A method and system for manufacturing silicone balls for rehabilitation training

Abstract

This invention relates to a method and system for manufacturing silicone balls for rehabilitation training in the field of silicone product manufacturing technology. The method includes: simulating the internal stress distribution and material flow characteristics of the silicone ball using finite element analysis based on the hardness gradient curve; obtaining the mold zoning scheme and vulcanization temperature field distribution through iterative calculations; acquiring the geometric data of the completed mold from the mold processing instructions; detecting the mold surface error and zoning accuracy through laser scanning to determine whether the mold meets the hardness gradient control requirements; adjusting the surface molecular structure of the silicone ball semi-finished product using plasma treatment technology; obtaining optimized surface friction parameters by controlling the treatment time and gas ratio; obtaining silicone ball performance data from the surface friction parameters; measuring the hardness and gripping force distribution using mechanical testing equipment to determine whether the silicone ball meets the needs of rehabilitation training.

Description

Technical Field

[0001] This invention relates to the field of silicone product manufacturing technology, and in particular to a method and system for manufacturing silicone balls for rehabilitation training. Background Technology

[0002] Silicone balls are crucial in sports rehabilitation and fitness, as their performance directly impacts training effectiveness and user experience. Solid silicone balls, with their controllable elasticity and durability, are ideal tools for stimulating muscle groups and improving coordination during rehabilitation training. However, silicone balls with a single hardness cannot meet the diverse needs of different rehabilitation stages or muscle groups.

[0003] While existing multi-layer injection molding processes can achieve hardness stratification, their complex production processes and cumbersome mold designs lead to high costs, hindering their widespread application in the medical and consumer markets. Furthermore, current methods suffer from deficiencies in the continuity of hardness transitions and the uniformity of material distribution, impacting user comfort and functionality. The core challenge lies in achieving precise gradient control of silicone ball hardness while reducing manufacturing costs. Single vulcanization processes struggle to create continuous hardness variations, and while multi-layer injection molding can achieve stratification, uneven interlayer bonding can lead to density deviations and performance instability.

[0004] Meanwhile, optimizing the grip strength of the ball's surface also needs to be considered to ensure safety during rehabilitation training. The unresolved technical factors limit the application of silicone balls in personalized rehabilitation needs, making it difficult to achieve differentiated stimulation of different muscle groups.

[0005] Therefore, the key issue of this study is how to improve the mold design and vulcanization process to achieve a continuous gradient change in hardness from the inside to the outside of the sphere while ensuring uniform material distribution and optimizing surface friction. Summary of the Invention

[0006] This invention provides a method for manufacturing silicone balls for rehabilitation training, comprising the following steps:

[0007] The hardness gradient requirements and surface friction standards are obtained from the silicone ball design parameter database. The sensitivity range of different muscle groups to hardness changes during rehabilitation training is analyzed by algorithm, and the hardness gradient curve and surface texture parameters are determined.

[0008] For the hardness gradient curve, finite element analysis was used to simulate the internal stress distribution and material flow characteristics of the silicone ball, and the mold partitioning scheme and vulcanization temperature field distribution were obtained through iterative calculation.

[0009] If the finite element analysis results show that the interlayer stress difference is greater than the preset threshold, the mold partition boundary and silicone formula ratio are adjusted, and the interlayer bonding uniformity is verified through secondary simulation to obtain the optimized mold design parameters.

[0010] Based on the optimized mold design parameters, a three-dimensional model of a multi-cavity mold is generated. A CNC machining path planning algorithm is used to calculate the mold machining trajectory and obtain high-precision mold machining instructions.

[0011] The geometric data of the finished mold is obtained from the mold processing instructions. The surface error and partitioning accuracy of the mold are detected by laser scanning to determine whether the mold meets the hardness gradient control requirements.

[0012] If the mold error is less than the preset threshold, the silicone ball molding process is controlled by a segmented vulcanization process. By monitoring the vulcanization temperature and pressure distribution in real time, a silicone ball semi-finished product with continuous hardness variation is obtained.

[0013] For semi-finished silicone balls, plasma treatment technology is used to adjust the surface molecular structure. By controlling the treatment time and gas ratio, optimized surface friction parameters are obtained.

[0014] The performance data of the silicone ball is obtained from the surface friction parameters. The hardness and grip force distribution are measured by mechanical testing equipment to determine whether the silicone ball meets the rehabilitation training needs.

[0015] If the test data deviation is less than the preset threshold, an automated sorting system is used to classify the silicone balls by quality, and surface defects are identified by machine vision to obtain finished silicone balls that meet the standards.

[0016] This invention provides a silicone ball manufacturing system for rehabilitation training, mainly comprising:

[0017] The hardness gradient requirement acquisition module is used to obtain hardness gradient requirements and surface friction standards from the silicone ball design parameter database.

[0018] The surface friction standard acquisition module is used to analyze the sensitivity range of different muscle groups to changes in hardness during rehabilitation training through algorithms, and to determine the hardness gradient curve and surface texture parameters.

[0019] The algorithm analysis module is used to simulate the internal stress distribution and material flow characteristics of silicone balls using finite element analysis based on the hardness gradient curve, and obtains the mold partitioning scheme and vulcanization temperature field distribution through iterative calculation.

[0020] The finite element analysis module is used to adjust the mold partition boundary and silicone formula ratio if the finite element analysis results show that the interlayer stress difference is greater than the preset threshold. It then verifies the interlayer bonding uniformity through secondary simulation to obtain the optimized mold design parameters.

[0021] The mold partitioning adjustment module is used to generate a three-dimensional model of a multi-cavity mold based on the optimized mold design parameters, and to calculate the mold machining trajectory using a CNC machining path planning algorithm to obtain high-precision mold machining instructions.

[0022] The mold processing instruction generation module is used to obtain the geometric data of the processed mold from the mold processing instructions, and to detect the surface error and partitioning accuracy of the mold by laser scanning to determine whether the mold meets the hardness gradient control requirements.

[0023] The mold detection module is used to control the silicone ball molding process by using a segmented vulcanization process if the mold error is less than a preset threshold. By monitoring the vulcanization temperature and pressure distribution in real time, a silicone ball semi-finished product with continuous hardness variation is obtained.

[0024] The vulcanization process control module is used to adjust the surface molecular structure of silicone ball semi-finished products using plasma treatment technology. By controlling the treatment time and gas ratio, optimized surface friction parameters are obtained.

[0025] The surface treatment module is used to obtain performance data of the silicone ball from surface friction parameters, and to measure hardness and grip force distribution through mechanical testing equipment to determine whether the silicone ball meets the needs of rehabilitation training.

[0026] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0027] This invention discloses a method for manufacturing silicone balls for rehabilitation training. By analyzing the sensitivity of muscle groups to changes in hardness, the hardness gradient curve and surface texture parameters are determined. Finite element analysis is used to optimize the mold design, enabling the molding of silicone balls with continuously varying hardness. Plasma treatment technology is used to adjust surface friction, and mechanical testing verifies the performance of the silicone balls. This invention solves the problems of traditional silicone balls having a single hardness and uncontrollable surface friction, achieving the manufacturing of silicone balls with adjustable hardness gradients and controllable surface friction, thus improving the targeting and effectiveness of rehabilitation training. Automated sorting and machine vision inspection ensure consistent product quality, providing a more precise and personalized auxiliary tool for rehabilitation training. Attached Figure Description

[0028] Figure 1 This is a flowchart of a method for manufacturing a silicone ball for rehabilitation training according to the present invention.

[0029] Figure 2 This is a schematic diagram of a method and system for manufacturing silicone balls for rehabilitation training according to the present invention.

[0030] Figure 3 This is another schematic diagram of a method and system for manufacturing silicone balls for rehabilitation training according to the present invention.

[0031] Figure 4 This is a schematic diagram of the framework of a silicone ball manufacturing system for rehabilitation training according to the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0033] like Figure 1-4 This embodiment of a method for manufacturing a silicone ball for rehabilitation training may specifically include:

[0034] S101. Obtain the hardness gradient requirements and surface friction standards from the silicone ball design parameter database. Analyze the sensitivity range of different muscle groups to hardness changes during rehabilitation training using algorithms to determine the hardness gradient curve and surface texture parameters.

[0035] Hardness gradient data and surface friction standard data are obtained from the silicone ball design parameter database to generate a structured dataset. A linear regression algorithm is used to analyze the correlation between hardness gradient and surface friction in the structured dataset, obtaining the mapping relationship between hardness gradient and friction. Based on this mapping relationship, training data for different muscle groups in rehabilitation training is obtained, and the impact of hardness gradient changes on the stimulation intensity of muscle groups is analyzed to determine the hardness change sensitivity range for each muscle group. If the sensitivity range covers a preset number of muscle groups, a clustering algorithm is used to group the sensitivity range, obtaining hardness gradient curves for different muscle groups. A texture generation algorithm is used to optimize the texture parameters corresponding to surface friction based on the hardness gradient curves, generating a texture parameter set. Through matching analysis between the texture parameter set and the hardness gradient curves, it is determined whether the parameter combination meets the friction standard for rehabilitation training, determining the final hardness gradient and surface texture parameters. If the final hardness gradient and surface texture parameters meet the standard, they are stored in the database, generating a silicone ball design parameter configuration file.

[0036] Specifically, hardness gradient requirements are extracted from the silicone ball design parameter database, such as a hardness range of 20 to 80 Shore A and a surface friction standard of 0.3 to 0.8. Machine learning algorithms, such as Support Vector Machine (SVM), are then used to analyze the sensitivity ranges of different muscle groups to changes in hardness during rehabilitation training.

[0037] For example, the sensitivity range of upper limb muscles to changes in hardness is 30 to 50 Shore A, while that of lower limb muscles is 40 to 60 Shore A. Based on these data, a multinomial regression algorithm is used to determine the hardness gradient curve; for example, the hardness gradient curve for upper limb muscles is y = 0.02x. 2 +0.5x+20, lower limb muscle strength is y=0.03x 2+0.4x+30. Simultaneously, surface texture parameters are determined using surface texture analysis algorithms, such as Fourier transform. For example, the surface texture parameters for upper limb muscles are 0.5 to 0.7, and for lower limb muscles, they are 0.6 to 0.8. These parameters and curves will be used to optimize the design of the silicone balls to meet the rehabilitation training needs of different muscle groups.

[0038] S102. For the hardness gradient curve, finite element analysis is used to simulate the internal stress distribution and material flow characteristics of the silicone ball. The mold partitioning scheme and vulcanization temperature field distribution are obtained through iterative calculation.

[0039] Acquire 3D model data and hardness gradient curve data of silicone spheres to generate an initial silicone sphere model. Use finite element analysis (FEM) to mesh the initial silicone sphere model and calculate the internal stress distribution to obtain stress distribution data. Analyze the material flow characteristics based on the stress distribution data and extract flow characteristic parameters. If the flow characteristic parameters exceed a preset threshold, adjust the hardness gradient curve, perform finite element analysis again, and iteratively optimize to obtain optimized stress distribution data and flow characteristic parameters. Determine a mold zoning scheme based on the optimized stress distribution data. Simulate the vulcanization process for the mold zoning scheme, calculate the temperature field distribution, and obtain the vulcanization temperature field distribution. Use a genetic algorithm to optimize the vulcanization temperature field distribution, adjust the mold zoning scheme, and determine the final mold zoning scheme and vulcanization temperature field distribution.

[0040] Specifically, in the finite element analysis, a three-dimensional model of the silicone sphere was first established using tetrahedral meshing with a mesh size of 0.5 mm to ensure computational accuracy. The material properties were defined by inputting the elastic modulus of silicone (2.5 MPa) and Poisson's ratio (0.49). In the boundary condition settings, the outer surface of the silicone sphere was fixed to simulate constraints in actual use. A nonlinear statics analysis module was used, with 100 iterations and a convergence accuracy of 0.001, to calculate stress distribution. The analysis results showed that the maximum stress inside the silicone sphere occurred in the central region, reaching 1.8 MPa, while the stress in the edge region was lower, approximately 0.5 MPa. Based on the stress distribution results, a K-means clustering algorithm was used to partition the mold, with 3 clusters and 50 iterations to obtain the partitioning scheme. In the vulcanization temperature field simulation, a heat conduction analysis module was used, with an initial temperature of 25 degrees Celsius, a vulcanization temperature of 150 degrees Celsius, a heating rate of 5 degrees Celsius per minute, a calculation time step of 1 minute, and a total calculation time of 30 minutes. Through iterative calculations, the temperature distribution in each region of the mold was obtained. The central region had the highest temperature, reaching 145 degrees Celsius, while the edge regions had a lower temperature, approximately 130 degrees Celsius. Based on the temperature field distribution results, the heating power was adjusted, with 500 watts set for the central region and 300 watts for the edge regions to achieve uniform vulcanization. Finally, through finite element analysis and iterative calculations, the mold zoning scheme and vulcanization temperature field distribution were optimized, improving the quality and performance of the silicone balls.

[0041] S103. If the finite element analysis results show that the interlayer stress difference is greater than the preset threshold, adjust the mold partition boundary and silicone formula ratio, and verify the interlayer bonding uniformity through secondary simulation to obtain the optimized mold design parameters.

[0042] Interlayer stress difference data is obtained through finite element analysis. It is then determined whether the interlayer stress difference exceeds a preset threshold, resulting in stress distribution analysis. If the interlayer stress difference exceeds the preset threshold, an optimization algorithm is used to adjust the mold partition boundaries, determining the mold partition boundary parameters. Based on the mold partition boundary parameters, an initial value for the silicone formula ratio is obtained, and an optimized formula ratio is obtained through matching with a material property database. A secondary finite element simulation is performed using the mold partition boundary parameters and the optimized formula ratio to obtain interlayer bonding uniformity data. By comparing the interlayer bonding uniformity data with a preset uniformity standard, it is determined whether the requirements are met, obtaining verification results. If the verification results meet the preset uniformity standard, optimized mold design parameters are output, determining the final design parameters. Based on the final design parameters, a digital model of the mold partitions and silicone formula is generated, yielding executable production parameters.

[0043] Specifically, in the finite element analysis, if the interlayer stress difference exceeds a preset threshold (e.g., 50 MPa), the system will automatically adjust the mold partition boundaries, increasing the number of partitions from 5 to 8, and recalculate the stress distribution of each partition. Simultaneously, the silicone formula ratio will be optimized based on the stress distribution, adjusting the ratio of silicone A to silicone B from 7:3 to 6:4 to reduce the interlayer stress difference. After adjustment, the system will perform a secondary simulation using the nonlinear contact algorithm in ANSYS software to calculate the interlayer bonding uniformity. The simulation results show that the interlayer stress difference is reduced to 30 MPa, below the preset threshold, and the interlayer bonding uniformity is improved by 15%. Finally, the system outputs optimized mold design parameters, including partition boundary coordinates, silicone formula ratio, and interlayer bonding uniformity index, providing precise guidance for subsequent mold manufacturing.

[0044] S104. Based on the optimized mold design parameters, generate a three-dimensional model of the multi-cavity mold, and use a CNC machining path planning algorithm to calculate the mold machining trajectory to obtain high-precision mold machining instructions.

[0045] A set of mold design parameters is obtained, including the geometric and material parameters of a multi-cavity mold. The set of mold design parameters is analyzed to determine the structural layout of the multi-cavity mold, obtaining structural description data. A geometric modeling algorithm is used to construct a three-dimensional model of the multi-cavity mold based on the structural description data, obtaining three-dimensional model data. It is determined whether the three-dimensional model data meets preset geometric constraints. If it does, a path planning algorithm is used to generate a machining trajectory, obtaining trajectory data. If it does not meet the constraints, the structural description data is adjusted, and the three-dimensional model is reconstructed. Based on the trajectory data and precision control rules, the machining trajectory is optimized to obtain high-precision trajectory data. Machining features are extracted from the high-precision trajectory data to generate machining instructions, obtaining an instruction sequence. The consistency between the machining instructions and the three-dimensional model data is verified using the instruction sequence to obtain the final machining instructions. It is determined whether the final machining instructions meet a preset precision threshold. If they do, a high-precision mold machining instruction is output; if they do not, the optimization parameters are adjusted, and the machining trajectory is regenerated.

[0046] Specifically, in the optimized mold design parameters, the stress distribution of the mold was first determined through finite element analysis. The mold was simulated using ANSYS software, with the material set as 45# steel, an elastic modulus of 210 GPa, and a Poisson's ratio of 0.3. The analysis results showed that the maximum stress occurred at the corner of the mold, reaching 350 MPa. Based on this, the mold wall thickness was adjusted to 15 mm, and a 5 mm radius fillet was added at the corner to reduce stress concentration. Next, a 3D model of the multi-cavity mold was generated using UGNX software. The mold dimensions were set to 200 mm × 150 mm × 50 mm, with 4 cavities, each measuring 50 mm × 50 mm × 30 mm. A 10 mm spacing between cavities was ensured to avoid interference during machining. In CNC machining path planning, a path optimization method based on the A* algorithm is adopted. The machining accuracy is set to 0.01 mm, the tool diameter to 10 mm, the cutting speed to 200 mm / min, and the feed rate to 0.1 mm / rev. The algorithm calculates the optimal machining trajectory, ensuring the shortest toolpath and no collisions. Finally, the generated machining trajectory is converted into G-code, and the spindle speed is set to 3000 rpm and the coolant flow rate to 5 L / min. High-precision mold machining instructions are generated to ensure that the surface roughness of the mold reaches Ra 0.8 μm, meeting the high-precision requirements.

[0047] S105. Obtain the geometric data of the completed mold from the mold processing instructions, and use laser scanning to detect the surface error and partitioning accuracy of the mold to determine whether the mold meets the hardness gradient control requirements.

[0048] The three-dimensional geometric data of the completed mold is extracted from the mold processing instructions. A preset analytical algorithm is used to decompose the instruction content to obtain the three-dimensional geometric data of the mold. Surface point cloud data of the mold's three-dimensional geometric data is obtained through laser scanning. A point cloud registration algorithm is used to generate mold surface error data and partition accuracy data. The mold surface error data is compared with a preset error threshold. If the surface error data exceeds the preset error threshold, the mold surface is determined to be unqualified, resulting in a surface error analysis result. The partition accuracy data is compared with a preset accuracy partition requirement. If the partition accuracy data meets the preset accuracy partition requirement, the partition accuracy is determined to be qualified, resulting in a partition accuracy analysis result. The mold surface is scanned using laser hardness testing technology to obtain hardness gradient distribution data. The hardness gradient distribution data is compared with a preset hardness gradient threshold. If the hardness gradient distribution data meets the preset hardness gradient threshold, the mold hardness control is determined to be qualified, resulting in a hardness control analysis result. Based on the surface error analysis result, the partition accuracy analysis result, and the hardness control analysis result, a logical AND operation is performed. If all three results are qualified, the mold is determined to meet the overall requirements, resulting in a final judgment result.

[0049] Specifically, the geometric data of the completed mold is extracted from the mold processing instructions. First, the machining path and cutting parameters of the mold are obtained through the G-code parsing module of the CNC machine tool, such as a machining depth of 2.5 mm and a cutting speed of 120 m / min. Next, the processed mold is geometrically reconstructed using 3D modeling software to generate an STL format model file, with the model accuracy controlled within 0.01 mm. Subsequently, a laser scanner is used to perform high-precision scanning of the mold surface, with a point cloud density of 1000 points per square millimeter. The scanned data is compared with the design model using a point cloud processing algorithm (such as the ICP algorithm) to calculate the surface error distribution, with an error threshold set at ±0.05 mm. For zonal accuracy detection, a region segmentation algorithm (such as K-means clustering) is used to divide the mold surface into several regions, each with a size of 10 mm × 10 mm. The average error value of each region is calculated, requiring that the error of each region does not exceed ±0.03 mm. Finally, the hardness distribution in different areas of the mold is analyzed using a hardness gradient control algorithm. For example, the surface hardness requirement is HRC58-62, and the internal hardness is HRC45-50. The heat treatment process is simulated using finite element analysis software to ensure that the hardness gradient meets the design requirements and the gradient change rate is controlled within the range of 1-2 HRC per millimeter.

[0050] S106. If the mold error is less than the preset threshold, the silicone ball molding process is controlled by a segmented vulcanization process. By monitoring the vulcanization temperature and pressure distribution in real time, a silicone ball semi-finished product with continuous hardness variation is obtained.

[0051] Mold error data is collected by sensors to determine if the mold error is less than a preset threshold. If the mold error is less than the preset threshold, the error distribution characteristics of the mold error data are determined. Based on the error distribution characteristics, a segmented vulcanization process is adopted to generate initial control parameters for vulcanization temperature and pressure distribution. Temperature and pressure data are acquired from the vulcanization equipment through a real-time monitoring system to determine if the temperature and pressure data deviate from the initial control parameters. If the temperature or pressure data deviates from the initial control parameters, the vulcanization process parameters are adjusted using a support vector machine algorithm to obtain optimized control parameters. Based on the optimized control parameters, a closed-loop control system is used to adjust the vulcanization equipment to generate silicone ball semi-finished products with continuous hardness variation. Hardness distribution data is acquired from the silicone ball semi-finished products using a hardness detection device to determine if the hardness distribution data meets the continuity requirement. If the hardness distribution data meets the continuity requirement, the optimized control parameters and the hardness distribution data are recorded through a data storage system to generate a process control log for the silicone ball semi-finished products.

[0052] Specifically, when the mold error is less than a preset threshold of 0.05 mm, the system automatically initiates a segmented vulcanization process to control the molding of silicone balls. First, high-precision sensors collect vulcanization temperature data in real time, controlling the temperature between 150℃ and 180℃. A PID algorithm dynamically adjusts the heating power based on the temperature change rate, ensuring temperature fluctuations do not exceed ±2℃. Simultaneously, pressure sensors monitor the vulcanization pressure, set within a range of 10MPa to 15MPa. A fuzzy control algorithm adjusts the pressure valve in real time to ensure uniform pressure distribution, with deviations controlled within ±0.5MPa. During vulcanization, the system collects temperature and pressure data every 5 seconds and uses a multivariate regression analysis model to predict the hardness trend of the silicone balls. When the hardness value reaches the preset range of 70±5 Shore A, the system automatically switches to the next vulcanization stage until the entire vulcanization process is completed. Finally, a continuous hardness tester analyzes the hardness distribution of the semi-finished silicone balls to ensure the hardness gradient meets design requirements and the hardness change rate does not exceed 2 Shore A / mm. All data is uploaded to a cloud database in real time for subsequent quality analysis and process optimization.

[0053] S107. For silicone ball semi-finished products, plasma treatment technology is used to adjust the surface molecular structure. By controlling the treatment time and gas ratio, optimized surface friction parameters are obtained.

[0054] Obtain the initial surface molecular structure data and baseline friction force parameters of the silicone ball semi-finished product. The initial surface molecular structure data is obtained through a plasma processing device, and the baseline friction force parameters are obtained through a friction force testing device. Use a machine learning model to analyze the relationship between the initial surface molecular structure data and the baseline friction force parameters to obtain a preliminary optimization range of the processing time and gas ratio. According to the preliminary optimization range, adjust the processing time and gas ratio of the plasma processing device to generate the first set of processed surface molecular structure data. Obtain the friction force parameters of the first set of processed surface molecular structure data through a friction force testing device, and determine whether the friction force parameters reach a preset threshold. If the friction force parameters do not reach the preset threshold, adjust the processing time and gas ratio through an iterative optimization algorithm to generate the second set of processed surface molecular structure data. Obtain the friction force parameters of the second set of processed surface molecular structure data through a friction force testing device, and determine whether the optimization target is met. According to the judgment result, use data fusion technology to integrate the first set of processed surface molecular structure data, the second set of processed surface molecular structure data, and the corresponding friction force parameters to determine the final processing time and gas ratio configuration. Apply the final processing time and gas ratio configuration through a plasma processing device to process the silicone ball semi-finished product to obtain optimized surface friction force parameters.

[0055] Specifically, during the surface treatment process of the silicone ball semi-finished product, first set the basic parameters through a plasma device. For example, use a radio frequency power of 300 watts and a chamber pressure of 50 pascals as the initial conditions. Then introduce a mixed gas for surface activation, where the volume ratio of argon to oxygen is controlled at 4:1. The gas ratio fluctuation is monitored in real time through a mass spectrometer and fed back to the control system to ensure that the ratio deviation does not exceed plus or minus two percent. The processing time is dynamically adjusted according to the surface energy target value. For example, when the contact angle needs to be reduced from 110 degrees to 70 degrees, a gradient time algorithm is used. In the first 5 minutes, it is processed with a linear model of 2 degrees per minute, and in the next 3 minutes, it is switched to an exponential decay model. Finally, the surface roughness is detected by an atomic force microscope. If the measured Ra value is in the range of 0.15 to 0.25 microns, it is judged to be qualified. For friction optimization, the Hertz contact theory is used to calculate the contact area, combined with the Coulomb friction model. When the normal load is 1 newton, the processing parameters are adjusted through an iterative algorithm until the measured friction coefficient is stable in the range of 0.35 plus or minus 0.02. In the data processing stage, the cubic spline interpolation method is used to fit the non-linear relationship between the process parameters and the friction coefficient, and finally a process mapping matrix is generated for batch production control. All sensor data during the process is uploaded to the MES system through the OPC-UA protocol to achieve closed-loop optimization of the process parameters.

[0056] S108. Obtain the performance data of the silicone ball from the surface friction parameters, measure the hardness and gripping force distribution through mechanical testing equipment, and determine whether the silicone ball meets the rehabilitation training needs.

[0057] A friction coefficient measuring instrument was used to sample multiple points on the surface of the silicone ball to obtain friction parameter datasets for different regions. Based on these datasets, a friction distribution map of the silicone ball surface was generated. A Shore hardness tester was used to perform hardness tests at uniformly distributed locations on the silicone ball surface, obtaining a hardness value dataset, and the average hardness value and standard deviation were calculated. A pressure sensor array was used to measure the force distribution of the silicone ball under different gripping forces, generating a gripping force distribution heatmap. A support vector machine algorithm was used to compare the friction parameter dataset with a pre-established set of rehabilitation training standard parameters to determine whether the friction meets the rehabilitation training requirements. A correlation analysis was performed on the hardness value dataset and the gripping force distribution heatmap to establish a hardness-grip force relationship model and determine the quantitative influence relationship between the two. A decision tree algorithm was used to comprehensively analyze the friction distribution map, hardness value dataset, and gripping force distribution heatmap to generate a multi-dimensional evaluation result of the silicone ball's performance. The matching degree between the evaluation results and the rehabilitation training standard requirements was calculated. If the matching degree was higher than a preset threshold, the silicone ball was determined to meet the rehabilitation training requirements; if it was lower than the preset threshold, it was determined not to meet the requirements.

[0058] Specifically, when obtaining performance data of the silicone ball through surface friction parameters, the friction coefficient between the silicone ball and the standard contact surface is first measured using a friction coefficient tester. For example, a friction coefficient of 0.75 indicates that its surface has good anti-slip properties. Next, the hardness of the silicone ball is measured using mechanical testing equipment such as a hardness tester, and a Shore hardness of 50A is obtained. This value indicates that the silicone ball has moderate elasticity, suitable for grasping and squeezing movements in rehabilitation training. To further evaluate the gripping force distribution of the silicone ball, a pressure distribution testing system is used to conduct multi-point pressure tests on the silicone ball. The analysis results show that when a force of 10N is applied, the pressure distribution on the surface of the silicone ball is uniform, with the maximum pressure point at 12kPa and the minimum pressure point at 8kPa. This indicates that the silicone ball can evenly distribute pressure when subjected to force, reducing local stress concentration. Based on this data, the comprehensive performance of the silicone balls was analyzed using algorithms. For example, a weighted average method was used to comprehensively score the friction coefficient, hardness, and pressure distribution data, setting the weight of friction coefficient at 30%, hardness at 40%, and pressure distribution at 30%. The calculated comprehensive score was 85 points, indicating that the silicone balls have good applicability in rehabilitation training. Finally, by comparing with rehabilitation training requirements, such as a friction coefficient greater than 0.7, a hardness between 40A and 60A, and a pressure distribution uniformity within ±20%, it was confirmed that the silicone balls meet all requirements and are suitable for use in rehabilitation training.

[0059] S109. If the test data deviation is less than the preset threshold, an automated sorting system is used to classify the silicone balls by quality, and surface defects are identified by machine vision to obtain finished silicone balls that meet the standards.

[0060] Test data of silicone balls is acquired through sensors, and the deviation between the test data and a preset threshold is calculated. If the deviation is less than the preset threshold, silicone balls that meet the deviation requirements are obtained. A machine vision system scans the surface of the qualified silicone balls to acquire a first image. The first image is processed using a convolutional neural network algorithm to identify surface defects, resulting in a second image containing defect features. Based on the defect features in the second image, the size and location information of the defects are extracted, and it is determined whether the defects exceed a preset defect threshold, thus determining the severity of the defects. Using a pre-established quality classification model and inputting the defect severity, a decision tree algorithm is used to classify the silicone balls according to their quality, obtaining quality grade labels. Based on the quality grade labels, an automated sorting system is driven to perform a classification operation, assigning the silicone balls to the collection areas corresponding to the quality grades, resulting in sorted silicone balls. The classification accuracy rate fed back by the sorting system is obtained. If the accuracy rate is lower than a preset standard, the sorting parameters are optimized using a support vector machine algorithm to obtain an optimized sorting configuration. Using the optimized sorting configuration, the automated sorting operation is re-executed to determine whether all silicone balls meet the standard, resulting in final silicone ball products that meet the standard.

[0061] Specifically, when the test data deviation is less than a preset threshold of 0.05, the system automatically initiates the automated sorting process. First, high-precision sensors collect physical parameters such as the diameter and weight of the silicone balls to ensure they meet the standard range (diameter 20±0.1mm, weight 5±0.05g). Then, the machine vision system uses a defect detection algorithm based on a convolutional neural network (CNN) to perform a comprehensive scan of the silicone ball surface, identifying defects such as scratches and bubbles, with defect areas exceeding 0.1mm. 2 Inferior silicone balls will be marked as defective. The system uses image processing technology to capture multi-angle images of each silicone ball, generating high-resolution images, and uses edge detection algorithms (such as the Canny algorithm) to accurately extract defect contours and calculate defect areas. For silicone balls that meet the standards, the system automatically classifies them as finished products and transports them to the packaging area via conveyor belt. Simultaneously, it records the quality data of each silicone ball and generates a quality report for subsequent traceability and analysis. The entire sorting process employs closed-loop control to ensure sorting accuracy of over 99.9%, and a real-time monitoring system dynamically adjusts sorting efficiency and quality to address variations in the characteristics of different batches of silicone balls.

[0062] This invention provides a silicone ball manufacturing system for rehabilitation training, mainly comprising:

[0063] The hardness gradient requirement acquisition module is used to obtain hardness gradient requirements and surface friction standards from the silicone ball design parameter database.

[0064] The surface friction standard acquisition module is used to analyze the sensitivity range of different muscle groups to changes in hardness during rehabilitation training through algorithms, and to determine the hardness gradient curve and surface texture parameters.

[0065] The algorithm analysis module is used to simulate the internal stress distribution and material flow characteristics of silicone balls using finite element analysis based on the hardness gradient curve, and obtains the mold partitioning scheme and vulcanization temperature field distribution through iterative calculation.

[0066] The finite element analysis module is used to adjust the mold partition boundary and silicone formula ratio if the finite element analysis results show that the interlayer stress difference is greater than the preset threshold. It verifies the interlayer bonding uniformity through secondary simulation and obtains the optimized mold design parameters.

[0067] The mold partitioning adjustment module is used to generate a three-dimensional model of a multi-cavity mold based on the optimized mold design parameters, and to calculate the mold machining trajectory using a CNC machining path planning algorithm to obtain high-precision mold machining instructions.

[0068] The mold processing instruction generation module is used to obtain the geometric data of the processed mold from the mold processing instructions, and to detect the surface error and partition accuracy of the mold by laser scanning to determine whether the mold meets the hardness gradient control requirements.

[0069] The mold detection module is used to control the silicone ball molding process by using a segmented vulcanization process if the mold error is less than a preset threshold. By monitoring the vulcanization temperature and pressure distribution in real time, a silicone ball semi-finished product with continuous hardness variation is obtained.

[0070] The vulcanization process control module is used to adjust the surface molecular structure of silicone ball semi-finished products using plasma treatment technology. By controlling the treatment time and gas ratio, optimized surface friction parameters are obtained.

[0071] The surface treatment module is used to obtain performance data of the silicone ball from surface friction parameters, and to measure hardness and grip force distribution through mechanical testing equipment to determine whether the silicone ball meets the rehabilitation training needs.

[0072] The above embodiments are merely one of the preferred embodiments of the present invention and should not be used to limit the scope of protection of the present invention. Any modifications or refinements made to the main design concept and spirit of the present invention that are not of substantial significance, but solve the same technical problem as the present invention, should be included within the scope of protection of the present invention.

Claims

1. A method for manufacturing a silicone ball for rehabilitation training, characterized in that, The method includes the following steps: S101. Obtain the hardness gradient requirements and surface friction standards from the silicone ball design parameter database, and analyze the sensitivity range of different muscle groups to hardness changes during rehabilitation training through algorithm analysis to determine the hardness gradient curve and surface texture parameters. S102. For the hardness gradient curve, finite element analysis is used to simulate the internal stress distribution and material flow characteristics of the silicone ball. The mold partitioning scheme and vulcanization temperature field distribution are obtained through iterative calculation. S103. If the finite element analysis results show that the interlayer stress difference is greater than the preset threshold, adjust the mold partition boundary and silicone formula ratio, verify the interlayer bonding uniformity through secondary simulation, and obtain the optimized mold design parameters. S104. Based on the optimized mold design parameters, generate a three-dimensional model of the multi-cavity mold, and use a CNC machining path planning algorithm to calculate the mold machining trajectory to obtain high-precision mold machining instructions. S105. Obtain the geometric data of the completed mold from the mold processing instructions, and use laser scanning to detect the surface error and partitioning accuracy of the mold to determine whether the mold meets the hardness gradient control requirements. S106. If the mold error is less than the preset threshold, the silicone ball molding process is controlled by segmented vulcanization process. By monitoring the vulcanization temperature and pressure distribution in real time, a silicone ball semi-finished product with continuous hardness variation is obtained. S107. For silicone ball semi-finished products, plasma treatment technology is used to adjust the surface molecular structure. By controlling the treatment time and gas ratio, optimized surface friction parameters are obtained. S108. Obtain silicone ball performance data from surface friction parameters, measure hardness and grip force distribution using mechanical testing equipment, and determine whether the silicone ball meets the rehabilitation training needs. S109. If the test data deviation is less than the preset threshold, an automated sorting system is used to classify the silicone balls by quality, and surface defects are identified by machine vision to obtain finished silicone balls that meet the standards.

2. The method for manufacturing a silicone ball for rehabilitation training according to claim 1, characterized in that, S101 includes: Hardness gradient data and surface friction standard data are obtained from the silicone ball design parameter database to generate a structured dataset; The correlation between hardness gradient and surface friction in the structured dataset was analyzed using a linear regression algorithm to obtain the mapping relationship between hardness gradient and friction. Based on the mapping relationship, training data for different muscle groups in rehabilitation training are obtained, the influence of hardness gradient changes on the stimulation intensity of muscle groups is analyzed, and the sensitivity range of hardness changes for each muscle group is determined. If the sensitivity interval covers a preset number of muscle groups, then the sensitivity interval is grouped by a clustering algorithm to obtain hardness gradient curves for different muscle groups. A texture generation algorithm is used to optimize the texture parameters corresponding to the surface friction force based on the hardness gradient curve, thereby generating a texture parameter set. By matching the texture parameter set with the hardness gradient curve, it is determined whether the parameter combination meets the friction standard for rehabilitation training, and the final hardness gradient and surface texture parameters are determined. If the final hardness gradient and surface texture parameters meet the standards, the final hardness gradient and surface texture parameters are stored in the database to generate a silicone ball design parameter configuration file.

3. The method for manufacturing a silicone ball for rehabilitation training according to claim 1, characterized in that, S102 includes: Obtain the 3D model data and hardness gradient curve data of the silicone ball to generate the initial silicone ball model; The initial silicone sphere model was meshed using the finite element analysis method, and the internal stress distribution was calculated to obtain stress distribution data. Analyze the material flow characteristics based on the stress distribution data and extract flow characteristic parameters; If the flow characteristic parameters exceed the preset threshold, the hardness gradient curve is adjusted, the finite element analysis is performed again, and iterative optimization is carried out to obtain the optimized stress distribution data and flow characteristic parameters. The mold zoning scheme is determined based on the optimized stress distribution data; The vulcanization process was simulated for the mold partitioning scheme, and the temperature field distribution was calculated to obtain the vulcanization temperature field distribution. A genetic algorithm is used to optimize the vulcanization temperature field distribution, adjust the mold zoning scheme, and determine the final mold zoning scheme and vulcanization temperature field distribution.

4. A method for manufacturing a silicone ball for rehabilitation training according to any one of claims 1-3, characterized in that, S103 includes: By obtaining interlayer stress difference data through finite element analysis, it is determined whether the interlayer stress difference exceeds a preset threshold, and stress distribution analysis results are obtained. If the interlayer stress difference exceeds a preset threshold, an optimization algorithm is used to adjust the mold partition boundary and determine the mold partition boundary parameters. Based on the mold partition boundary parameters, the initial value of the silicone formula ratio is obtained, and the optimized formula ratio is obtained by matching with the material property database. Using the mold partition boundary parameters and the optimized formula ratio, a secondary finite element simulation was performed to obtain interlayer bonding uniformity data. By comparing the interlayer bonding uniformity data with the preset uniformity standard, it is determined whether the requirements are met, and the verification result is obtained. If the verification result meets the preset uniformity standard, the optimized mold design parameters are output to determine the final design parameters; Based on the final design parameters, a digital model of the mold partitions and silicone formula is generated to obtain executable production parameters.

5. A method for manufacturing a silicone ball for rehabilitation training according to any one of claims 1-3, characterized in that, S104 includes: Obtain a set of mold design parameters, which includes the geometric parameters and material parameters of a multi-cavity mold; The set of mold design parameters is analyzed to determine the structural layout of the multi-cavity mold and obtain structural description data. Using a geometric modeling algorithm, a three-dimensional model of a multi-cavity mold is constructed based on the structural description data, resulting in three-dimensional model data; Determine whether the three-dimensional model data meets the preset geometric constraints. If it does, use a path planning algorithm to generate a processing trajectory and obtain trajectory data. If the conditions are not met, the structural description data is adjusted and the 3D model is reconstructed. Based on the trajectory data and combined with precision control rules, the processing trajectory is optimized to obtain high-precision trajectory data; Processing features are extracted from the high-precision trajectory data to generate processing instructions, resulting in an instruction sequence; By verifying the consistency between the processing instructions and the three-dimensional model data through the instruction sequence, the final processing instructions are obtained; Determine whether the final processing instruction meets the preset accuracy threshold. If it does, output a high-precision mold processing instruction. If the requirements are not met, adjust and optimize the parameters, and regenerate the machining trajectory.

6. A method for manufacturing a silicone ball for rehabilitation training according to any one of claims 1-3, characterized in that, S105 includes: The three-dimensional geometric data of the mold after processing is extracted from the mold processing instructions, and the instructions are decomposed using a preset analytical algorithm to obtain the three-dimensional geometric data of the mold. The surface point cloud data of the mold's three-dimensional geometry is obtained by laser scanning detection, and the mold surface error data and partition accuracy data are generated by point cloud registration algorithm. The surface error data of the mold is compared with a preset error threshold. If the surface error data exceeds the preset error threshold, the surface of the mold is determined to be unqualified, and the surface error analysis result is obtained. The partition accuracy data is compared with the preset accuracy partition requirements. If the partition accuracy data meets the preset accuracy partition requirements, the partition accuracy is determined to be qualified, and the partition accuracy analysis result is obtained. The surface of the mold is scanned using laser hardness testing technology to obtain hardness gradient distribution data; The hardness gradient distribution data is compared with the preset hardness gradient threshold. If the hardness gradient distribution data meets the preset hardness gradient threshold, the mold hardness control is deemed qualified, and the hardness control analysis result is obtained. Based on the surface error analysis results, the partitioning accuracy analysis results, and the hardness control analysis results, a logical AND operation is performed. If all three results are qualified, the mold is determined to meet the overall requirements, and the final determination result is obtained.

7. A method for manufacturing a silicone ball for rehabilitation training according to any one of claims 1-3, characterized in that, S106 includes: The system collects mold error data using sensors to determine whether the mold error is less than a preset threshold. If the mold error is less than a preset threshold, then the error distribution characteristics of the mold error data are determined; Based on the error distribution characteristics, a segmented vulcanization process is adopted to generate initial control parameters for the vulcanization temperature and pressure distribution; The system uses a real-time monitoring system to obtain temperature and pressure data from the vulcanizing equipment and determines whether the temperature and pressure data deviate from the initial control parameters. If the temperature or pressure data deviates from the initial control parameters, the vulcanization process parameters are adjusted using a support vector machine algorithm to obtain optimized control parameters. Based on the optimized control parameters, a closed-loop control system is used to adjust the vulcanization equipment to generate silicone ball semi-finished products with continuous hardness variation. Hardness distribution data is obtained from the silicone ball semi-finished product using a hardness testing device to determine whether the hardness distribution data meets the continuity requirements. If the hardness distribution data meets the continuity requirement, the optimized control parameters and the hardness distribution data are recorded through the data storage system to generate the process control log of the silicone ball semi-finished product.

8. A method for manufacturing a silicone ball for rehabilitation training according to any one of claims 1-3, characterized in that, S107 includes: The initial surface molecular structure data and baseline friction parameters of the silicone ball semi-finished product are obtained. The initial surface molecular structure data is obtained by plasma treatment equipment, and the baseline friction parameters are obtained by friction testing equipment. A machine learning model was used to analyze the relationship between the initial surface molecular structure data and the baseline friction parameters to obtain a preliminary optimization range for the processing time and gas ratio. Based on the preliminary optimization range, the processing time and gas ratio of the plasma treatment equipment are adjusted to generate the first set of surface molecular structure data after treatment. The friction parameters of the first set of processed surface molecular structure data are obtained by a friction testing device, and it is determined whether the friction parameters reach a preset threshold. If the frictional force parameter does not reach the preset threshold, the processing time and gas ratio are adjusted by an iterative optimization algorithm to generate a second set of processed surface molecular structure data. The friction parameters of the second set of processed surface molecular structure data are obtained using a friction testing device to determine whether the optimization objective is met. Based on the judgment results, data fusion technology is used to integrate the first set of processed surface molecular structure data and the second set of processed surface molecular structure data and corresponding friction parameters to determine the final processing time and gas ratio configuration. By applying the final processing time and gas ratio configuration using a plasma treatment device, the semi-finished silicone balls are processed to obtain optimized surface friction parameters.

9. A method for manufacturing a silicone ball for rehabilitation training according to any one of claims 1-3, characterized in that, S108 includes: A friction coefficient measuring instrument was used to sample multiple points on the surface of the silicone ball to obtain a dataset of friction force parameters in different areas. Based on the friction parameter dataset, a friction distribution map of the silicone ball surface is generated; Hardness tests were performed on evenly distributed locations on the surface of a silicone ball using a Shore hardness tester to obtain a dataset of hardness values, and the average hardness value and standard deviation were calculated. A pressure sensor array was used to measure the force distribution of a silicone ball under different gripping forces, and a heat map of gripping force distribution was generated. The support vector machine algorithm is used to compare the friction parameter dataset with the pre-established rehabilitation training standard parameter set to determine whether the friction meets the rehabilitation training requirements. A correlation analysis was performed on the hardness value dataset and the gripping force distribution heatmap to establish a hardness-grip force relationship model and determine the quantitative influence relationship between the two. Based on the decision tree algorithm, the friction force distribution map, hardness value dataset and grip force distribution heat map are comprehensively analyzed to generate multi-dimensional evaluation results of silicone ball performance. The matching degree is calculated based on the assessment results and the requirements of rehabilitation training standards. If the matching degree is higher than the preset threshold, the silicone ball is determined to meet the rehabilitation training requirements. If the value is below the preset threshold, it is determined that the requirement is not met.

10. A silicone ball manufacturing system for rehabilitation training, characterized in that, This system is used to implement a method for manufacturing a silicone ball for rehabilitation training as described in any one of claims 1-9, the system comprising: The hardness gradient requirement acquisition module is used to obtain hardness gradient requirements and surface friction standards from the silicone ball design parameter database. The surface friction standard acquisition module is used to analyze the sensitivity range of different muscle groups to changes in hardness during rehabilitation training through algorithms, and to determine the hardness gradient curve and surface texture parameters. The algorithm analysis module is used to simulate the internal stress distribution and material flow characteristics of silicone balls using finite element analysis based on the hardness gradient curve, and obtains the mold partitioning scheme and vulcanization temperature field distribution through iterative calculation. The finite element analysis module is used to adjust the mold partition boundary and silicone formula ratio if the finite element analysis results show that the interlayer stress difference is greater than the preset threshold. It verifies the interlayer bonding uniformity through secondary simulation and obtains the optimized mold design parameters. The mold partitioning adjustment module is used to generate a three-dimensional model of a multi-cavity mold based on the optimized mold design parameters, and to calculate the mold machining trajectory using a CNC machining path planning algorithm to obtain high-precision mold machining instructions. The mold processing instruction generation module is used to obtain the geometric data of the processed mold from the mold processing instructions, and to detect the surface error and partition accuracy of the mold by laser scanning to determine whether the mold meets the hardness gradient control requirements. The mold detection module is used to control the silicone ball molding process by using a segmented vulcanization process if the mold error is less than a preset threshold. By monitoring the vulcanization temperature and pressure distribution in real time, a silicone ball semi-finished product with continuous hardness variation is obtained. The vulcanization process control module is used to adjust the surface molecular structure of silicone ball semi-finished products using plasma treatment technology. By controlling the treatment time and gas ratio, optimized surface friction parameters are obtained. The surface treatment module is used to obtain performance data of the silicone ball from surface friction parameters, and to measure hardness and grip force distribution through mechanical testing equipment to determine whether the silicone ball meets the needs of rehabilitation training.

Images