A control system and method for oiling the surface of a fuel tank housing stamping part

Abstract

The present invention discloses a control system and a control method for the surface oiling of a fuel tank housing stamping part, which relates to the technical field of intelligent manufacturing control systems. The method includes the following steps: first, extract key process parameters in the oiling process from historical data, and establish the correlation between the parameters to form a knowledge graph; on the basis of establishing the knowledge graph, the reinforcement learning model is initialized based on the associated data in the knowledge graph; the collected real-time data will be input into the knowledge graph and the reinforcement learning model, and the knowledge graph and the reinforcement learning model will be analyzed collaboratively to jointly determine the oiling strategy. By utilizing the reinforcement learning algorithm, the present invention enables the system to autonomously learn and optimize the oiling strategy, reduces manual intervention, and significantly improves production efficiency and product quality; through the application of the knowledge graph, the system can generate the best oiling strategy that adapts to different production conditions based on comprehensive historical data and real-time data.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent manufacturing control systems, and in particular to a control system for oiling the surface of a fuel tank housing stamping part and a control method thereof. Background Art

[0002] As the modern manufacturing industry continues to increase its requirements for precision machining and automation, the surface treatment process of metal stamping parts also requires more intelligent management and optimization. In the production process of fuel tank housing stamping parts, surface oiling is one of the key steps, which directly affects the surface quality, anti-corrosion performance of the product and the smooth progress of subsequent processing links.

[0003] Traditional oiling methods usually rely on manual operation or simple automated equipment for control, which is difficult to accurately adapt to changes in different materials, environments and process conditions. This low-intelligence oiling method often leads to excessive or insufficient oiling, which affects the surface quality of metal parts, increases production costs, and has adverse effects on subsequent stamping processes and product performance. In addition, the adjustment of process parameters depends on the experience of workers and lacks effective decision support and automatic optimization mechanisms, which makes it difficult to ensure the repeatability and stability of the process. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above problems existing in the prior art, the present invention is proposed.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] A method for controlling the oiling of a fuel tank housing stamping part surface, characterized in that the method comprises the following steps:

[0008] Step 1: First, extract the key process parameters in the oiling process from the historical data, and establish the correlation between the parameters to form a knowledge graph;

[0009] Step 2: On the basis of establishing the knowledge graph, the reinforcement learning model is initialized based on the associated data in the knowledge graph;

[0010] Step 3: The collected real-time data will be input into the knowledge graph and reinforcement learning model. The knowledge graph and reinforcement learning model will jointly analyze and determine the oiling strategy.

[0011] Step 4: Establish a feedback update mechanism. After each oiling process, the system will generate feedback data based on the oiling effect. The feedback data will be updated to the knowledge graph to enrich the content of the knowledge graph. As the knowledge graph is continuously updated, the reinforcement learning model will automatically acquire new data for learning to ensure the latest and comprehensiveness of its strategy.

[0012] As a preferred solution of the method for controlling the surface oiling of a fuel tank housing stamping part of the present invention, the method for constructing the knowledge graph in step 1 is:

[0013] S101: extract key process parameters, including: material type M, oiling time T, stamping pressure P, and oiling amount O;

[0014] S102: Analyze the multivariate relationship in the historical data and establish a knowledge graph for expressing parameter correlation:

[0015] Among them, K(M,T,P,O) represents the correlation between key parameters in the knowledge graph, which serves as the input basis of the reinforcement learning model; f1(M i ,t) represents the influence of material type at different time points; f2(P i ) is the influence of stamping pressure on the oiling effect, which is modeled by the physical properties of stamping; α is the time decay coefficient, ln(1+O i ) represents the logarithmic relationship between the amount of oil applied and its effect.

[0016] As a preferred solution of the method for controlling the surface oiling of a fuel tank housing stamping part of the present invention, the influence f1(M i ,t) is calculated as:

[0017] f1(M i ,t)=α 0i +α 1i t+α 2i t 2 +…+α ni t n

[0018] Among them, α 0i ,α 1i ,α 2i ,…α ni For material type M i The relevant polynomial coefficients can be obtained by fitting historical data;

[0019] The influence of stamping pressure on the oiling effect f2(P i ) is calculated as:

[0020]

[0021] Among them, β, γ, and δ are physical coefficients related to materials and pressure, which respectively represent the influence of different physical properties.

[0022] As a preferred solution of the method for controlling the surface oiling of a fuel tank housing stamping part described in the present invention, the goal of reinforcement learning is to maximize the expected return of strategy π in the environment, and to obtain the associated data extracted from the knowledge graph as the input of the model:

[0023]

[0024] where π * (X) represents the optimal strategy generated by reinforcement learning, X(t) is the feature vector input of real-time process parameters, μ t Represents the discount factor, which describes the discounting effect of future returns.

[0025] As a preferred solution of the method for controlling the surface oiling of a fuel tank housing stamping part of the present invention, the feedback update mechanism includes:

[0026] S401: Within a certain period of time, collect the following feedback data: Oiling effect O eff :Measure the uniformity, thickness and other physical characteristics of the oiled surface; quality pass rate Q eff : Indicates the qualified proportion of oil-coated products;

[0027] S402: After the feedback data is generated, the knowledge graph will be updated, and its expression is:

[0028]

[0029] Among them, g1(O eff ) is used to measure the contribution of oiling quality to the knowledge graph; g2(O eff )Quality pass rate Q eff Contribution to knowledge graph updates;

[0030] S403: After the knowledge graph is updated, the reinforcement learning model will automatically acquire new data for learning to ensure the latest and comprehensiveness of the oiling strategy. The update formula is:

[0031]

[0032] Among them, π new * According to the definition of strategy space, the range of (X) is a multidimensional vector, reflecting the oiling strategy selection at different time steps.

[0033] As a preferred solution of the method for controlling the surface oiling of a fuel tank housing stamping part of the present invention, wherein:eff ) is calculated as:

[0034] g1(O eff )=θ·log(1+O eff )

[0035] Among them, θ is the adjustment coefficient of the influence of adjusting the oiling effect.

[0036] As a preferred solution of the method for controlling the surface oiling of a fuel tank housing stamping part of the present invention, wherein: eff ) is calculated as:

[0037]

[0038] Here, θ2 is an adjustment coefficient for adjusting the influence of the oiling effect.

[0039] A control system applied to the above-mentioned method for controlling the surface oiling of the fuel tank housing stamping part, the system comprising:

[0040] The historical data analysis module is used to collect, organize and analyze the historical data of the oiling process and extract key process parameters. The knowledge graph construction module builds a knowledge graph based on historical data to systematically store and manage different parameters and their relationships.

[0041] Real-time data acquisition module provides the latest process data for real-time oiling decision-making, ensuring that the decision-making process is dynamically adjusted according to the current production environment;

[0042] The reinforcement learning decision module uses the reinforcement learning algorithm to process the data obtained from the real-time data acquisition module and the knowledge graph construction module to generate the best oiling strategy;

[0043] As well as the feedback data collection and update module, by collecting feedback data, the knowledge graph is updated in real time, the database content is enriched, and it is provided to the reinforcement learning decision module to further optimize the oiling strategy.

[0044] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the above-mentioned method for controlling the surface oiling of a fuel tank housing stamping part are implemented.

[0045] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the above-mentioned method for controlling the surface oiling of a fuel tank housing stamping part are implemented.

[0046] Beneficial effects of the present invention:

[0047] 1. The present invention utilizes a reinforcement learning algorithm, and the system can autonomously learn and optimize the oiling strategy, reducing manual intervention and significantly improving production efficiency and product quality. Through the application of knowledge graphs, the system can generate the best oiling strategy adapted to different production conditions based on comprehensive historical data and real-time data, thereby ensuring the consistency and reliability of oiling.

[0048] 2. It can dynamically adjust the oiling strategy based on real-time data to ensure that the process parameters are always in the best state and avoid oiling problems caused by environmental changes or other factors; through the continuous collection and updating of feedback data, the reinforcement learning model and knowledge graph can continuously optimize the strategy to ensure the continuous improvement and optimization of the oiling process and improve the surface quality and process stability of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0050] Figure 1 This is a schematic diagram of the overall structure of a method for controlling the oiling of a stamped part surface of a fuel tank housing proposed by the present invention;

[0051] Figure 2 A cumulative reward variation diagram in a method for controlling the surface oiling of a fuel tank housing stamping part proposed by the present invention;

[0052] Figure 3 This is a diagram showing the uniformity of the oiling effect in a method for controlling the surface oiling of a fuel tank housing stamping part proposed by the present invention. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0055] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0056] Reference Figure 1 , as an embodiment of the present invention, provides a method for controlling the surface oiling of a fuel tank housing stamping part, the method comprising the following steps:

[0057] Step 1: First, extract the key process parameters in the oiling process from the historical data, and establish the correlation between the parameters to form a knowledge graph;

[0058] Specifically, the method for constructing the knowledge graph is:

[0059] S101: extract key process parameters, including: material type M, oiling time T, stamping pressure P, and oiling amount O;

[0060] S102: Analyze the multivariate relationship in the historical data and establish a knowledge graph for expressing parameter correlation:

[0061] Among them, K(M,T,P,O) represents the correlation between key parameters in the knowledge graph, which serves as the input basis of the reinforcement learning model; f1(M i ,t) represents the influence of material type at different time points; f2(P i ) is the influence of stamping pressure on the oiling effect, which is modeled by the physical properties of stamping; α is the time decay coefficient, ln(1+O i ) represents the logarithmic relationship between the amount of oil applied and its effect.

[0062] The influence of material type at different time points f1(M i ,t) is calculated as:

[0063] f1(M i ,t)=α 0i +α 1i t+α 2i t 2 +…+α ni t n

[0064] Among them, α 0i ,α 1i ,α 2i ,…α ni For material type M i The relevant polynomial coefficients can be obtained by fitting historical data;

[0065] The influence of stamping pressure on the oiling effect f2(P i ) is calculated as:

[0066]

[0067] Among them, β, γ, and δ are physical coefficients related to materials and pressure, which respectively represent the influence of different physical properties.

[0068] The role of the knowledge graph is to record historical experience and provide a comprehensive background database to help the system understand the best oiling solution under different production conditions. These historical data cover a variety of variables and provide data support for subsequent decision-making.

[0069] Step 2: Based on the establishment of the knowledge graph, the reinforcement learning model is initialized based on the associated data in the knowledge graph; the knowledge graph serves as the experience library of the reinforcement learning model, helping the model to make reasonable oiling strategy choices in the initial stage. By inputting the associations in the knowledge graph into the reinforcement learning model, the model can effectively shorten the learning time and generate guiding decisions in the early stages.

[0070] Step 3: The collected real-time data will be input into the knowledge graph and reinforcement learning model. The knowledge graph and reinforcement learning model will work together to analyze and jointly determine the oiling strategy.

[0071] Specifically, the goal of reinforcement learning is to maximize the expected return of the strategy π in the environment and use the associated data extracted from the knowledge graph as the input of the model:

[0072]

[0073] where π * (X) represents the optimal strategy generated by reinforcement learning, X(t) is the feature vector input of real-time process parameters, μ t Represents the discount factor, which describes the discounting effect of future returns.

[0074] Step 4: Establish a feedback update mechanism. After each oiling process, the system will generate feedback data based on the oiling effect. The feedback data will be updated to the knowledge graph to enrich the content of the knowledge graph. As the knowledge graph is continuously updated, the reinforcement learning model will automatically acquire new data for learning to ensure the latest and comprehensiveness of its strategy.

[0075] Specifically, the feedback update mechanism includes:

[0076] S401: Within a certain period of time, collect the following feedback data: Oiling effect O eff :Measure the uniformity, thickness and other physical characteristics of the oiled surface; quality pass rate Q eff: Indicates the qualified proportion of oil-coated products;

[0077] S402: After the feedback data is generated, the knowledge graph will be updated, and its expression is:

[0078]

[0079] Among them, g1(O eff ) is used to measure the contribution of oiling quality to the knowledge graph; g2(Q eff )Quality pass rate Q eff Contribution to knowledge graph updates;

[0080] The g1(O eff ) is calculated as:

[0081] g1(O eff )=θ·log(1+O eff )

[0082] Among them, θ is the adjustment coefficient of the influence of adjusting the oiling effect.

[0083] The g2(Q eff ) is calculated as:

[0084]

[0085] Here, θ2 is an adjustment coefficient for adjusting the influence of the oiling effect.

[0086] S403: After the knowledge graph is updated, the reinforcement learning model will automatically acquire new data for learning to ensure the latest and comprehensiveness of the oiling strategy. The update formula is:

[0087]

[0088] Among them, π new * According to the definition of strategy space, the range of (X) is a multidimensional vector, reflecting the oiling strategy selection at different time steps.

[0089] The reinforcement learning model obtains feedback data through the above formula and continuously updates the oiling strategy. As the knowledge graph gradually becomes richer, the system will have more accurate correlations between process parameters, thereby making more precise strategy adjustments for each oiling process. In the long run, the synergy between the knowledge graph and the reinforcement learning model will improve the system's oiling efficiency and product quality.

[0090] Reference Figure 2-3 , showing the effect of changes before and after the reinforcement learning strategy is updated;

[0091] Cumulative Rewards Changes: Figure 1Shows the cumulative reward of the reinforcement learning policy before and after it is updated. The updated policy improves over the course of training, and the cumulative reward increases, indicating that the model is making better decisions.

[0092] Variation of uniformity of oiling effect: Figure 2 The updated strategy gradually improved the uniformity of the oiling effect, indicating that the updated oiling strategy can more accurately control the oiling quality and achieve better uniformity.

[0093] This embodiment also provides a control system applied to the above-mentioned method for controlling the surface oiling of the fuel tank housing stamping part, characterized in that the system comprises:

[0094] The historical data analysis module is used to collect, organize and analyze the historical data of the oiling process and extract key process parameters. The knowledge graph construction module builds a knowledge graph based on historical data to systematically store and manage different parameters and their relationships.

[0095] Real-time data acquisition module provides the latest process data for real-time oiling decision-making, ensuring that the decision-making process is dynamically adjusted according to the current production environment;

[0096] The reinforcement learning decision module uses the reinforcement learning algorithm to process the data obtained from the real-time data acquisition module and the knowledge graph construction module to generate the best oiling strategy;

[0097] And the feedback data collection and update module, by collecting feedback data, updating the knowledge graph in real time, enriching the database content, and providing it to the reinforcement learning decision module to further optimize the oiling strategy

[0098] This embodiment also provides a computer device, which is suitable for a method for controlling the oiling of the surface of a fuel tank housing stamping part, and includes: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement a method for controlling the oiling of the surface of a fuel tank housing stamping part as proposed in the above embodiment.

[0099] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0100] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, a method for controlling the surface oiling of a fuel tank housing stamping part as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (Static Random Access Memory, referred to as SRAM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), an erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), a programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), a read-only memory (Read-Only Memory, referred to as ROM), a magnetic storage device, a flash memory, a disk or an optical disk.

[0101] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for controlling the surface oiling of a fuel tank housing stamping part, characterized in that: The method comprises the following steps: Step 1: First, extract the key process parameters in the oiling process from the historical data, and establish the correlation between the parameters to form a knowledge graph; Step 2: On the basis of building the knowledge graph, the reinforcement learning model is initialized based on the associated data in the knowledge graph; Step 3: The collected real-time data will be input into the knowledge graph and reinforcement learning model. The knowledge graph and reinforcement learning model will jointly analyze and determine the oiling strategy. Step 4: Establish a feedback update mechanism. After each oiling process, the system will generate feedback data based on the oiling effect. The feedback data will be updated to the knowledge graph to enrich the content of the knowledge graph. As the knowledge graph is continuously updated, the reinforcement learning model will automatically acquire new data for learning to ensure the latest and comprehensiveness of its strategy. The method for constructing the knowledge graph in step 1 is as follows: S101: extract key process parameters, including: material type M, oiling time T, stamping pressure P, and oiling amount O; S102: Analyze the multivariate relationship in the historical data and establish a knowledge graph for expressing parameter correlation: Among them, K(M,T,P,O) represents the correlation between key parameters in the knowledge graph, which serves as the input basis of the reinforcement learning model; f1(M i ,t) represents the influence of material type at different time points; f2(P i ) is the influence of stamping pressure on the oiling effect, which is modeled by the physical properties of stamping; α is the time decay coefficient, ln(1+O i ) represents the logarithmic relationship between the amount of oil applied and its effect; The goal of reinforcement learning is to maximize the expected return of the strategy π in the environment and use the associated data extracted from the knowledge graph as the input of the model: where π * (X) represents the optimal strategy generated by reinforcement learning, X(t) is the feature vector input of real-time process parameters, μ t Represents the discount factor, which describes the discounting effect of future returns.

2. A method for controlling the oiling of a fuel tank housing stamping surface according to claim 1, characterized in that: The influence of material type at different time points f1(M i ,t) is calculated as: f1(M i ,t)=α 0i +α 1i t+α 2i t 2 +…+α ni t n Among them, α 0i ,α 1i ,α 2i ,…α ni For material type M i The relevant polynomial coefficients are obtained by fitting historical data; The influence of stamping pressure on the oiling effect f2(P i ) is calculated as: Among them, β, γ, and δ are physical coefficients related to materials and pressure, which respectively represent the influence of different physical properties.

3. A method for controlling the surface oiling of a fuel tank housing stamping according to claim 2, characterized in that: The feedback update mechanism includes: S401: Within a certain period of time, collect the following feedback data: Oiling effect O eff :Measure the uniformity and thickness of the oiled surface; quality pass rate Q eff : Indicates the qualified proportion of oil-coated products; S402: After the feedback data is generated, the knowledge graph will be updated, and its expression is: Among them, g1(O eff ) is used to measure the contribution of oiling quality to the knowledge graph; g2(Q eff ) is used to measure the quality pass rate Q eff Contribution to knowledge graph updates; S403: After the knowledge graph is updated, the reinforcement learning model will automatically acquire new data for learning to ensure the latest and comprehensiveness of the oiling strategy. The update formula is: Among them, π new * According to the definition of strategy space, the range of (X) is a multidimensional vector, reflecting the oiling strategy selection at different time steps.

4. A method for controlling the oiling of a fuel tank housing stamping surface according to claim 3, characterized in that: The g1(O eff ) is calculated as: in, An adjustment factor for adjusting the influence of the oiling effect.

5. A method for controlling the oiling of a fuel tank housing stamping surface according to claim 4, characterized in that: The g2(Q eff ) is calculated as: in, An adjustment factor for adjusting the influence of the oiling effect.

6. A control system applied to the method for controlling the surface oiling of a fuel tank housing stamping part according to any one of claims 1 to 5, characterized in that: The system includes: The historical data analysis module is used to collect, organize and analyze the historical data of the oiling process and extract key process parameters. The knowledge graph construction module builds a knowledge graph based on historical data to systematically store and manage different parameters and their relationships. Real-time data acquisition module provides the latest process data for real-time oiling decision-making, ensuring that the decision-making process is dynamically adjusted according to the current production environment; The reinforcement learning decision module uses the reinforcement learning algorithm to process the data obtained from the real-time data acquisition module and the knowledge graph construction module to generate the best oiling strategy; As well as the feedback data collection and update module, by collecting feedback data, the knowledge graph is updated in real time, the database content is enriched, and it is provided to the reinforcement learning decision module to further optimize the oiling strategy.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a method for controlling the surface oiling of a fuel tank housing stamping part according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for controlling the surface oiling of a fuel tank housing stamping part according to any one of claims 1 to 5 are implemented.

Images