FAST carbon fiber double-prevention split bulletproof helmet research and development system and method

Abstract

The invention discloses a research and development system and method for a FAST carbon fiber double-prevention split bulletproof helmet, and aims at achieving efficient, high-quality and personalized helmet manufacturing, creating a three-dimensional model of the helmet through the CAD technology, finely designing an inner helmet, an outer helmet and a connecting structure of the inner helmet and the outer helmet, and conducting material pretreatment. An automatic laying machine is adopted to accurately control the fiber direction, and laying data are collected in real time through a sensor. X-ray tomography and machine vision technologies are introduced, nondestructive testing is carried out on internal defects, the defect rate Dr is predicted through machine learning, and the manufacturing process is optimized in combination with real-time feedback so as to reduce interlayer delamination. The protection performance of the helmet is verified through performance testing, the production bottleneck is recognized through generated data, and technological parameters are optimized. Performance indexes are set and adjusted based on user feedback, product quality and user experience are ensured, and intelligent management from design to production is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent manufacturing of helmets, and specifically relates to a FAST carbon fiber double-protection split bulletproof helmet research and development system and method. Background Art

[0002] The special feature of the FAST bulletproof helmet is its double-protection split design, that is, the separation design of the inner helmet and the outer helmet. Although the split structure can improve comfort and protection, its requirements on manufacturing process and structural connection are also more complicated. Therefore, how to effectively control defects in the manufacturing process, such as delamination and porosity between layers, is an issue that must be focused on in the research and development of the FAST helmet. The application of intelligent detection technology can ensure that the quality of each link is controllable. And as users' demand for personalized products increases, traditional mass production methods are difficult to meet the diversified needs of the market. The application of intelligent manufacturing technology can greatly improve production efficiency, reduce waste, and thus reduce costs. The intelligent research and development method and system can effectively solve these background problems and realize efficient, high-quality and low-cost bulletproof helmet manufacturing. Summary of the invention

[0003] A FAST carbon fiber double-defense split bulletproof helmet development method comprises the following steps: S1. Digital design and modeling: Use CAD software for 3D modeling to create a three-dimensional model of the bulletproof helmet. The model includes two parts: the inner helmet and the outer helmet, as well as the design of connection points and interfaces to optimize the helmet structure; S2. Material selection and pretreatment: Select high-strength carbon fiber and aramid fiber hybrid fabric as the matrix material, use filler-modified composite resin glue as the binder, pre-treat the material: clean the fiber fabric and activate the surface to ensure the cleanliness of the fiber fabric and the quality of the prepreg; S3. Automated fiber placement: Use an automated fiber placement machine (AFP) or an automated fiber placement (AFP) system to accurately lay carbon fiber prepreg to ensure the optimal configuration of fiber directions to enhance ballistic performance in specific directions, and collect corresponding data through sensors, wireless communication modules and energy harvesting devices; S4. Intelligent inspection and quality control: Use non-destructive testing technology: X-ray tomography, and apply machine vision for automatic inspection to check the surface quality and internal defects of the helmet body. Record data: raw material characteristics, simulation and emulation, design parameters, interlayer delamination, bubbles, pores and fine defects. Use simulation software to simulate the stress and strain of composite materials under different manufacturing conditions to understand their stress distribution and delamination risks. Use historical data to train machine learning models and integrate them to obtain formulas. ,in represents the defect rate, is the weight vector,b is the bias term, σ(·) is the sigmoid function, ρ Indicates the density of carbon fiber and aramid fiber, H Indicates hardness, U Indicates the impact energy absorption, It is the impact test result, Bp For ballistic performance test data, Based on the fatigue test results, the formula predicts the results, and combined with the real-time feedback of the interlayer delamination defect rate on the production line, the manufacturing process is adjusted in time to avoid and reduce the occurrence of interlayer delamination defects. S5. Testing and verification: Conduct a series of tests, including ballistic performance test, environmental adaptability test, and electronic component function test. Based on the material characteristics and structural design of the helmet, use finite element analysis tools to establish a mechanical model to predict the stress distribution and deformation of the helmet under different impact conditions. Use LS-DYNA software to predict the helmet's response to different types of impacts in actual use: stone and debris impacts, compare the simulation results with the results of the actual impact test, and adjust the model parameters until the two match to ensure the accuracy of the model; S6. Data analysis and continuous optimization: Collect data from the production process: material properties, process parameters and test data, and analyze the generated data through big data Dr The results identified the key steps in production that are prone to interlayer delamination, and adjusted the automated fiber laying parameters, process temperature, pressure and layer angle production process to continuously optimize the manufacturing process and improve efficiency and product quality; S7. Performance indicator setting: Performance indicator system: covers bulletproof performance, wearing comfort, and electronic component reliability. User feedback collection: collect actual feedback from users through questionnaires and on-site tests to understand the performance of the helmet in a real environment. Combine laboratory test data with user feedback to continuously adjust the design and production process to form a PDCA: Plan-Do-Check-Act continuous improvement cycle.

[0004] Furthermore, a FAST carbon fiber double-protection split bulletproof helmet development method is provided. In step S4, non-destructive testing technology is used: X-ray tomography, and machine vision is used for automatic testing to check the surface quality of the helmet body and record the internal defects. The specific steps are as follows: X-ray tomography step: placing the manufactured bulletproof helmet on the working table of the X-ray CT scanner; Parameter setting: Set appropriate scanning parameters according to the material characteristics and size of the bulletproof helmet: X-ray energy, exposure time and scanning angle; Scanning process: Start the X-ray CT scanner to perform a full-scale scan of the bulletproof helmet; Image reconstruction: Use computer software to reconstruct the acquired projection data into cross-sectional images; Data analysis: Analyze images using specialized software such as MATLAB, LabVIEW, and OpenCV to check for pores, cracks, or other internal defects, and record relevant data; Machine vision inspection steps: Install a high-resolution industrial camera and aim it at the bulletproof helmet to be inspected; Image capture: Start the machine vision system to capture high-definition images of the bulletproof helmet; Image processing: Use image processing software to perform image analysis, including edge detection and contrast enhancement; Feature extraction: Extract key features from the processed image: shape, size and location of defects; Classification and recognition: Use a pre-trained convolutional neural network to classify features and determine whether there are surface defects; Result output: Output qualified and unqualified judgments based on the classification results, and record the specific circumstances of the defects; Data collection and organization steps: Record all defect data obtained from X-ray CT scanner and machine vision inspection, including defect type, location, and size; Data cleaning: clean up outliers or erroneous records in the data to ensure data accuracy; Data organization: Organize and classify data according to certain standards to facilitate subsequent analysis and modeling.

[0005] Furthermore, a FAST carbon fiber double-protection split bulletproof helmet development method is provided, in step S4, historical data is used to train a machine learning model, and the specific steps are as follows: S41. Data preparation: Divide the collected data into training set and test set; S42. Model training: Define features based on the test results: type, location and size of defects, use the training set data to train the deep learning model, and integrate the formula ,in represents the defect rate, is the weight vector, b is the bias term, σ(·) is the sigmoid function, ρ represents density, H Indicates hardness, U Indicates the impact energy absorption, It is the impact test result, Bp For ballistic performance test data, is the fatigue test result; S43. Model verification and optimization: Verify the accuracy and generalization ability of the model on the test set, adjust the model parameters according to the verification results, and optimize the model performance; S44. Model deployment: Deploy the optimized model to the production environment to predict in real time whether new products have potential manufacturing defects and failure points; S45. Real-time detection: Continuously use X-ray CT scanners and machine vision to detect newly produced bulletproof helmets, and input the detected new data into the trained machine learning model; S46. Prediction output: The model outputs the probability of whether each sample has a defect and predicts the category. Based on the model prediction results, appropriate measures are taken to correct problems in the production process and avoid similar defects in the future.

[0006] A FAST carbon fiber double-defense split bulletproof helmet research and development system, the FAST carbon fiber double-defense split bulletproof helmet research and development system is used to implement a FAST carbon fiber double-defense split bulletproof helmet research and development method; the FAST carbon fiber double-defense split bulletproof helmet research and development system includes: a digital design and modeling module, an automated fiber laying module, an intelligent detection and quality control module, a personalized customization and assembly module, a testing and verification module, and a data analysis and continuous optimization module; The digital design and modeling module is used to create a three-dimensional digital model of the bulletproof helmet, and the AI-assisted design system helps optimize the structural design and improve the performance of the bulletproof helmet. Automated fiber laying module: used to lay carbon fiber prepreg intelligent component embedding workstation, embedding sensors, wireless communication modules and other intelligent components in the helmet body; Intelligent inspection and quality control module: used to inspect internal structures, use machine vision to inspect surface quality, collect data during the production process, and use trained machine learning models to predict manufacturing defects; Personalized customization and assembly module: obtain the user's head size data and assemble communication equipment and night vision goggles accessories; Testing and verification module: conduct bulletproof performance test, simulate various environmental conditions for testing, electronic component function test bench to ensure the normal operation of electronic components; Data analysis and continuous optimization module: Stores data from the production process, analyzes the data to identify production bottlenecks and improvement opportunities, and develops improvement plans based on the analysis results.

[0007] The beneficial effects of the present invention are as follows: by digital design and modeling, using CAD software to create a three-dimensional model, the design cycle can be accelerated and the product development time can be shortened. The application of automated fiber laying greatly improves the manufacturing speed and accuracy, and reduces the time and errors of manual intervention. By embedding sensors and other intelligent components in the helmet body, such as temperature, humidity, and pressure sensors, as well as wireless communication modules and energy collection devices, the bulletproof helmet has the ability to monitor its own status and external environment. Intelligent functions help to monitor the wearer's condition in real time and can provide emergency support when necessary. Automatic inspection using non-destructive testing technology (such as X-ray tomography) and machine vision technology can ensure the surface quality and internal structural integrity of the product. By using machine learning models to predict manufacturing defects and failure points, measures can be taken in advance to avoid quality problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flow chart of the development method of a FAST carbon fiber double-protection split bulletproof helmet; DETAILED DESCRIPTION

[0009] A FAST carbon fiber double-defense split bulletproof helmet development method comprises the following steps: S1. Digital design and modeling: Use CAD software for 3D modeling to create a three-dimensional model of the bulletproof helmet. The model includes two parts: the inner helmet and the outer helmet, as well as the design of connection points and interfaces. AI-assisted design is used to optimize the helmet structure. S2. Material selection and pretreatment: Select high-strength carbon fiber and aramid fiber hybrid fabric as the matrix material, use filler-modified composite resin glue as the binder, and pretreat the material to ensure the cleanliness of the fiber fabric and the quality of the prepreg; S3. Automated fiber placement: Use an automated fiber placement machine (AFP) or an automated fiber placement (AFP) system to accurately lay carbon fiber prepreg to ensure the optimal configuration of fiber directions to enhance ballistic performance in specific directions, and collect corresponding data through sensors, wireless communication modules and energy harvesting devices; S4. Intelligent inspection and quality control: Use non-destructive testing technology: X-ray tomography, and apply machine vision for automatic inspection to check the surface quality and internal defects of the helmet body. Record data: raw material characteristics, simulation and emulation, design parameters, interlayer delamination, bubbles, pores and fine defects. Use simulation software to simulate the stress and strain of composite materials under different manufacturing conditions to understand their stress distribution and delamination risks. Use historical data to train machine learning models and integrate them to obtain formulas. ,in represents the defect rate, is the weight vector, bis the bias term, σ(·) is the sigmoid function, ρ Indicates the density of carbon fiber and aramid fiber, H Indicates hardness, U Indicates the impact energy absorption, It is the impact test result, Bp For ballistic performance test data, For fatigue test results, the formula is used to predict the results, and combined with the real-time feedback of the interlayer delamination defect rate on the production line, the manufacturing process is adjusted in time to avoid and reduce the occurrence of interlayer delamination defects; S5. Testing and verification: Conduct a series of tests, including ballistic performance test, environmental adaptability test, and electronic component function test. Based on the material characteristics and structural design of the helmet, use finite element analysis tools to establish a mechanical model to predict the stress distribution and deformation of the helmet under different impact conditions. Use LS-DYNA software to predict the helmet's response to different types of impacts in actual use: stone and debris impacts, compare the simulation results with the results of the actual impact test, and adjust the model parameters until the two match to ensure the accuracy of the model; S6. Data analysis and continuous optimization: Collect data from the production process: material properties, process parameters and test data, and analyze the generated data through big data Dr The results identified the key steps in production that are prone to interlayer delamination, and adjusted the automated fiber laying parameters, process temperature, pressure and layer angle production process to continuously optimize the manufacturing process and improve efficiency and product quality; S7. Performance indicator setting: a performance indicator system covering ballistic performance, wearing comfort, and electronic component reliability. User feedback collection: actual feedback from users is collected through questionnaires and on-site tests to understand the performance of the helmet in a real environment. The design and production process are continuously adjusted by combining laboratory test data with user feedback to form a PDCA: Plan-Do-Check-Act continuous improvement cycle.

[0010] Furthermore, a FAST carbon fiber double-protection split bulletproof helmet development method is provided. In step S4, non-destructive testing technology is used: X-ray tomography, and machine vision is used for automatic testing to check the surface quality of the helmet body and record the internal defects. The specific steps are as follows: X-ray tomography step: placing the manufactured bulletproof helmet on the working table of the X-ray CT scanner; Parameter setting: Set appropriate scanning parameters according to the material characteristics and size of the bulletproof helmet: X-ray energy, exposure time and scanning angle; Scanning process: Start the X-ray CT scanner to perform a full-scale scan of the bulletproof helmet; Image reconstruction: Use computer software to reconstruct the acquired projection data into cross-sectional images; Data analysis: Analyze images using specialized software such as MATLAB, LabVIEW, and OpenCV to check for pores, cracks, or other internal defects, and record relevant data; Machine vision inspection steps: Install a high-resolution industrial camera and aim it at the bulletproof helmet to be inspected; Image capture: Start the machine vision system to capture high-definition images of the bulletproof helmet; Image processing: Use image processing software to perform image analysis, including edge detection and contrast enhancement; Feature extraction: Extract key features from the processed image: shape, size and location of defects; Classification and recognition: Use a pre-trained convolutional neural network to classify features and determine whether there are surface defects; Result output: Output qualified and unqualified judgments based on the classification results, and record the specific circumstances of the defects; Data collection and organization steps: Record all defect data obtained from X-ray CT scanner and machine vision inspection, including defect type, location, and size; Data cleaning: clean up outliers or erroneous records in the data to ensure data accuracy; Data organization: Organize and classify data according to certain standards to facilitate subsequent analysis and modeling.

[0011] Furthermore, a FAST carbon fiber double-protection split bulletproof helmet development method is provided, in step S4, historical data is used to train a machine learning model, and the specific steps are as follows: S41. Data preparation: Divide the collected data into training set and test set; S42. Model training: Define features based on the test results: type, location and size of defects, use the training set data to train the deep learning model, and integrate the formula ,in represents the defect rate, is the weight vector, b is the bias term, σ(·) is the sigmoid function, ρ represents density, H Indicates hardness, U Indicates the impact energy absorption, It is the impact test result, Bp For ballistic performance test data, is the fatigue test result; S43. Model verification and optimization: Verify the accuracy and generalization ability of the model on the test set, adjust the model parameters according to the verification results, and optimize the model performance; S44. Model deployment: Deploy the optimized model to the production environment to predict in real time whether new products have potential manufacturing defects and failure points; S45. Real-time detection: Continuously use X-ray CT scanners and machine vision to detect newly produced bulletproof helmets, and input the detected new data into the trained machine learning model; S46. Prediction output: The model outputs the probability of whether each sample has a defect and predicts the category. Based on the model prediction results, appropriate measures are taken to correct problems in the production process and avoid similar defects in the future.

[0012] A FAST carbon fiber double-defense split bulletproof helmet research and development system, the FAST carbon fiber double-defense split bulletproof helmet research and development system is used to implement a FAST carbon fiber double-defense split bulletproof helmet research and development method; the FAST carbon fiber double-defense split bulletproof helmet research and development system includes: a digital design and modeling module, an automated fiber laying module, an intelligent detection and quality control module, a personalized customization and assembly module, a testing and verification module, and a data analysis and continuous optimization module; The digital design and modeling module is used to create a three-dimensional digital model of the bulletproof helmet, helping to optimize the structural design and improve the performance of the bulletproof helmet; Automated fiber laying module: used to lay carbon fiber prepreg intelligent component embedding workstation, embedding sensors, wireless communication modules and other intelligent components in the helmet body; Intelligent inspection and quality control module: used to inspect internal structures, use machine vision to inspect surface quality, collect data during the production process, and use trained machine learning models to predict manufacturing defects; Personalized customization and assembly module: obtain the user's head size data and assemble communication equipment and night vision goggles accessories; Testing and verification module: conduct bulletproof performance test, simulate various environmental conditions for testing, electronic component function test bench to ensure the normal operation of electronic components; Data analysis and continuous optimization module: Stores data from the production process, analyzes the data to identify production bottlenecks and improvement opportunities, and develops improvement plans based on the analysis results.

Claims

1. A method for developing a FAST carbon fiber double-defense split bulletproof helmet, characterized in that: The following steps are included: S1. Digital design and modeling: Use CAD software for 3D modeling to create a three-dimensional model of the bulletproof helmet. The model includes two parts: the inner helmet and the outer helmet, as well as the design of connection points and interfaces, and optimize the helmet structure; S2. Material selection and pretreatment: Select carbon fiber and aramid fiber hybrid fabric as the matrix material, use filler-modified resin glue as the binder, pre-treat the material: clean the fiber fabric, activate the surface, and evaluate the cleanliness of the fiber fabric and the quality of the prepreg; S3. Automated fiber placement: The automated fiber placement machine AFP is used to accurately lay carbon fiber prepreg, optimize the configuration of fiber direction, and collect corresponding data through sensors, wireless communication modules and energy collection devices; S4. Intelligent inspection and quality control: Use non-destructive testing technology: X-ray tomography, and apply machine vision for automatic inspection to check the surface quality and internal defects of the helmet body. Record data: raw material characteristics, simulation and emulation, design parameters, interlayer delamination, bubbles, pores and fine defects. Use simulation software to simulate the stress and strain of composite materials under different manufacturing conditions to understand their stress distribution and delamination risks. Use historical data to train machine learning models and integrate them to obtain formulas. ,in represents the defect rate, is the weight vector, b is the bias term, σ(·) is the sigmoid function, ρ Indicates the density of carbon fiber and aramid fiber, H Indicates hardness, U Indicates the impact energy absorption, It is the impact test result, B For ballistic performance test data, For fatigue test results, the formula is used to predict the results, and combined with the real-time feedback of the interlayer delamination defect rate on the production line, the manufacturing process is adjusted in time to avoid and reduce the occurrence of interlayer delamination defects; S5. Testing and verification: Conduct a series of tests, including ballistic performance test, environmental adaptability test, and electronic component function test. Based on the material characteristics and structural design of the helmet, use finite element analysis tools to establish a mechanical model to predict the stress distribution and deformation of the helmet under different impact conditions. Use LS-DYNA software to predict the helmet's response to different types of impacts in actual use: stone and debris impacts, compare the simulation results with the results of the actual impact test, and adjust the model parameters until the two match to ensure the accuracy of the model; S6. Data analysis and continuous optimization: Collect data from the production process: material properties, process parameters and test data, and analyze the generated data through big data Dr The results identified the key steps in production that are prone to interlayer delamination, and adjusted the automated fiber laying parameters, process temperature, pressure and layer angle production process to continuously optimize the manufacturing process and improve efficiency and product quality; S7. Performance indicator setting: Performance indicator system: covers bulletproof performance and wearing comfort. Through questionnaires and on-site tests, actual feedback from users is collected to understand the performance of the helmet in a real environment. By combining laboratory test data with user feedback, the design and production process are continuously adjusted to form a PDCA: Plan-Do-Check-Act continuous improvement cycle.

2. A FAST carbon fiber double-defense split bulletproof helmet development method as claimed in claim 1, characterized in that: In step S4, non-destructive testing technology is used: X-ray tomography, and machine vision is used for automatic testing to check the surface quality of the helmet body and record the internal defects. The specific steps are as follows: X-ray tomography step: placing the manufactured bulletproof helmet on the working table of the X-ray CT scanner; Parameter setting: Set appropriate scanning parameters according to the material characteristics and size of the bulletproof helmet: X-ray energy, exposure time and scanning angle; Scanning process: Start the X-ray CT scanner to perform a full-scale scan of the bulletproof helmet; Image reconstruction: Use computer software to reconstruct the acquired projection data into cross-sectional images; Data analysis: Analyze the images through dedicated software to check for pores, cracks, and other internal defects, and record relevant data; Machine vision inspection steps: Install a high-resolution industrial camera and aim it at the bulletproof helmet to be inspected; Image capture: Start the machine vision system to capture high-definition images of the bulletproof helmet; Image processing: Use image processing software to perform image analysis, including edge detection and contrast enhancement; Feature extraction: Extract key features from the processed image: shape, size and location of defects; Classification and recognition: Use a pre-trained convolutional neural network to classify features and determine whether there are surface defects; Result output: Output qualified and unqualified judgments based on the classification results, and record the specific circumstances of the defects; Data collection and organization steps: Record defect data obtained from X-ray CT scanner and machine vision inspection, including defect type, location, and size; Data cleaning: clean up outliers and erroneous records in the data; Data collation: Organize and classify data according to certain standards.

3. A FAST carbon fiber double-defense split bulletproof helmet development method as claimed in claim 1, characterized in that: In step S4, historical data is used to train the machine learning model. The specific steps are as follows: S41.Data preparation: Divide the collected data into training set and test set; S42. Model training: Define features based on the test results: type, location and size of defects, use the training set data to train the deep learning model, and integrate the formula ,in represents the defect rate, is the weight vector, b is the bias term, σ(·) is the sigmoid function, ρ represents density, H Indicates hardness, U Indicates the impact energy absorption, It is the impact test result, B For ballistic performance test data, is the fatigue test result; S43. Model verification and optimization: Verify the accuracy and generalization ability of the model on the test set, adjust the model parameters according to the verification results, and optimize the model performance; S44. Model deployment: Deploy the optimized model to the production environment to predict in real time whether new products have potential manufacturing defects and failure points; S45. Real-time detection: Continuously use X-ray CT scanners and machine vision to detect newly produced bulletproof helmets, and input the detected new data into the trained machine learning model; S46. Prediction output: The model outputs the probability of whether each sample has a defect and predicts the category. Based on the model prediction results, appropriate measures are taken to correct problems in the production process and avoid similar defects in the future.

4. A FAST carbon fiber double-defense split bulletproof helmet development system, characterized in that: The FAST carbon fiber double-defense split bulletproof helmet development system is used to implement the FAST carbon fiber double-defense split bulletproof helmet development method as described in any one of claims 1 to 3; The FAST carbon fiber double-protection split bulletproof helmet research and development system includes: a digital design and modeling module, an automated fiber laying module, an intelligent detection and quality control module, a personalized customization and assembly module, a testing and verification module, and a data analysis and continuous optimization module; The digital design and modeling module is used to create a three-dimensional digital model of the bulletproof helmet, helping to optimize the structural design and improve the performance of the bulletproof helmet; Automated fiber laying module: used to lay carbon fiber prepreg intelligent component embedding workstation, embed sensors and wireless communication module intelligent components in the helmet body; Intelligent inspection and quality control module: used to inspect internal structures, use machine vision to inspect surface quality, collect data during the production process, and use trained machine learning models to predict manufacturing defects; Personalized customization and assembly module: obtain the user's head size data and assemble communication equipment and night vision goggles accessories; Testing and verification module: conduct bulletproof performance test, simulate various environmental conditions for testing, electronic component function test bench to ensure the normal operation of electronic components; Data analysis and continuous optimization module: Stores data from the production process, analyzes the data to identify production bottlenecks and improvement opportunities, and develops improvement plans based on the analysis results.

Images