Methods, equipment, media, and products for determining the curing process curve of composite materials
By combining deep Q-networks and long short-term memory networks, the curing state of composite materials is monitored in real time, and the temperature change is adjusted autonomously. This solves the problem of the difficulty in dynamically adjusting the curing process curve of composite materials, and realizes real-time optimization of the process and improvement of quality and efficiency.
Patent Information
- Application Number
- CN202611551951.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-09
- Publication Date
- 2026-11-10
AI Technical Summary
Existing composite material curing process curves are difficult to dynamically adjust, leading to problems such as excessive local temperature gradients, residual stress accumulation, and long manufacturing cycles.
By combining deep Q-networks with long short-term memory networks, the curing state parameters of the composite material are monitored in real time, and the temperature change is adjusted autonomously to achieve dynamic optimization of the process curve.
It achieves a synergistic improvement in the quality and efficiency of the composite material curing process, reduces computational costs, adapts to the nonlinear evolution characteristics of the composite material curing process, and improves manufacturing efficiency and molding quality.
Smart Images

Figure CN123069922A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of composite material manufacturing, and in particular to a method, equipment, medium, and product for determining the curing and molding process curve of composite materials. Background Technology
[0002] Carbon fiber reinforced polymer (CFRP) composites are widely used in aerospace vehicle structure manufacturing due to their high specific strength, high specific stiffness, and excellent fatigue resistance. During the autoclave curing process of CFRP, the externally applied temperature process profile directly affects the resin curing reaction rate, the internal temperature distribution of the component, the evolution of the degree of curing, and the formation of residual stress. Therefore, rationally designing the curing process profile is of great significance for improving the molding quality and manufacturing efficiency of composite components.
[0003] Currently, composite material curing processes mainly employ fixed heating-holding-cooling process curves based on experience. These processes typically determine multiple isothermal plateaus according to parameters recommended by the material supplier, ensuring the final degree of curing meets requirements to achieve component molding. However, the composite material curing process involves various nonlinear factors such as heat conduction, curing exothermics, resin rheology, and material property evolution. The internal state of the component continuously changes over time, making it difficult to dynamically adjust the fixed process curve according to the actual curing state. This can easily lead to problems such as excessive local temperature gradients, residual stress accumulation, and long manufacturing cycles.
[0004] In recent years, optimization methods for composite material curing processes have mainly included parameter optimization methods based on numerical simulation and process optimization methods based on intelligent algorithms. Among them, the finite element simulation method can describe the evolution of temperature field, degree of curing field, and stress field during the curing process. However, since the composite material curing process involves complex multi-physics coupling, a single calculation usually takes a long time, which is difficult to meet the needs of a large number of iterative calculations in the optimization process.
[0005] Intelligent optimization methods based on genetic algorithms and multi-objective optimization algorithms can improve traditional process curve design methods to a certain extent. However, these methods usually require pre-defining a limited set of process parameters, such as heating rate, holding temperature, and holding time. The optimization results are still limited by the manually set process structure and cannot achieve continuous dynamic adjustment during the curing process.
[0006] Deep reinforcement learning methods can autonomously select control strategies based on the current state through the interaction between the agent and the environment, demonstrating significant advantages in the optimization of complex dynamic systems. However, the curing process of composite materials is highly time-dependent. If a high-precision numerical model is directly used as the reinforcement learning environment, a large amount of repetitive computation is required, resulting in low training efficiency and making it difficult to apply to practical process optimization.
[0007] Therefore, based on the above problems, there is an urgent need to provide a method for determining the curing process curve of composite materials, which can quickly predict the evolution law of the curing state of composite materials, further autonomously adjust the temperature change, and achieve a synergistic improvement in the quality and efficiency of the curing process. Summary of the Invention
[0008] The purpose of this application is to provide a method, equipment, medium, and product for determining the curing process curve of composite materials, which can quickly predict the evolution law of the curing state of composite materials, further autonomously adjust the temperature change, and achieve a synergistic improvement in the quality and efficiency of the curing process.
[0009] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a method for determining the curing and molding process curve of a composite material, including: The curing state parameters of the composite material under different process parameters are obtained; the process parameters are the temperature changes; the curing state parameters include: mold temperature, temperature at characteristic locations inside the component, degree of curing, and residual stress state. Based on the curing state parameters, the Q values corresponding to different temperature changes are determined using a deep Q-network, and the temperature change corresponding to the maximum Q value is taken as the final temperature change. Based on the final temperature change and the corresponding curing state parameters, the curing state parameters at the next moment are determined using a long short-term memory network. Based on the fixed state parameters at the next time step, the comprehensive reward value is determined based on the reward function; and the parameters of the deep Q network are updated using the comprehensive reward value to obtain the trained deep Q network. The current curing state parameters are collected in real time, and the final temperature change is determined based on the trained deep Q network. Based on the final temperature change and the current curing state parameters, the curing state parameters at the next moment are determined based on the long short-term memory network. A process curve is constructed based on the final temperature change at each moment and the corresponding curing state parameters.
[0010] Secondly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for determining the curing process curve of composite materials.
[0011] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for determining the curing process curve of composite materials.
[0012] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for determining the curing process curve of composite materials.
[0013] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, equipment, medium, and product for determining the curing process curve of composite materials. By acquiring the curing state parameters of the composite material under different process parameter conditions, and using a deep Q-network to determine the final temperature change, and using a long short-term memory network to determine the curing state parameters at the next moment, this application combines a deep Q-network with a long short-term memory network, transforming the traditional offline calculation-based process optimization method into a dynamic decision-making method based on state feedback. It can automatically adjust the temperature change according to the real-time curing state parameters, realizing closed-loop control of the curing process. Attached Figure Description
[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0015] Figure 1 This is a flowchart illustrating a method for determining the curing and molding process curve of a composite material in one embodiment of this application. Detailed Implementation
[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0018] In one exemplary embodiment, such as Figure 1 As shown, a method for determining the curing and molding process curve of a composite material is provided, including the following S1 to S6. Wherein: S1: Obtain the curing state parameters of the composite material under different process parameter conditions.
[0019] The curing process of composite materials is a dynamic evolution process with a significant time dependence. Its internal temperature, degree of curing, and stress state are influenced not only by the current temperature but also by historical temperature variations. Therefore, a time-series database containing the relationship between process parameters and the evolution of the curing state is first established. This database includes curing state parameters under different process parameter conditions. The process parameter is the change in mold temperature, referred to simply as temperature change. Based on the temperature change, the heating rate, cooling rate, and holding time are further derived. The curing state parameters include mold temperature, temperature at characteristic locations inside the component, degree of curing, and residual stress state.
[0020] Each curing process is discretized into multiple time steps, and in the first... The process parameters for each time step are: ; in, For the first Process parameters for each time step, For the first Temperature change at each time step.
[0021] The curing state parameters can be obtained through experimental testing, online monitoring, or numerical simulation methods, and the expressions are as follows: .
[0022] in, For the first Curing state parameters at each time step For the first The mold temperature at each time step For the first Temperature at characteristic locations inside the component at each time step For the first Curing degree at each time step, No. The residual stress state at each time step.
[0023] By continuously collecting curing state parameters under different process parameter conditions, a time series database is formed for training.
[0024] S2: Based on the curing state parameters, determine the Q value corresponding to different temperature changes using a deep Q network (DQN), and take the temperature change corresponding to the maximum Q value as the final temperature change.
[0025] The composite material curing process is defined as a Markov Decision Process (MDP), including a state space, action space, state transition model, and reward function. In this application, process parameters are defined as the action space, curing state parameters are defined as the state space, and a Long Short-Term Memory (LSTM) network is used as the state transition model. The action space describes the next mold temperature adjustment, and the state space describes the current stage of the curing process. The expressions for the action space and state space are as follows: ; ; in, For the action space, The change in temperature For state space, For mold temperature, Temperature at a characteristic location inside the component. For degree of curing, It is in a residual stress state.
[0026] Based on the temperature control capability of the curing equipment, the operating space is limited to Within the range, among which, The minimum temperature change This represents the maximum temperature change.
[0027] To meet the requirements of deep Q-networks for a discrete action space, the continuous temperature adjustment range is discretized into multiple temperature adjustment actions, each corresponding to a temperature change strategy, as expressed below: ; in, For discrete action space, The temperature difference between the available time steps is denoted as .
[0028] S3: Based on the final temperature change and the corresponding curing state parameters, determine the curing state parameters for the next moment using a long short-term memory network.
[0029] Because the curing process of composite materials exhibits time-dependent characteristics such as delayed thermal conduction, exothermic curing reaction, and evolution of material properties, Long Short-Term Memory (LSTM) networks are used as state transition models. LSM networks consist of input gates, forget gates, output gates, and memory units. Through a recurrent structure, they store historical curing state parameters, enabling the learning of complex time-series evolution patterns.
[0030] The Long Short-Term Memory (LSTM) network in this application is obtained by training the original LTM network using a time-series database. Specifically, process parameters and solidification state parameters are input into the original LTM network to obtain predicted values of the solidification state parameters at the next time step. These predicted values are then compared with the measured values, and the original LTM network is updated using an error loss function to obtain the final LTM network. Before constructing the deep Q-network, the LTM network serves as a reinforcement learning environment during the training process.
[0031] S3 specifically includes: S31: Combine the final temperature change and the corresponding curing state parameters to obtain the combined variable.
[0032] The combined variable expression is as follows: ; in, For the first Combination variables at each time step For the first Curing state parameters at each time step For the first Process parameters for each time step, For the first The mold temperature at each time step For the first Temperature at characteristic locations inside the component at each time step For the first Curing degree at each time step, No. The residual stress state at each time step For the first Temperature change at each time step.
[0033] S32: Based on the combined variables and the long short-term memory network, determine the change in solidified state parameters.
[0034] The change in solidified state parameters at the next time step is predicted using a long short-term memory network. The expression for the change in solidified state parameters is as follows: ; in, For the first The change in curing state parameters at each time step For the first Temperature change at characteristic locations inside the component at each time step For the first The change in curing degree at each time step For the first The change in residual stress state at each time step.
[0035] S33: Determine the curing state parameters at the next moment based on the curing state parameters and the changes in the curing state parameters.
[0036] Specifically, the expression for the solidification state parameters at the next moment is as follows: ; in, For the first Curing state parameters at each time step For the first The mold temperature at each time step For the first Temperature at characteristic locations inside the component at each time step For the first Curing degree at each time step, No. The residual stress state at each time step For the first Temperature change at each time step For the first Temperature change at characteristic locations inside the component at each time step For the first The change in curing degree at each time step For the first The change in residual stress state at each time step.
[0037] In summary, this application employs a fast state prediction model (i.e., a long short-term memory network), which eliminates the need to repeatedly call time-consuming computational models to obtain the state changes of composite materials under different temperature control strategies. The expression is as follows: ; in, It is a long short-term memory network.
[0038] S4: Based on the fixed state parameters at the next time step, determine the comprehensive reward value based on the reward function; and use the comprehensive reward value to update the parameters of the deep Q network to obtain the trained deep Q network.
[0039] Specifically, the curing state parameters are first normalized to obtain normalized curing state parameters. Then, the comprehensive reward value (reward function) is determined based on the normalized curing state parameters. The calculation formula is as follows: ; in, For the overall reward value, Temperature weighting, This represents the normalized internal temperature gradient of the component. This is a temperature uniformity evaluation item, used to reduce the temperature difference between the inside and outside of the component during the curing process. , For residual stress weight, The normalized maximum residual stress, This is a residual stress evaluation item used to reduce the maximum residual stress in the component during the curing process. , As time weight, This represents the total time of the normalized curing process. This is a curing cycle evaluation item used to reduce the total curing time. , As the curing degree weight, For the final degree of curing, To solidify the completion of the reward item, when the final solidification degree reaches the set requirement ( (This will grant additional rewards.)
[0040] After normalization, all dimensions in the reward function are 1. Since the degree of solidification itself has a dimension of 1, no normalization is needed. When performing deep Q-network reinforcement learning, setting all weights to be equal yields a more balanced optimization effect.
[0041] Based on the overall reward value, the parameters of the deep Q-network are updated using an experience replay mechanism. Specifically, this is done according to the target parameters of the deep Q-network. The Bellman target value is calculated using the following formula: ; in, For Bellman's target value, The deep Q-network is in a fixed state. Take action when temperature changes The value (Q value) This is the discount factor.
[0042] Based on the current parameters of the deep Q network Sure Compared with Bellman's target value The squared error between the two values is used to construct the loss function, and the calculation formula is as follows: in, Representative from experience pool The mini-batch samples are drawn according to uniform distribution. ; This represents the mathematical expectation of a uniform sampling distribution in the experience replay pool. In actual training, this expectation is approximated by calculating the average loss on randomly selected small batches of experience samples.
[0043] The parameters of the current depth Q-network are then updated using gradient descent, calculated as follows: ; in, The learning rate is used for the deep Q-network. By repeatedly performing empirical sampling, Bellman target calculation, and deep Q-network parameter updates, the current Q-network gradually improves the estimation accuracy of the long-term cumulative benefits of various temperature regulation actions under different curing states.
[0044] The formula for the deep Q-network training process in this application comes from the paper "human-level control through deep reinforcement learning" in the journal Nature.
[0045] Through multiple rounds of iterative training, the deep Q-network acquires a dynamic temperature regulation strategy that can adapt to different curing states, thereby reducing the temperature gradient, minimizing residual stress, and shortening the manufacturing cycle.
[0046] S5: Real-time acquisition of current curing state parameters; determination of the final temperature change based on a trained deep Q-network; determination of the next curing state parameters based on the final temperature change and current curing state parameters using a long short-term memory network.
[0047] Specifically, the trained deep Q-network and long short-term memory network are deployed to collect current solidified state parameters in real time, and the trained deep Q-network is used to calculate the value function corresponding to different actions (different temperature changes). (Q value), and select the action corresponding to the maximum value (maximum Q value) as the current final action. That is, temperature adjustment amount It is used to control the equipment to adjust the heating, heat preservation, or cooling process. The current final action and the current curing state (curing state parameters) are input into the long short-term memory network to obtain the curing state at the next moment. .
[0048] S6: Construct a process curve based on the final temperature change at each moment and the corresponding curing state parameters.
[0049] The closed-loop control process of repeatedly executing status acquisition → action decision → temperature adjustment → status update is ultimately used to obtain a real-time optimized curing process curve.
[0050] In an exemplary embodiment, based on the characteristics of the composite material curing process, curing state data samples (curing state parameters) under different process conditions are established. The curing state data samples are divided into training and validation sets to train an initial Long Short-Term Memory (LSTM) network, resulting in a fast prediction model capable of describing the dynamic evolution of the composite material's curing state. The LSTM network is used as a reinforcement learning environment, defining a state space and an action space. To meet the temperature control requirements of actual curing equipment, the action range is constrained. Actions are discretized to enable the deep Q-network to select actions. The curing state parameters are initialized, and the deep Q-network calculates the Q-value corresponding to each action based on the current curing state parameters. The action corresponding to the largest Q-value is selected and input into the LSTM network to obtain the curing state parameters for the next time step. A comprehensive reward value is calculated based on the curing state parameters for the next time step; the deep Q-network parameters are updated using an experience replay pool. After multiple rounds of training, the deep Q-network can obtain the optimal temperature control strategy.
[0051] By utilizing a trained deep Q-network to output a sequence of temperature adjustment actions corresponding to continuous time steps, a fully optimized curing process curve is obtained. This process curve is not limited by the traditional fixed insulation platform but dynamically adjusts according to changes in the internal state of the composite material. A long short-term memory network (LSTM) and the trained deep Q-network are deployed in the composite material molding control system. During curing, temperature sensors, dielectric detection devices, or other state monitoring devices acquire real-time information on mold temperature, component internal temperature, and curing state. This collected data is input into the trained deep Q-network, which outputs the next-stage temperature adjustment strategy based on the real-time state and controls the heating equipment to execute it. Through continuous cycling of state acquisition → action decision → temperature adjustment → state update, real-time intelligent optimization of the composite material curing process is achieved.
[0052] This application establishes a time-series database to obtain the mapping relationship between temperature regulation actions and the evolution of curing state parameters under different process parameter conditions. A long short-term memory (LSTM) network is used to establish a rapid prediction model for the curing state, enabling rapid prediction of the curing state parameters from the current curing state parameters and temperature regulation actions to the next time step. The LTM network is used as a reinforcement learning environment model, and a deep reinforcement learning algorithm is introduced to achieve real-time optimization of the curing process curve. Compared with traditional fixed process curve optimization methods, this application does not limit the heating-holding-cooling curve form, but dynamically adjusts the temperature change process according to the actual curing state of the material, enabling the process curve to adapt to the nonlinear evolution characteristics of the composite material curing process.
[0053] In summary, this application has significant advantages over existing technologies. First, by learning the time-series evolution of the curing process, this application achieves rapid mapping from process temperature changes to internal temperature, degree of curing, and stress state. Compared to directly using high-precision numerical models for optimization calculations, this reduces computational costs and improves optimization efficiency. Second, this application employs deep reinforcement learning to dynamically optimize the curing process curve. It does not require pre-setting a fixed heating-holding-cooling curve; instead, it autonomously adjusts its temperature change strategy based on the current curing state of the composite material, enabling the optimized process curve to adapt to the nonlinear evolution characteristics of the curing process. Third, by constructing a reward function that comprehensively considers temperature uniformity, residual stress, and manufacturing cycle, this application enables the agent to achieve synergistic optimization of molding quality and manufacturing efficiency while ensuring the final degree of curing meets requirements. Furthermore, this application uses a combination of long short-term memory networks and deep Q-networks to transform the traditional offline computation-based process optimization method into a dynamic decision-making method based on state feedback, enabling closed-loop control of the curing process based on real-time monitoring data. In addition, this application has strong versatility and can be applied to the optimization of composite material molding processes under different material systems, different structural forms and different curing equipment conditions. It can be applied to autoclave curing, compression molding and automated lay-up composite material manufacturing processes.
[0054] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores composite material curing process curve data. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for determining a composite material curing process curve.
[0055] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
[0056] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0057] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0058] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.
[0059] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
[0060] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, etc., and are not limited to these.
[0061] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0062] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method for determining the curing and molding process curve of a composite material, characterized in that, The method includes: The curing state parameters of the composite material under different process parameters are obtained; the process parameters are the temperature changes; the curing state parameters include: mold temperature, temperature at characteristic locations inside the component, degree of curing, and residual stress state. Based on the curing state parameters, the Q values corresponding to different temperature changes are determined using a deep Q-network, and the temperature change corresponding to the maximum Q value is taken as the final temperature change. Based on the final temperature change and the corresponding curing state parameters, the curing state parameters at the next moment are determined using a long short-term memory network. Based on the fixed state parameters at the next time step, the comprehensive reward value is determined based on the reward function; and the parameters of the deep Q network are updated using the comprehensive reward value to obtain the trained deep Q network. The current curing state parameters are collected in real time, and the final temperature change is determined based on the trained deep Q network. Based on the final temperature change and the current curing state parameters, the curing state parameters at the next moment are determined based on the long short-term memory network. A process curve is constructed based on the final temperature change at each moment and the corresponding curing state parameters.
2. The method for determining the curing and molding process curve of composite materials according to claim 1, characterized in that, The acquisition of the curing state parameters of the composite material under different process parameter conditions specifically includes: Using formula Determine process parameters; Using formula Determine the curing state parameters; in, For the first Process parameters for each time step, For the first Temperature change at each time step For the first Curing state parameters at each time step For the first The mold temperature at each time step For the first Temperature at characteristic locations inside the component at each time step For the first Curing degree at each time step, No. The residual stress state at each time step.
3. The method for determining the curing and molding process curve of composite materials according to claim 1, characterized in that, The step of determining the curing state parameters for the next moment based on the final temperature change and the corresponding curing state parameters, using a long short-term memory network, specifically includes: The final temperature change and the corresponding curing state parameters are combined to obtain the combined variables; Based on the combined variables and the long short-term memory network, the changes in solidified state parameters are determined. Based on the curing state parameters and the changes in the curing state parameters, determine the curing state parameters at the next moment.
4. The method for determining the curing and molding process curve of composite materials according to claim 3, characterized in that, The process of combining the final temperature change and the corresponding curing state parameters to obtain combined variables specifically includes: Using formula Obtain the combined variables; in, For the first Combination variables at each time step For the first Curing state parameters at each time step For the first Process parameters for each time step, For the first The mold temperature at each time step For the first Temperature at characteristic locations inside the component at each time step For the first Curing degree at each time step, No. The residual stress state at each time step For the first Temperature change at each time step.
5. The method for determining the curing and molding process curve of composite materials according to claim 3, characterized in that, The determination of the changes in solidified state parameters based on the combined variables and the long short-term memory network specifically includes: Using formula Determine the changes in curing state parameters; in, For the first The change in curing state parameters at each time step For the first Temperature change at characteristic locations inside the component at each time step For the first The change in curing degree at each time step For the first The change in residual stress state at each time step.
6. The method for determining the curing and molding process curve of composite materials according to claim 3, characterized in that, The step of determining the curing state parameters at the next moment based on the curing state parameters and the changes in the curing state parameters specifically includes: Using formula Determine the curing state parameters for the next moment; in, For the first Curing state parameters at each time step For the first The mold temperature at each time step For the first Temperature at characteristic locations inside the component at each time step For the first Curing degree at each time step, No. The residual stress state at each time step For the first Temperature change at each time step For the first Temperature change at characteristic locations inside the component at each time step For the first The change in curing degree at each time step For the first The change in residual stress state at each time step.
7. The method for determining the curing and molding process curve of composite materials according to claim 1, characterized in that, The step of determining the comprehensive reward value based on the solidified state parameters at the next moment and the reward function specifically includes: Using formula Determine the overall reward value; in, For the overall reward value, Temperature weighting, This represents the normalized internal temperature gradient of the component. For residual stress weight, The normalized maximum residual stress, As time weight, This represents the total time of the normalized curing process. As the curing degree weight, This represents the final degree of curing.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for determining the composite material curing molding process curve according to any one of claims 1-7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for determining the curing process curve of composite materials as described in any one of claims 1-7.
10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for determining the curing process curve of composite materials as described in any one of claims 1-7.
