameter information and combined with the PID feedback regulator, the automatic opening and closing mechanism of the casting mold is controlled to melt and bond the sprayed outer shell and the filling inner core to obtain a composite material shoe sole.

[0042] In the specific implementation of this invention, the melting and bonding process refers to pressing and steam heating to expand the inner core, bonding the inner core to the outer shell. The automatic opening and closing mechanism of the casting mold slowly and gradually closes the sole mold. When the upper and lower molds are pressed together, steam is discharged from the vent holes of the mold pins into the lower mold filled with thermoplastic polyurethane foam elastomer. The hot steam and thermoplastic polyurethane foam elastomer work together to expand and fill the inner core and bond it to the sprayed outer shell. After the set mold closing time, the automatic opening and closing mechanism of the casting mold automatically opens the mold and shuts off the steam. At the same time, a PID feedback regulator is used for real-time adjustment and control during the melting and bonding process. Finally, a composite material sole combining non-foamed polyurethane spraying and thermoplastic polyurethane foam elastomer is obtained. This composite material sole is aesthetically pleasing, lightweight, and combines beauty with wearing comfort. It has high physical properties such as flexural resistance, wear resistance, slip resistance, and high elasticity, while using less material and being economical.

[0043] In this embodiment of the invention, a multiphysics coupling model is established to construct a multidimensional variable dataset. A backpropagation neural network is trained based on this dataset to obtain a trained backpropagation neural network. This trained backpropagation neural network then generates process parameter information using the shoe sole mold model information, effectively improving the accuracy and efficiency of process parameter analysis and significantly reducing reliance on manual labor. The target equipment, based on the process parameter information and combined with a PID feedback regulator, performs non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole mold. This effectively improves the precision of shoe sole manufacturing. Furthermore, by performing non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole, the resulting composite material shoe sole is not only aesthetically pleasing and lightweight but also possesses excellent physical properties such as flexural resistance, abrasion resistance, slip resistance, and high resilience.

[0044] Example 3 Please see Figure 3 , Figure 3 This is a schematic diagram of the structural composition of a composite material shoe sole manufacturing device according to an embodiment of the present invention. The device includes: Shoe sole mold model identification module 31: used to identify shoe sole mold models based on radio frequency identification technology and obtain shoe sole mold model information; Dataset construction module 32: used to establish a multiphysics coupling model and construct a multidimensional variable dataset based on the multiphysics coupling model; Network training module 33: used to train the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network; Parameter generation module 34: used to generate process parameter information based on the trained backpropagation neural network using the shoe sole mold model information; Shoe sole manufacturing module 35: This module is used by the target equipment to perform non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole mold based on the process parameter information and a proportional-integral-derivative PID feedback regulator, in order to obtain a composite material shoe sole. The PID feedback regulator is used to adjust the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment processes of the shoe sole mold in real time.

[0045] In the specific implementation of this invention, the specific implementation of the device item can be referred to the implementation of the method item above, and will not be repeated here.

[0046] In this embodiment of the invention, a multiphysics coupling model is established to construct a multidimensional variable dataset. A backpropagation neural network is trained based on this dataset to obtain a trained backpropagation neural network. This trained backpropagation neural network then generates process parameter information using the shoe sole mold model information, effectively improving the accuracy and efficiency of process parameter analysis and significantly reducing reliance on manual labor. The target equipment, based on the process parameter information and combined with a PID feedback regulator, performs non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole mold. This effectively improves the precision of shoe sole manufacturing. Furthermore, by performing non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole, the resulting composite material shoe sole is not only aesthetically pleasing and lightweight but also possesses excellent physical properties such as flexural resistance, abrasion resistance, slip resistance, and high resilience.

[0047] This invention provides a computer-readable storage medium storing a computer program. When executed by a processor, the program implements the composite material shoe sole manufacturing method of any of the above embodiments. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic cards, or optical cards. In other words, the storage device includes any medium that can store or transmit information in a readable form by a device (e.g., a computer, a mobile phone), and can be a read-only memory, a disk, or an optical disk, etc.

[0048] Furthermore, the above provides a detailed description of a composite material shoe sole manufacturing method and related apparatus provided by the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for manufacturing a composite material shoe sole, characterized in that, The method includes: Shoe sole mold model identification is performed based on radio frequency identification technology to obtain shoe sole mold model information; Establish a multiphysics coupling model and construct a multidimensional variable dataset based on the multiphysics coupling model; The backpropagation neural network is trained based on the multidimensional variable dataset to obtain a trained backpropagation neural network. Based on the trained backpropagation neural network, process parameter information is generated using the shoe sole mold model information; The target equipment uses a proportional-integral-derivative (PID) feedback regulator based on the process parameter information to perform non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the shoe sole mold to obtain a composite material shoe sole. The PID feedback regulator is used to adjust the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment processes of the shoe sole mold in real time.

2. The method for manufacturing composite material shoe soles according to claim 1, characterized in that, The process of identifying the shoe sole mold model based on radio frequency identification (RFID) technology to obtain shoe sole mold model information includes: Obtain the radio frequency identification (RFID) tag of the shoe sole mold, and identify the shoe sole mold model based on the RFID tag to obtain the shoe sole mold model information.

3. The method for manufacturing composite material shoe soles according to claim 1, characterized in that, The establishment of a multiphysics coupling model and the construction of a multidimensional variable dataset based on the multiphysics coupling model include: Determine the physical field interface, coupling logic, and boundary conditions; Geometric partitioning information is determined based on fluid domain, solid domain, and thermal boundary, and material parameter information is obtained; Determine the meshing strategy based on the geometric partitioning information; Based on the physical field interface, coupling logic, boundary conditions, geometric meshing information, material parameter information, and mesh strategy, the solver and output variables are configured, and a multiphysics coupling model is established based on the solver and output variables. Based on the multiphysics coupling model, the interaction relationship between the temperature field, stress field and flow field of each shoe sole mold model is determined, and a multidimensional variable dataset is constructed based on the interaction relationship.

4. The method for manufacturing composite material shoe soles according to claim 3, characterized in that, The determination of the physical field interface, coupling logic, and boundary conditions includes: Obtain the laminar flow interface, turbulent flow interface, solid mechanics interface, and heat transfer interface, and determine the physical field interface based on the laminar flow interface, turbulent flow interface, solid mechanics interface, and heat transfer interface; Obtain fluid-structure interaction, thermal-fluid interaction, and thermal-structure interaction, and generate coupling logic based on the fluid-structure interaction, thermal-fluid interaction, and thermal-structure interaction; Obtain the fluid inlet, fluid outlet, solid fixed constraint, heat source, and heat sink; determine the coupling boundary based on the arbitrary Lagrange-Euler (ALE) method; and determine the boundary conditions based on the coupling boundary, fluid inlet, fluid outlet, solid fixed constraint, heat source, and heat sink.

5. The method for manufacturing composite material shoe soles according to claim 1, characterized in that, The step of training the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network includes: The multidimensional variable dataset is cleaned and labeled to obtain a labeled multidimensional variable dataset; The backpropagation neural network is trained based on the labeled multidimensional variable dataset to obtain a trained backpropagation neural network.

6. The method for manufacturing composite material shoe soles according to claim 1, characterized in that, The process parameter information generated using the shoe sole mold model information based on the trained backpropagation neural network includes: The shoe sole mold model information is input into a trained backpropagation neural network. The trained backpropagation neural network analyzes process parameters based on the shoe sole mold model information using neuron connection weights, bias values, and activation functions to obtain process parameter information.

7. The method for manufacturing composite material shoe soles according to claim 1, characterized in that, The target equipment, based on the process parameter information and combined with a proportional-integral-derivative PID feedback regulator, performs non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the sole mold to obtain a composite material sole, including: The automatic rotary machine aligns the robotic arm with the shoe sole mold at the workstation, and the spray foaming machine delivers non-foamed polyurethane to the spray head of the robotic arm. Based on the process parameter information and combined with the PID feedback regulator, the spraying head of the robotic arm is controlled to perform non-foamed polyurethane spraying on the surface of the shoe sole mold at the workstation to obtain a sprayed shell. Based on the process parameter information, the screw feeder and hot air generator are controlled by a PID feedback regulator to fill the shoe sole mold cavity at the workstation with thermoplastic polyurethane foam elastomer to obtain a filled inner core. Based on the process parameter information and combined with the PID feedback regulator, the automatic opening and closing mechanism of the casting mold is controlled to melt and bond the sprayed outer shell and the filling inner core to obtain a composite material shoe sole.

8. The method for manufacturing composite material shoe soles according to claim 1, characterized in that, The real-time adjustment of the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting process of the shoe sole mold includes: Real-time detection values ​​of the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling and melting process of shoe sole mold are obtained based on a combination of sensors. The difference between the real-time detection value and the process parameter information is calculated, and a feedback control quantity is generated based on the difference using a PID feedback regulator. The operating parameters of the target equipment are then adjusted in real time based on the feedback control quantity.

9. A composite material shoe sole manufacturing apparatus, characterized in that, The device includes: Shoe sole mold model identification module: used to identify shoe sole mold models based on radio frequency identification technology and obtain shoe sole mold model information; Dataset construction module: used to establish a multiphysics coupling model and construct a multidimensional variable dataset based on the multiphysics coupling model; Network training module: used to train the backpropagation neural network based on the multidimensional variable dataset to obtain a trained backpropagation neural network; Parameter generation module: used to generate process parameter information based on the trained backpropagation neural network and the shoe sole mold model information; The sole manufacturing module is used by the target equipment to perform non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment on the sole mold based on the process parameter information and a proportional-integral-derivative PID feedback regulator to obtain a composite material sole. The PID feedback regulator is used to adjust the non-foamed polyurethane spraying, thermoplastic polyurethane foam elastomer filling, and melting treatment process of the sole mold in real time.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on an electronic device, cause the electronic device to perform the composite material sole manufacturing method as described in any one of claims 1 to 8.

Images