A tire lifecycle supply chain management method and system

Abstract

This application relates to a tire lifecycle supply chain management method and system, including intelligent sensor deployment, data processing, parallel computing, data standardization, traceability map construction, storage, and visualization steps. This invention collects tire pressure and internal structure data in real time, utilizes edge computing and distributed computing clusters for data processing and feature extraction to ensure data consistency and anomaly handling, constructs a data traceability map to locate problem nodes, stores the processed data in a distributed database, generates a tire 3D model and performs rasterization processing to obtain a tire 3D image, and realizes visualized monitoring of tire status to optimize supply chain management.

Description

Technical Field

[0001] This application relates to the field of information technology, specifically to a method and system for tire lifecycle supply chain management. Background Technology

[0002] In electronic digital data processing and vehicle tire lifecycle supply chain management systems, there are technical challenges in the real-time processing and accuracy assurance of massive tire lifecycle data. Tires generate a large amount of data at every stage, including production, transportation, storage, use, maintenance, and scrapping. This data is characterized by its high timeliness, diverse formats, and dispersed sources. How to efficiently collect, transmit, and process this data is a huge challenge. At the same time, due to the harsh environment in which tires are used, problems such as noise, packet loss, and distortion are prone to occur during data collection, leading to inaccurate data and affecting subsequent analysis and decision-making.

[0003] In addition, tire life cycle data has obvious time-series characteristics. How to utilize the time sequence and correlation of the data to build a reasonable data model and algorithm, realize real-time data verification and error correction, and thus improve data quality and system reliability is also a technical problem that urgently needs to be solved.

[0004] The solution proposed by this invention to address the shortcomings of the above-mentioned content is as follows: intelligent sensors collect tire data, edge computing nodes perform data preprocessing, distributed computing clusters perform parallel computing, and problem nodes are located through data traceability maps. Finally, the tire data is stored and rendered to generate a 3D model. This invention includes steps such as data collection, filtering, preprocessing, parallel computing, data unification, problem location, storage, and visualization to ensure the efficiency and accuracy of data processing. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, the purpose of this application is to provide a tire full life cycle supply chain management method and system.

[0006] The tire lifecycle supply chain management method described in this application includes the following steps:

[0007] S1. Deploy intelligent sensors with multi-core heterogeneous parallel processing capabilities. The intelligent sensors collect raw data of tire pressure in real time and obtain data on the internal structure of the tire.

[0008] S2. The edge computing node filters the raw tire pressure data in step S1 to remove redundant pressure information and preprocesses the tire internal structure data in step S1 to generate a tire internal structure feature map.

[0009] S3. The raw data of tire pressure and the tire internal structure feature map from step S2 are uploaded to the distributed computing cluster. The distributed computing framework distributes the raw data of tire pressure and the tire internal structure feature map to different computing nodes for parallel computing.

[0010] S4. If the multi-threaded hardware of the compute node fails, the dynamic branch prediction mechanism is activated, which distributes the data that the node is responsible for processing to other normal nodes, and uses the cache consistency protocol to ensure data consistency.

[0011] S5. After each calculation node completes the calculation of the raw data of tire pressure and the tire internal structure feature map, the format of the raw data of tire pressure and the tire internal structure feature map is unified through a unified data standard.

[0012] S6. Based on the unified original tire pressure data and the tire internal structure feature map, construct a tire data traceability map to record the entire process from data collection to processing. If an abnormality occurs in the data processing stage, locate the problem node based on the traceability map.

[0013] S7. Store the processed raw tire pressure data and tire internal structure feature map into a distributed database, and render the raw tire pressure data and tire internal structure feature map to obtain visualization results.

[0014] S8. The visualization server generates a three-dimensional tire model based on the original tire pressure data and the tire's internal structural feature map. The three-dimensional model is then rasterized to obtain a three-dimensional tire image.

[0015] S9. Map the raw tire pressure data and the tire internal structure feature map to the tire 3D model, render the mapped model, and obtain the tire visualization results.

[0016] Preferably, in step S1, an intelligent sensor integrating a multi-core heterogeneous processor, pressure and image sensors is deployed to collect tire pressure data and internal structure images in real time, upload them to the data processing center, monitor pressure anomalies and issue alarms, process images to extract texture features, perform spectrum analysis for early warning, and fuse pressure and image data.

[0017] Preferably, in step S2, the edge node receives and groups the tire pressure data uploaded by the vehicle-mounted sensor, filters and fuses the effective data to remove noise, analyzes the trend of changes in the data inside the tire, processes the features through a convolutional neural network, performs pooling and classification to determine the internal structural features of the tire, uses a recurrent neural network for time series analysis, constructs a structural feature change model and determines the state, and finally, the processed data and feature map are stored and visualized.

[0018] Preferably, in step S3, the edge node receives and groups the tire pressure data uploaded by the vehicle-mounted sensor, filters and removes noise to obtain effective data, processes the data features through a convolutional neural network, performs pooling and classification to identify the internal structural features of the tire, uploads the data to a distributed computing cluster for parallel computing, identifies the degree of deformation in the feature map based on the tire pressure data and internal structure relationship model, and uses a regression model to correct the pressure data.

[0019] Preferably, in step S4, the following steps are taken: monitoring the running information of computing nodes, recording thread status and handling exceptions, determining the set of normal nodes, formulating a data allocation scheme and triggering data migration based on node load and data volume, in order to achieve data synchronization and cache consistency maintenance between nodes, collecting tire pressure and structural feature data, assigning unique identifiers and recording information.

[0020] Preferably, in step S5, the raw tire pressure data and tire internal structure feature map format uploaded by each computing node are collected and standardized. If the data format does not conform to the standard, the format is converted, the structural feature map is analyzed using an image recognition algorithm, the pressure data is corrected by combining a regression model, the corrected pressure data and feature map are converted into a standardized format, feature vectors are extracted, and they are packaged into a data package with a unified data standard according to preset rules for subsequent processing.

[0021] Preferably, in step S6, raw tire pressure data and tire internal structure feature maps are collected, a unique identifier is assigned and a timestamp is recorded, the data preprocessing module cleans and calibrates the data, extracts key features to form feature vectors, stores normal data in the database, and generates a data traceability map. Abnormal data triggers an alarm, and the problem node is traced back and located through the identifier. Abnormal pressure data is further analyzed to determine the degree of deformation, and the traceability map and abnormal nodes are displayed through a graphical interface.

[0022] Preferably, in step S7, tire pressure data and internal structural feature maps completed by the computing nodes are collected. If the pressure data is abnormal, the deformation in the structural map is analyzed by an image recognition algorithm, and the pressure data is corrected. The corrected pressure data is standardized, and the feature vector of the structural map is extracted. The standardized pressure data and the structural map are combined into a data packet, which is then segmented and verified according to the database protocol and stored. The pressure data is visualized into a curve graph by the rendering engine, and the structural feature map is rendered.

[0023] Preferably, in step S8, the processor obtains a tire model from the model library, uses the graphics library and combines it with the internal structural feature map information to construct a three-dimensional tire model. If the pressure is abnormal, the model is adjusted to reflect the deformation. The model is rasterized to obtain an initial image. If the resolution is insufficient, it is improved by a super-resolution algorithm. The processed image is used to eliminate jagged edges, reduce noise and enhance color to obtain a three-dimensional tire image, which is then output by the visualization server.

[0024] Preferably, in step S9, tire pressure data and internal structural feature maps are acquired, and the data format is determined. These data are used to construct the mapping relationship and initial model data of the tire's three-dimensional model, thereby forming the tire's geometric model. After verifying the geometric model, attribute mapping is performed to generate an attribute model, and a lighting model is created to achieve the lighting effect. If the lighting effect meets the visual characteristics, rasterization processing is performed to generate a two-dimensional image, and image rendering is completed to visualize the tire's three-dimensional model. Finally, the pressure data, structural feature maps, and visualization results are stored in a distributed database, and the storage structure is determined.

[0025] The tire lifecycle supply chain management system described in this application includes: an intelligent sensor module, an edge computing node module, a distributed computing cluster module, a data standardization module, a data traceability construction module, a distributed database storage module, a tire 3D model generation module, and a tire 3D model optimization module.

[0026] The intelligent sensor module is connected to the edge computing node module to collect tire pressure data and internal structure images in real time, and uploads them to the data processing center through the communication module in the intelligent sensor.

[0027] The edge computing node module is connected to the smart sensor module and the distributed computing cluster module. It is used to receive tire pressure data from the smart sensor, process filtering, noise reduction, and feature extraction, and perform pooling and classification through a convolutional neural network to determine the internal structural features of the tire. The processed tire pressure data and internal structural feature map are stored in a distributed database and uploaded to the distributed computing cluster. The distributed computing framework distributes the pressure data and internal structural data to different computing nodes.

[0028] The distributed computing cluster module is connected to the edge computing node module and the data standardization module. It is used to receive data uploaded by the edge nodes, distribute it to different computing nodes for parallel computing, and perform exception handling. If a computing node fails, a dynamic branch prediction mechanism is activated to reallocate the data.

[0029] The data standardization module is connected to the distributed computing cluster module and the data traceability construction module, and is used to convert data into data with a unified data standard.

[0030] The data traceability construction module is connected to the data standardization module and the distributed database storage module. It is used to construct a tire data traceability map based on the standardized data. When data processing is abnormal, the problem node is located through the traceability map. The original data is transmitted to the data preprocessing module, which judges the data integrity according to the rule base; normal data is stored, and the data is associated with the corresponding identifier and timestamp and stored in the distributed database.

[0031] The distributed database storage module is connected to the data traceability construction module and the tire 3D model generation module. It is used to store the processed data in the distributed database, visualize the pressure data through the rendering engine, and render the structural feature map. In this process, the standard data packet is divided into multiple data blocks according to the data storage protocol of the distributed database.

[0032] The tire 3D model generation module is connected to the distributed database storage module and the tire 3D model optimization module. It is used to construct a tire 3D model using a graphics library and internal structural feature map information. The tire 3D image after rasterization is output by the visualization server. The processor accesses the model library to obtain the tire model file and uses the graphics library to modify the tire 3D model to obtain a deformable tire 3D model.

[0033] The tire 3D model optimization module is connected to the tire 3D model generation module. It is used to construct the mapping relationship and initial model data of the tire 3D model, create a lighting model, realize the lighting effect, and perform rasterization processing to generate a 2D image. Pressure data, internal structural feature maps and visualization result images are stored in a distributed database.

[0034] The tire lifecycle supply chain management method and system described in this application have the following advantages: Addressing the challenges of processing multi-source heterogeneous tire data and the requirements for reliability and traceability during data processing, this invention deploys intelligent sensors with multi-core heterogeneous parallel processing capabilities to collect tire pressure and internal structure data in real time. It utilizes edge computing nodes for data preprocessing, a distributed computing cluster for parallel computation, and introduces a dynamic branch prediction mechanism and a high-speed cache consistency protocol to ensure the reliability of the computation process. Simultaneously, it constructs a tire data traceability map to achieve traceability of the data processing process. Finally, by rendering the processed data, a three-dimensional tire model is generated and visualized, providing an intuitive presentation of the tire's condition.

[0035] This invention effectively solves the challenges of tire lifecycle data collection, processing, reliability assurance, traceability, and visualization, thereby improving the efficiency and accuracy of tire condition monitoring. Attached Figure Description

[0036] Figure 1This application describes a tire lifecycle supply chain management method. Figure 1 ;

[0037] Figure 2 This application describes a tire lifecycle supply chain management method. Figure 2 ;

[0038] Figure 3 This is a structural block diagram of a tire lifecycle supply chain management system as described in this application. Detailed Implementation

[0039] like Figures 1-2 As shown, the tire lifecycle supply chain management method described in this application includes the following steps:

[0040] S1. Deploy intelligent sensors with multi-core heterogeneous parallel processing capabilities. The intelligent sensors collect raw data of tire pressure in real time and obtain data on the internal structure of the tire.

[0041] S2. The edge computing node filters the raw tire pressure data in step S1 to remove redundant pressure information and preprocesses the tire internal structure data in step S1 to generate a tire internal structure feature map.

[0042] S3. The raw data of tire pressure and the tire internal structure feature map from step S2 are uploaded to the distributed computing cluster. The distributed computing framework distributes the raw data of tire pressure and the tire internal structure feature map to different computing nodes for parallel computing.

[0043] S4. If the multi-threaded hardware of the compute node fails, the dynamic branch prediction mechanism is activated, which distributes the data that the node is responsible for processing to other normal nodes, and uses the cache consistency protocol to ensure data consistency.

[0044] S5. After each calculation node completes the calculation of the raw data of tire pressure and the tire internal structure feature map, the format of the raw data of tire pressure and the tire internal structure feature map is unified through a unified data standard.

[0045] S6. Based on the unified original tire pressure data and the tire internal structure feature map, construct a tire data traceability map to record the entire process from data collection to processing. If an abnormality occurs in the data processing stage, locate the problem node based on the traceability map.

[0046] S7. Store the processed raw tire pressure data and tire internal structure feature map into a distributed database, and render the raw tire pressure data and tire internal structure feature map to obtain visualization results.

[0047] S8. The visualization server generates a three-dimensional tire model based on the original tire pressure data and the tire's internal structural feature map. The three-dimensional model is then rasterized to obtain a three-dimensional tire image.

[0048] S9. Map the raw tire pressure data and the tire internal structure feature map to the tire 3D model, render the mapped model, and obtain the tire visualization results.

[0049] like Figures 1-2 As shown, in step S1, a smart sensor is deployed, which integrates a multi-core heterogeneous processor, a pressure sensor, and an image sensor;

[0050] The intelligent sensor contains a pressure sensor that collects tire pressure data in real time to obtain the raw tire pressure data.

[0051] The image sensor inside the smart sensor acquires images of the tire's internal structure, obtaining the original structural image.

[0052] The intelligent sensor integrates a communication module to upload raw tire pressure data to the data processing center. If the raw tire pressure data is lower than a preset pressure threshold, an alarm mechanism is triggered.

[0053] The image processor within the smart sensor segments the original image and identifies multiple sub-images.

[0054] The data processing center receives raw tire pressure data. If the raw tire pressure data deviates from the historical data by more than a set value, the data is marked as abnormal.

[0055] The data processing center adopts a multi-core heterogeneous parallel processing architecture to obtain the texture feature information of each sub-image.

[0056] The data processing center performs spectrum analysis based on texture feature information. If the difference between the spectrum analysis result and the standard spectrum is greater than a set range, an early warning mechanism is triggered.

[0057] The data processing center merges the original pressure data and the original structural image to obtain fused data.

[0058] Specifically, in step S1, a smart sensor is deployed that integrates a multi-core heterogeneous processor, a pressure sensor, and an image sensor, including a Renesas Electronics R-Car series chip that integrates multiple ARM Cortex-A57 and Cortex-A53 cores, and an Imagination Technologies PowerVR GX6650 GPU.

[0059] The intelligent sensor uses a piezoresistive pressure sensor with a sampling frequency of 100Hz to collect tire pressure data in real time and obtain the raw tire pressure data, including 2.5 bar.

[0060] The image sensor inside the smart sensor uses a CMOS image sensor with a resolution of 1280x720 pixels to capture images of the internal structure of the tire and obtain the original image of the structure.

[0061] The smart sensor integrates a 4G communication module to upload raw pressure data to the data processing center. If the raw pressure data is lower than the preset pressure threshold of 2.0 bar, an alarm mechanism is triggered.

[0062] The image processor in the smart sensor uses an edge detection-based image segmentation algorithm, including the Canny algorithm, to segment the original structural image into multiple sub-images, identifying nine sub-images.

[0063] The data processing center receives raw pressure data. If the deviation between the raw pressure data and historical data is greater than the set value of 0.2 bar, the data is marked as abnormal.

[0064] The data processing center adopts a multi-core heterogeneous parallel processing architecture, utilizes the OpenCL framework, calls GPU resources, and employs the gray-level co-occurrence matrix algorithm to obtain texture feature information such as contrast, energy, and entropy of each sub-image;

[0065] The data processing center performs fast Fourier transform spectrum analysis based on texture feature information. If the proportion of high-frequency components in the spectrum analysis results differs from the proportion of high-frequency components in the standard spectrum by more than 10%, an early warning mechanism is triggered.

[0066] The data processing center uses a wavelet transform-based image fusion algorithm to fuse the original pressure data and the original structural image to obtain fused data.

[0067] like Figures 1-2 As shown, in step S2, the edge node receives the raw tire pressure data uploaded by the vehicle sensor, groups the data according to the time series, and obtains a set of tire pressure data for different time periods.

[0068] The tire pressure data set for each time period is analyzed by data filtering. If the fluctuation range of the pressure data value is less than the tire pressure threshold, the pressure data set for that time period is considered redundant information. The redundant information is discarded and the effective pressure data is obtained. The effective pressure data is then fused using a preprocessing layer to obtain the tire internal data after removing noise signals.

[0069] Based on the trend of changes in the intratibial data, extract the characteristics of changes in the intratibial data. If the trend of changes in the intratibial data shows that the fluctuation range of the tire pressure data value is greater than the tire pressure threshold, then the pressure data set within this time period is determined to be valid tire pressure data.

[0070] The first convolutional neural network is used to perform convolution calculation on the changes in fetal data to obtain multiple first convolutional feature maps. The first pooling process is then performed on the first convolutional feature maps to obtain the first pooling feature map.

[0071] The first pooling feature map is convolved by the second convolutional neural network to obtain multiple second convolutional feature maps. The second pooling is then performed on the second convolutional feature maps to obtain the second pooling feature map.

[0072] The second pooling feature map is classified by a fully connected layer to obtain a third feature map. Based on the third feature map, the internal structural features of the tire are determined.

[0073] By performing time-series analysis on structural features using a recurrent neural network, a time-series variation model of the internal structural features of the tire is constructed, and the internal structural state of the tire is determined based on the time-series variation model.

[0074] The processed tire pressure data and internal structure feature maps are stored in a distributed database, and the tire pressure data and internal structure feature maps are processed to obtain visualization results.

[0075] Specifically, in step S2, the edge node receives raw tire pressure data uploaded by the vehicle sensor in real time, including receiving 100 pressure data points per second, and grouping these data points into 10 time periods according to the timestamp to obtain a set of tire pressure data for 10 time periods.

[0076] The tire pressure data set for each time period is analyzed, including calculating the standard deviation of the tire pressure data for each time period. If the standard deviation for a certain time period is less than the preset threshold of 0.1 PSI, the pressure data set for that time period is determined to be redundant information.

[0077] Discarding this redundant information, only retaining the effective pressure data with a standard deviation greater than 0.1 PSI, and using the Kalman filter algorithm to fuse the effective pressure data, we obtain the tire intra-tire data after removing the noise signal;

[0078] Based on the trend of changes in in-tire data, including tire pressure values ​​rising by more than 2 PSI within 1 minute, the characteristics of changes in in-tire data are extracted. If the trend of changes in in-tire data shows that the fluctuation range of tire pressure data values ​​is greater than the tire pressure threshold, then the pressure data set within this time period is determined to be valid tire pressure data.

[0079] The intrauterine data variation features are calculated by convolution using a first convolutional neural network containing 64 3x3 convolutional kernels to obtain 64 first convolutional feature maps. Max pooling with a stride of 2 is then performed on the first convolutional feature maps to obtain the first pooling feature maps.

[0080] The first pooling feature map is convolved by a second convolutional neural network containing 128 3x3 convolutional kernels to obtain 128 second convolutional feature maps. Then, average pooling with a stride of 2 is performed on the second convolutional feature maps to obtain the second pooling feature map.

[0081] The second pooling feature map is classified by a fully connected layer containing 1024 neurons to obtain the third feature map. Based on the third feature map, the internal structural features of the tire, including the deformation patterns of the tire crown or sidewall, are determined.

[0082] By performing time-series analysis on structural features using a Long Short-Term Memory (LSTM) network, a time-series change model of the internal structural features of the tire is constructed. Based on the time-series change model, the internal structural state of the tire is determined, including whether there is slow air leakage.

[0083] The processed tire pressure data and internal structure feature maps are stored in a Hadoop-based distributed database. The tire pressure data and internal structure feature maps are then rendered using WebGL to obtain visualization results.

[0084] like Figures 1-2 As shown, in step S3, the edge node receives the raw tire pressure data uploaded by the vehicle sensor, groups the data according to the time series, and obtains a set of tire pressure data for different time periods.

[0085] By analyzing the tire pressure data set within each time period through data filtering, if the fluctuation range of the pressure data value is less than the tire pressure threshold, the pressure data set within that time period is determined to be redundant information, and the redundant information is discarded to obtain the valid pressure data.

[0086] The effective pressure data is fused using a preprocessing layer to remove noise signals and obtain denoised in-tire data. Based on the trend of in-tire data changes, the characteristics of in-tire data changes are extracted.

[0087] The first convolutional neural network is used to perform convolution calculation on the changes in fetal data to obtain multiple first convolutional feature maps. The first pooling process is then performed on the first convolutional feature maps to obtain the first pooling feature map.

[0088] The first pooling feature map is convolved by the second convolutional neural network to obtain multiple second convolutional feature maps. The second pooling is then performed on the second convolutional feature maps to obtain the second pooling feature map.

[0089] The second pooling feature map is classified by a fully connected layer to obtain a third feature map. Based on the third feature map, the internal structural features of the tire are determined.

[0090] Tire pressure data and tire internal structure feature map are uploaded to a distributed computing cluster. The distributed computing framework distributes the pressure data and internal structure data to different computing nodes for parallel computing to obtain the tire pressure data calculated by each computing node.

[0091] Based on the pre-established model of the relationship between tire pressure data and internal structure, the range of tire pressure data is determined. If the range of tire pressure data exceeds the preset threshold range, the internal structure feature map of the tire at the corresponding moment of the calculation node is obtained, and the internal structure feature map of the calculation node is obtained.

[0092] Pre-trained image recognition algorithms identify pre-defined marked regions in feature maps, determine the degree of deformation in these regions, and then use a regression model based on the relationship between the degree of deformation and pressure data to obtain corrected pressure data.

[0093] Specifically, in step S3, the edge node receives the raw tire pressure data uploaded by the vehicle sensor, including sampling 100 times per second, and grouping the data according to the time series, including grouping every 5 minutes to obtain tire pressure data sets for different time periods;

[0094] By analyzing the tire pressure data set within each time period through data filtering, if the fluctuation range of the pressure data value is less than the preset 0.5PSI tire pressure threshold, the pressure data set within that time period is determined to be redundant information, and the redundant information is discarded to obtain the valid pressure data.

[0095] The effective pressure data is fused using a preprocessing layer, and noise signals are removed by applying a Kalman filter algorithm to obtain denoised in-tire data. Based on the trend of in-tire data changes, features of in-tire data changes, such as the rate of pressure increase or decrease, are extracted.

[0096] The first convolutional neural network is used to perform convolutional calculations on the changes in fetal data. A 3x3 convolutional kernel is used with a stride of 1 to obtain multiple first convolutional feature maps. The first pooling process is then performed based on the first convolutional feature maps using 2x2 max pooling to obtain the first pooling feature map.

[0097] The first pooling feature map is convolved by the second convolutional neural network, using a 3x3 convolutional kernel with a stride of 1, to obtain multiple second convolutional feature maps. The second pooling is then performed on the second convolutional feature maps using 2x2 average pooling to obtain the second pooling feature map.

[0098] The second pooling feature map is classified through a fully connected layer, including the use of a Softmax classifier, to obtain a third feature map. Based on the third feature map, the internal structural features of the tire are determined.

[0099] Tire pressure data and tire internal structure feature maps are uploaded to a distributed computing cluster, including a Hadoop cluster. The distributed computing framework distributes the pressure data and internal structure data to different computing nodes, including 10 computing nodes, for parallel computing. The MapReduce framework is used to obtain the tire pressure data calculated by each computing node.

[0100] Based on the pre-established model of the relationship between tire pressure data and internal structure, including the support vector machine model, the range of tire pressure data is determined. If the range of tire pressure data exceeds the preset threshold range of 2.0-2.5 PSI, the internal structure feature map of the tire at the corresponding time of the calculation node is obtained, and the internal structure feature map of the calculation node is obtained.

[0101] Using pre-trained image recognition algorithms, including Faster R-CNN, the system identifies pre-defined marked regions in the feature map and determines the degree of deformation of these regions, including a deformation degree of 10%. Based on the degree of deformation of the marked regions, a regression model based on the relationship between the degree of deformation and the stress data, including a linear regression model, is used to obtain corrected stress data.

[0102] like Figures 1-2 As shown, in step S4, the runtime information of the computing node is obtained, the thread state is recorded in the processor register, and if the thread state is abnormal, the interrupt handler is triggered to obtain the abnormal thread information.

[0103] Based on the abnormal thread information, the interrupt handler reads the pre-established node status list. If the node status information in the list indicates that there are normal nodes, the set of normal nodes participating in the calculation is determined.

[0104] Based on the set of normal nodes participating in the computation, the execution unit counts the amount of data to be processed for each thread, and obtains the average amount of data based on the amount of data to be processed and the number of normal nodes participating in the computation.

[0105] The average data volume and the current load of each normal node are used to determine the data allocation scheme. The data allocation scheme includes the mapping relationship between data blocks and target nodes, and a data block redistribution list is obtained.

[0106] Based on the data block reallocation list, the fault detection module triggers a data migration instruction. The data migration instruction contains the data block identifier and the target node address, thus obtaining the set of data blocks to be migrated.

[0107] Based on the set of data blocks to be migrated, after the data migration instruction is triggered, the original node sends the data blocks to the target node. After receiving the data blocks, the target node sends a cache update request through the internal bus and obtains the inter-node data synchronization instruction.

[0108] According to the data synchronization instructions between nodes, the target node's cached data is updated. The synchronization module compares the cached information of each node. If the cached information is different, the cache consistency protocol is triggered, and a cache consistency maintenance instruction is obtained.

[0109] Obtain raw tire pressure data and tire internal structure feature map. Assign a unique identifier to each data point. The identifier set is S. Record the data collection time and data source information based on the identifier and timestamp to obtain the raw tire data.

[0110] According to the cache consistency maintenance instructions, update the cache data of each node to ensure data consistency, and dynamically adjust the branch prediction mechanism according to the data allocation scheme to determine the data processing flow.

[0111] Specifically, in step S4, the computing node continuously monitors the thread status through internal sensors, including a register status value of 0x00FF indicating a thread abnormality, triggering an interrupt handler with interrupt vector number 15 to obtain abnormal thread information.

[0112] The interrupt handler accesses the node status list. If the list shows that node A is in a normal state, node B is in an abnormal state, and node C is in a normal state, then it determines that nodes A and C participate in the calculation.

[0113] The execution unit counts the amount of data to be processed for each thread, including 10MB for node A, 15MB for node B, and 5MB for node C. The average data amount is (10+15+5) / 2=15MB.

[0114] Node A is currently at 60% load and Node C is currently at 40% load. Based on the load balancing algorithm, including the weighted round-robin algorithm, it is determined that 15MB of data from Node B will be allocated to Node C for processing.

[0115] The fault detection module generates data migration instructions, including instructions in the format "MOVE DATA_BLOCK_ID=001TARGET_NODE=C", to obtain instructions for migrating data block 001 to node C;

[0116] Data block 001 is migrated from node B to node C. After receiving it, node C sends a cache update request, including a request to update the data at cache line address 0x1000.

[0117] The synchronization module compares the cache information of nodes A and C and finds that the data at cache line address 0x1000 of node C is inconsistent, triggering the MESI protocol.

[0118] Obtain tire pressure data, including a tire pressure value of 2.5 bar, an internal structure feature map resolution of 1024x1024 pixels, an assignment identifier of Tire_001, and a timestamp of 2023-10-27 10:00:00, to obtain the raw tire data;

[0119] According to the MESI protocol, the data at cache line address 0x1000 of node C is updated to be consistent with that of node A. Based on the data allocation scheme, the dynamic branch prediction mechanism reassigns the tasks originally allocated to node B to nodes A and C, thus determining a new data processing flow.

[0120] like Figures 1-2 As shown, in step S5, the original data of tire pressure and the tire internal structure feature map calculated by each computing node are obtained. According to the pre-established data format specifications, the standard format of the original data of tire pressure and the tire internal structure feature map is determined. The format of the pressure data uploaded by each computing node is identified through the preset data format template. If the format of the pressure data is inconsistent with the standard format, the format conversion program is started to obtain the pressure data in the standard format.

[0121] Based on the preset image format standard for internal structure feature maps, the format of the internal structure feature maps uploaded by each computing node is identified. If the format of the internal structure feature map is inconsistent with the standard format, the image format conversion program is started to obtain an internal structure feature map in a unified format.

[0122] Obtain the pressure data after format conversion, and determine the range of tire pressure data based on the pre-established model of the relationship between tire pressure data and internal structure;

[0123] If the tire pressure data exceeds the preset threshold range, the tire internal structure feature map at the corresponding moment of the calculation node is obtained, and the preset marked area in the feature map is identified by a pre-trained image recognition algorithm to determine the degree of deformation of the marked area.

[0124] Based on the degree of deformation in the marked area, a regression model based on the relationship between the degree of deformation and the pressure data is used to obtain corrected pressure data and update the original pressure data;

[0125] By using a preset data format template, the pressure data corrected by each computing node is transformed into the first pressure data in a standardized format. Then, an image feature extraction algorithm based on a convolutional neural network is used to extract the feature vectors of the internal structure feature maps uploaded by each computing node, and the feature vector set is determined.

[0126] Based on the preset image format standard of the internal structure feature map, the internal structure feature map uploaded by each computing node is converted into a second internal structure feature map in a unified format. The first pressure data and the second internal structure feature map are obtained to determine the next data processing content.

[0127] According to the preset data packaging rules, the first pressure data and the second internal structure feature map are combined into a standardized data package to obtain the final output result of the unified data standard.

[0128] Specifically, in step S5, the tire pressure data calculated by each calculation node is first obtained, including the pressure data of node A as 3.5MPa and node B as 2.8MPa, as well as their corresponding internal structure feature diagrams. According to the preset data format specifications, including the pressure data uniformly using MPa unit and the internal structure feature diagrams uniformly using JPEG format, the standard format of these data is determined.

[0129] The system identifies the format of the pressure data uploaded by each node by using preset data format templates, including XML templates. If the pressure data of node A conforms to the XML template, while the pressure data of node B is in CSV format, the system starts the format conversion program to convert the CSV format of node B to XML format, thus obtaining pressure data in a standardized format.

[0130] Based on the preset image format standards, including the requirement that all images be in JPEG format with 1024x768 pixels, the internal structure feature map format uploaded by each node is identified. If the feature map of node A is in PNG format and the feature map of node B conforms to the JPEG standard, then the image format conversion program is started to convert the PNG format of node A to JPEG format to obtain an internal structure feature map in a unified format.

[0131] Obtain the pressure data after format conversion, including 3.5MPa for node A and 2.8MPa for node B. Based on the tire pressure data and internal structure relationship model, including the condition that the pressure is greater than 3.0MPa, it is determined that the pressure data range of node A is high pressure and node B is normal.

[0132] If the pressure data of node A exceeds the preset threshold range (including 3.0 MPa), the tire internal structure feature map of node A at the corresponding time is obtained. The preset marked area in the feature map, including the marked point at a specific location, is identified by a pre-trained image recognition algorithm, including a ResNet-based image classification model. The degree of deformation of the marked area is determined, including a 10% change in the spacing between the marked points.

[0133] Based on the 10% deformation degree of the marked area, a regression model based on the relationship between the deformation degree and pressure data is adopted, including the linear regression model y = kx + b, where y is the corrected pressure, x is the deformation degree, and k and b are model parameters, to obtain corrected pressure data, including 3.6 MPa, and update the original pressure data;

[0134] Using a preset data format template, the corrected pressure data of each node, including 3.6 MPa for node A and 2.8 MPa for node B, is converted into first pressure data in a standardized format. An image feature extraction algorithm based on a convolutional neural network, including the VGG-16 network, is used to extract the feature vectors of the internal structure feature maps uploaded by each node, including obtaining a 1x4096-dimensional feature vector and determining the feature vector set.

[0135] Based on the preset image format standard of the internal structure feature map, the internal structure feature map uploaded by each node is converted into a second internal structure feature map in a unified format. The first pressure data and the second internal structure feature map are obtained, and the next data processing content is determined to be data packaging.

[0136] According to the preset data packaging rules, including packaging stress data and feature map data in JSON format, combining the first stress data with the second internal structure feature map into a standardized data package, the final output result of the unified data standard is obtained, including generating a JSON file containing stress data and feature map data.

[0137] like Figures 1-2 As shown, in step S6, the raw data of tire pressure and the internal structure feature map of the tire are obtained. Each data is assigned a unique identifier, the identifier set is S, the number of internal structure feature maps is N, the i-th feature map is represented as Xi, and the set is represented as X = {X1, X2, ... XN}. The data collection time and data source information are recorded according to the identifier and timestamp to obtain the raw data.

[0138] The raw data is transmitted to the data preprocessing module for data cleaning, removal of duplicate and erroneous data, and data integrity is determined based on the rule base. If data is missing, it is marked to obtain intermediate data.

[0139] Intermediate data undergoes format conversion to transform heterogeneous data into a unified format, and the data is calibrated according to calibration rules to obtain standard data;

[0140] A feature extraction algorithm is used to extract key features from standard data to form a feature vector. The feature vector is represented as V = {v1, v2, ... vn}, where n represents the feature dimension. According to the rules, it is determined whether there are outliers in the feature vector. If there are outliers, an anomaly alarm message is generated.

[0141] Store normal data, associate the data with corresponding identifiers and timestamps and store them in a distributed database, generate node information of the data traceability graph based on the data storage records, and obtain a set of node information;

[0142] Obtain abnormal alarm information, extract abnormal data identifiers, and trace back abnormal data in the database based on the identifiers to obtain the abnormal data processing flow;

[0143] If the abnormal data identifier exists in the tire pressure data, then obtain the tire pressure data calculated by each computing node, and determine the range of tire pressure data based on the pre-established tire pressure data and internal structure relationship model.

[0144] If the tire pressure data exceeds the preset threshold range, the tire internal structure feature map at the corresponding moment of the calculation node is obtained, and the preset marked area in the feature map is identified by a pre-trained image recognition algorithm to determine the degree of deformation of the marked area.

[0145] The data traceability diagram is displayed through a graphical interface. According to the processing flow of abnormal data, abnormal nodes are highlighted in the traceability diagram. Based on the abnormal node information, the corresponding data processing information is obtained, and the problem node location result is obtained.

[0146] Specifically, in step S6, tire pressure data and internal structure feature maps are acquired, and a unique identifier is assigned to each data point. The identifier set is S = {S1, S2, ... Sn}, the number of internal structure feature maps is 100, the 50th feature map is represented as X50, and the set is represented as X = {X1, X2, ... X100}. The data acquisition time and data source information are recorded according to the identifier S1 and the timestamp 2023-10-26 10:00:00 to obtain the raw data containing specific data information.

[0147] The raw data is transmitted to the data preprocessing module, where data cleaning algorithms are used to remove duplicate and erroneous data, including removing duplicate values ​​and erroneous values ​​that are obviously outside the normal range in the pressure data, including negative pressure or pressure exceeding 10MPa. Based on the rule base, the data integrity is judged. If more than 20% of the data is missing, it is marked to obtain the marked intermediate data.

[0148] Intermediate data is converted from different formats to JSON format using a format conversion algorithm. According to calibration rules, including calibrating sensor data using standard pressure values ​​to obtain standard data, principal component analysis (PCA) algorithm is used to extract key features from the standard data to form a feature vector V = {v1, v2, ... v10}. This includes reducing the 100 feature dimensions of the original data to 10. According to the quartile range rule, it is determined whether there are outliers in the feature vector. If the value of v3 is greater than the upper quartile plus 1.5 times the interquartile range, an anomaly alarm message is generated.

[0149] Normal data is associated with the corresponding identifier S1 and timestamp 2023-10-26, 10:00:00 and stored in the database. Node information for the data traceability graph is generated based on the data storage records, resulting in a set of node information.

[0150] Obtain abnormal alarm information, extract abnormal data identifier S3, and trace back abnormal data in the database based on identifier S3 to obtain the processing flow of abnormal data from collection to processing.

[0151] If the abnormal data identifier S3 exists in the tire pressure data, then obtain the tire pressure data calculated by each calculation node, including the pressure data of node A which is 3.5MPa. Based on the pre-established tire pressure data and internal structure relationship model, determine whether the pressure data is between 2.0MPa and 3.0MPa.

[0152] If the tire pressure data range of 3.5MPa exceeds the preset threshold range of 2.0MPa to 3.0MPa, then the tire internal structure feature map X30 at the corresponding time of the calculation node A is obtained. The preset marked area in the feature map X30 is identified by the pre-trained ResNet-50 image recognition algorithm, and the degree of deformation of the marked area is determined, including a deformation coefficient of 0.8.

[0153] The data traceability diagram is displayed through a graphical interface. According to the processing flow of abnormal data S3, the abnormal nodes are highlighted in the traceability diagram. Based on the abnormal node information, the corresponding data processing link is obtained as the data preprocessing link. The problem node location result indicates that an abnormality occurred in the data preprocessing link.

[0154] like Figures 1-2 As shown, in step S7, the tire pressure data calculated by each computing node and the tire internal structure feature map at the corresponding time are obtained to obtain the original tire pressure data and internal structure image.

[0155] If the tire pressure data exceeds the preset threshold range, the preset marked area in the tire internal structure feature map at the corresponding moment of the calculation node is identified by a pre-trained image recognition algorithm to determine the degree of deformation of the marked area.

[0156] Based on the degree of deformation in the marked area, a regression model based on the relationship between the degree of deformation and the pressure data is used to obtain corrected pressure data and update the original pressure data;

[0157] By using a preset data format template, the corrected pressure data of each computing node is transformed into the first pressure data in a standardized format, thus obtaining standardized pressure data.

[0158] An image feature extraction algorithm based on convolutional neural networks is used to extract feature vectors from the internal structure feature maps uploaded by each computing node, and the feature vector set is determined.

[0159] Based on the preset image format standard for internal structure feature maps, the internal structure feature maps uploaded by each computing node are converted into a second internal structure feature map in a unified format to obtain a standardized internal structure feature map.

[0160] Acquire first pressure data and second internal structure feature map, and combine the first pressure data and second internal structure feature map into a standardized data package according to preset data packaging rules to obtain a combined data package;

[0161] According to the data storage protocol of the distributed database, the standard data packet is divided into multiple data blocks, and verification information is generated through the data verification algorithm to obtain the data block to be stored and the verification information.

[0162] The rendering engine maps the first pressure data into a pressure curve and renders the second internal structure feature map to obtain a visualization result.

[0163] Specifically, in step S7, the tire pressure data calculated by each computing node is obtained, including the pressure value uploaded by node A as 60 PSI, the pressure value uploaded by node B as 65 PSI, and the tire internal structure feature map at the corresponding time of these nodes.

[0164] If the tire pressure data of node A is 60 PSI, which exceeds the preset threshold range of 55 PSI to 63 PSI, then the preset marked area in the tire internal structure feature map of node A at the corresponding time is identified by a pre-trained image recognition algorithm, including the use of the ResNet-50 model, and the deformation degree of the marked area is determined to be 5%.

[0165] Based on the 5% deformation degree of the marked area, a regression model based on the relationship between the deformation degree and pressure data is adopted, including the linear regression model y = ax + b, where y represents the corrected pressure data and x represents the deformation degree. The corrected pressure data is calculated to be 62 PSI, and the original pressure data is updated to 60 PSI.

[0166] Using a preset data format template, the corrected pressure data of each computing node, including 62 PSI for node A and 64 PSI for node B, is converted into first pressure data in a standardized format. In this example, the JSON format {“node”:“A”,“pressure”:62} is used.

[0167] An image feature extraction algorithm based on convolutional neural networks, including the VGG-16 model, is adopted to extract feature vectors from the internal structure feature maps uploaded by each computing node and determine the feature vector set.

[0168] Based on the preset image format standard of the internal structure feature map, including uniformly converting it to PNG format and adjusting the resolution to 512x512 pixels, the internal structure feature maps uploaded by each computing node are converted into a second internal structure feature map in a unified format.

[0169] Acquire the first pressure data and the second internal structure feature map, and according to the preset data packaging rules, including sorting and combining the data according to timestamp and node ID, combine the first pressure data and the second internal structure feature map into a standardized data package;

[0170] According to the data storage protocol of the distributed database, including HDFS using Hadoop, the standard data packet is divided into multiple data blocks, each with a size of 128MB, and verification information is generated through data verification algorithms, including MD5.

[0171] The first pressure data is mapped to a pressure curve using a rendering engine, including OpenGL, with time on the horizontal axis and pressure value on the vertical axis. The second internal structure feature map is then rendered to obtain a visualization result.

[0172] like Figures 1-2 As shown, in step S8, based on the tire pressure data and internal structure feature diagram, the processor accesses the model library to obtain the corresponding tire model file and obtains a preliminary tire data model.

[0173] Obtain a preliminary tire data model, use a graphics library and combine it with internal structural feature map information to construct a three-dimensional tire model, and determine the basic shape of the three-dimensional tire model;

[0174] Obtain the constructed 3D tire model. If the pressure value exceeds the preset range, determine the tire deformation parameters based on the pressure value, modify the 3D tire model using the graphics library, and obtain the deformed 3D tire model.

[0175] Obtain the 3D model of the deformed tire, and use a renderer to perform rasterization processing to obtain the first rasterized image of the 3D tire model.

[0176] A first rasterized image of the tire 3D model is obtained. If the resolution of the first rasterized image is lower than a preset resolution threshold, a super-resolution reconstruction algorithm is used to process it to obtain a second rasterized image of the tire 3D model with a high resolution.

[0177] A high-resolution 3D tire model is obtained as a second rasterized image. If there are jagged edges in the image, an anti-aliasing algorithm is used to process it, resulting in a smooth-edge 3D tire model as a third rasterized image.

[0178] Obtain the third rasterized image of the 3D tire model with smooth edges. If there is noise in the image, use an image denoising algorithm to process it and obtain the fourth rasterized image of the clear 3D tire model.

[0179] Obtain a clear fourth rasterized image of the tire 3D model. If the color saturation of the clear fourth rasterized image of the tire 3D model is lower than the preset value, then perform color enhancement processing to obtain a color-rich fifth rasterized image of the tire 3D model.

[0180] The fifth rasterized image of the tire's 3D model with rich colors is obtained, and the visualization server outputs the final 3D image of the tire, thus obtaining the tire visualization result.

[0181] Specifically, in step S8, the processor accesses the model library based on tire pressure data (including 1.5 Bar) and internal structure feature map (including tread depth of 8 mm and sidewall steel wire layers of 2 layers), and matches the corresponding tire model file (including Michelin Pilot Sport4 tire model) through the retrieval algorithm to obtain a preliminary tire data model.

[0182] The tire data model was obtained, and using the OpenGL graphics library, combined with the internal structural feature map information (including the angle of each tread block being 30 degrees), a three-dimensional model with an 8mm tread depth and 2 layers of sidewall steel wires was constructed to determine the basic shape of the tire three-dimensional model.

[0183] If the pressure value is 2.5 Bar, which is outside the preset range of 1.8 Bar to 2.2 Bar, then the tire deformation parameters (including the radial deformation coefficient of 0.05) are determined based on the pressure value of 2.5 Bar. The tire 3D model is modified using the OpenGL graphics library to obtain a 3D tire 3D model that conforms to the deformation under 2.5 Bar pressure.

[0184] The three-dimensional model of the deformed tire was obtained, and rasterization was performed using a ray tracing renderer. The sampling rate was set to 16 samples per pixel to obtain the first rasterized image of the three-dimensional tire model with a resolution of 1024x768.

[0185] If the resolution of the first rasterized image of the tire 3D model is 1024x768, which is lower than the preset resolution threshold of 1920x1080, then a deep learning-based super-resolution reconstruction algorithm (including the SRCNN algorithm, which contains 3 convolutional layers with kernel sizes of 9x9, 1x1 and 5x5) is used to process it to obtain a high-resolution tire 3D model second rasterized image with a resolution of 1920x1080.

[0186] If jagged edges are detected in the image by the edge detection algorithm (including the Sobel operator), the Fast Approximate Antialiasing (FXAA) algorithm is used to process the image to obtain the third rasterized image of the 3D tire model with smooth edges.

[0187] If the peak signal-to-noise ratio (PSNR) of the image is found to be less than 30dB after obtaining the smooth edge image, then the image denoising algorithm based on non-local means is used for processing. The search window size is set to 21x21 and the similarity window size is set to 7x7 to obtain the fourth rasterized image of the clear tire 3D model.

[0188] If the color saturation of the clear image is lower than the preset value of 0.8, then color enhancement processing is performed to increase the saturation to 0.9, resulting in a fifth rasterized image of the tire 3D model with rich colors.

[0189] The image with rich colors is obtained, and the visualization server outputs the final 3D image of the tire, thus obtaining the tire visualization result.

[0190] like Figures 1-2 As shown, in step S9, tire pressure data and internal structure feature map are acquired, and the data format of tire pressure data and internal structure feature map is determined;

[0191] Based on the data format of tire pressure data and internal structure feature map, a mapping relationship for the tire 3D model is constructed, and the mapping relationship parameters of the tire 3D model are obtained.

[0192] Using the mapping parameters of the tire 3D model, the tire pressure data and internal structure feature map are converted to obtain the initial model data of the tire 3D model.

[0193] Using the initial model data of the tire 3D model, the geometry of the tire 3D model is constructed to obtain the geometric model of the tire 3D model;

[0194] If the geometric model of the tire 3D model conforms to the mapping relationship between tire pressure data and internal structural feature map, then attribute mapping is performed on the geometric model of the tire 3D model to obtain the attribute model of the tire 3D model.

[0195] Based on the attribute model of the tire 3D model, a lighting model of the tire 3D model is generated to obtain the lighting effect of the tire 3D model;

[0196] If the lighting effect of the tire 3D model matches the visual characteristics of the tire pressure data and internal structural feature map, then the tire 3D model is rasterized to obtain a 2D image of the tire 3D model.

[0197] Based on the two-dimensional image of the tire's three-dimensional model, image rendering is performed to obtain the visualization result of the tire's three-dimensional model;

[0198] By using a distributed database, the storage structure of tire pressure data, internal structural feature maps, and visualization results of tire 3D models are stored, and the storage structure of tire visualization results is determined.

[0199] Specifically, in step S9, tire pressure data is acquired, including a pressure value of 3.5 bar recorded by each tire pressure sensor, and an internal structure feature map is acquired, including an image of the tire's internal structure obtained by X-ray scanning with a resolution of 1024x768 pixels. The pressure data is determined to be in floating-point format and the internal structure feature map is in grayscale image format.

[0200] Based on these data formats, a mapping relationship is constructed, including mapping pressure values ​​to color changes of the tire model (3.5 bar corresponds to RGB color value (255,0,0)) and mapping grayscale values ​​of internal structural feature maps to texture details of the model surface (grayscale value range 0-255 corresponds to texture depth 0.1-1.0 mm). A mapping relationship parameter file is obtained. Using this mapping relationship parameter file, the data is transformed, including using a Python script to read pressure data and image data, and using a linear interpolation algorithm to convert pressure values ​​and grayscale values ​​into vertex colors and texture coordinates of the tire model to obtain the initial model data file.

[0201] By reading the initial model data file, the geometry is constructed using graphics libraries such as OpenGL, including constructing a triangular facet model using vertex coordinates and facet indices, resulting in a geometric model containing 10,000 triangular faces.

[0202] If the vertex color and texture coordinates of each triangle facet in the geometric model match the definition in the mapping parameter file, then attribute mapping is performed, including setting the color attribute of each triangle facet to the corresponding RGB value and mapping the texture coordinates to the corresponding texture image, to obtain a model with color and texture attributes.

[0203] Based on the attribute model, a lighting model is generated, including using the Phong lighting model and setting the coefficients of ambient light, diffuse light, and specular light to 0.2, 0.5, and 0.3, respectively, to obtain a model with lighting effects;

[0204] If, under lighting conditions, the high-pressure area of ​​the tire model appears bright red and the internal structure and texture are clearly visible, conforming to visual characteristics, then rasterization processing is performed, including using the Z-buffer algorithm for depth testing, projecting the 3D model onto a 2D plane, and obtaining a 1920x1080 pixel 2D image.

[0205] Based on the two-dimensional image, the image is rendered using a rendering engine including Blender. The rendering parameters are set to 100 sampling times and PNG output format to obtain the final visualization result image.

[0206] Using a distributed database including MongoDB, pressure data, internal structure feature maps, and visualization result images are stored. The storage structure is determined to include a pressure value field, an image field, and a result image field for each tire record.

[0207] like Figure 3 As shown, the tire full life cycle supply chain management system described in this application includes: an intelligent sensor module, an edge computing node module, a distributed computing cluster module, a data standardization module, a data traceability construction module, a distributed database storage module, a tire 3D model generation module, and a tire 3D model optimization module.

[0208] The intelligent sensor module is connected to the edge computing node module to collect tire pressure data and internal structure images in real time, and uploads them to the data processing center through the communication module in the intelligent sensor.

[0209] The edge computing node module is connected to the smart sensor module and the distributed computing cluster module. It is used to receive tire pressure data from the smart sensor, process filtering, noise reduction, and feature extraction, and perform pooling and classification through a convolutional neural network to determine the internal structural features of the tire. The processed tire pressure data and internal structural feature map are stored in a distributed database and uploaded to the distributed computing cluster. The distributed computing framework distributes the pressure data and internal structural data to different computing nodes.

[0210] The distributed computing cluster module is connected to the edge computing node module and the data standardization module. It is used to receive data uploaded by the edge nodes, distribute it to different computing nodes for parallel computing, and perform exception handling. If a computing node fails, a dynamic branch prediction mechanism is activated to reallocate the data.

[0211] The data standardization module is connected to the distributed computing cluster module and the data traceability construction module, and is used to convert data into data with a unified data standard.

[0212] The data traceability construction module is connected to the data standardization module and the distributed database storage module. It is used to construct a tire data traceability map based on the standardized data. When data processing is abnormal, the problem node is located through the traceability map. The original data is transmitted to the data preprocessing module, which judges the data integrity according to the rule base; normal data is stored, and the data is associated with the corresponding identifier and timestamp and stored in the distributed database.

[0213] The distributed database storage module is connected to the data traceability construction module and the tire 3D model generation module. It is used to store the processed data in the distributed database, visualize the pressure data through the rendering engine, and render the structural feature map. In this process, the standard data packet is divided into multiple data blocks according to the data storage protocol of the distributed database.

[0214] The tire 3D model generation module is connected to the distributed database storage module and the tire 3D model optimization module. It is used to construct a tire 3D model using a graphics library and internal structural feature map information. The tire 3D image after rasterization is output by the visualization server. The processor accesses the model library to obtain the tire model file and uses the graphics library to modify the tire 3D model to obtain a deformable tire 3D model.

[0215] The tire 3D model optimization module is connected to the tire 3D model generation module. It is used to construct the mapping relationship and initial model data of the tire 3D model, create a lighting model, realize the lighting effect, and perform rasterization processing to generate a 2D image. Pressure data, internal structural feature maps and visualization result images are stored in a distributed database.

[0216] In the description of this specification, references to terms such as "some embodiments," "other embodiments," and "ideal embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative descriptions of the above terms do not necessarily refer to the same embodiments or examples.

[0217] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0218] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A tire lifecycle supply chain management method, characterized in that, Includes the following steps: S1. Deploying a multi-core heterogeneous parallel processing intelligent sensor, the intelligent sensor collects raw tire pressure data in real time and obtains tire internal structure data, specifically including: Deploy intelligent sensor nodes to monitor tire pressure, upload the data to the processing center, and trigger an alarm when the tire pressure is too low. Simultaneously, images of the tire's interior are acquired and segmented, and texture features are analyzed using a multi-core heterogeneous architecture. S2. The edge computing node filters the raw tire pressure data in step S1 to remove redundant pressure information and preprocesses the tire internal structure data in step S1 to generate a tire internal structure feature map. S3. The raw data of tire pressure and the tire internal structure feature map from step S2 are uploaded to the distributed computing cluster. The distributed computing framework distributes the raw data of tire pressure and the tire internal structure feature map to different computing nodes for parallel computing. S4. If the multi-threaded hardware of the compute node fails, the dynamic branch prediction mechanism is activated, which distributes the data that the node is responsible for processing to other normal nodes, and uses the cache consistency protocol to ensure data consistency. S5. After each calculation node completes the calculation of the raw data of tire pressure and the tire internal structure feature map, the format of the raw data of tire pressure and the tire internal structure feature map is unified through a unified data standard. S6. Based on the unified raw tire pressure data and tire internal structure feature diagram, construct a tire data traceability diagram to record the entire process from data collection to processing. If an anomaly occurs in the data processing stage, locate the problem node based on the traceability diagram. Specifically, this also includes: Collect tire pressure and internal image data, assign a unique ID and record the time, and transmit them to the preprocessing module for cleaning; S7. Store the processed raw tire pressure data and tire internal structure feature map into a distributed database, and render the raw tire pressure data and tire internal structure feature map to obtain visualization results. This also includes: The feature maps are standardized in format, combined with stress data, packaged, stored in blocks for verification, and then visualized. S8. The visualization server generates a three-dimensional tire model based on the original tire pressure data and the tire's internal structural feature map. The three-dimensional model is then rasterized to obtain a three-dimensional tire image. S9. Map the raw tire pressure data and the tire internal structure feature map to the tire 3D model, render the mapped model, and obtain the tire visualization results.

2. The tire lifecycle supply chain management method according to claim 1, characterized in that, In step S3, the raw tire pressure data and tire internal structure feature map from step S2 are uploaded to the distributed computing cluster. The distributed computing framework distributes the raw tire pressure data and tire internal structure feature map to different computing nodes for parallel computation, including: Obtain the timestamp associated with the tire ID to form an initial data pair. Based on the tire ID, combine the raw tire pressure data and the tire internal structure feature map into a data packet and upload it to the distributed computing cluster. The distributed computing cluster receives the data packet and, according to a pre-established allocation table, allocates the data packet to an idle node in the computing cluster. If a node is processing data, the data packet is cached in the node's queue. After receiving the data packet, the nodes in the computation group use the Fast Fourier Transform algorithm to extract the periodic variation features in the pressure value and obtain the pressure fluctuation frequency.

3. The tire lifecycle supply chain management method according to claim 1, characterized in that, In step S4, if the computing node hardware multithreading malfunctions, a dynamic branch prediction mechanism is activated to distribute the data processed by that node to other normal nodes. This is used by the cache consistency protocol to ensure data consistency, including: Obtain runtime information of the compute node, including the thread state recorded in the processor's registers. If the thread state is abnormal, trigger the interrupt handler. The interrupt handler reads a pre-established list of node states and determines the set of normal nodes to participate in the calculation based on the node state information in the list. The execution unit counts the amount of data to be processed for each thread, and obtains the average amount of data based on the amount of data to be processed and the number of normal nodes participating in the calculation. Based on the average data volume and the current load of each normal node, a data allocation scheme is determined, which includes the mapping relationship between data blocks and target nodes.

4. The tire lifecycle supply chain management method according to claim 1, characterized in that, In step S5, after each calculation node completes the calculation of the raw tire pressure data and the tire internal structure feature map, the format of the raw tire pressure data and the tire internal structure feature map is unified through a unified data standard, including: Obtain the tire pressure data calculated by each computing node, and determine the range of tire pressure data based on the pre-established model of the relationship between tire pressure data and internal structure. If the tire pressure data range exceeds the preset threshold range, the tire internal structure feature map at the corresponding time of the calculation node is obtained, and the preset marked area in the tire internal structure feature map is identified by a pre-trained image recognition algorithm to determine the degree of deformation of the marked area. Based on the degree of deformation of the marked area, a regression model based on the relationship between the degree of deformation and the pressure data is used to obtain corrected pressure data.

5. The tire lifecycle supply chain management method according to claim 1, characterized in that, In step S8, the visualization server generates a 3D model of the tire based on the raw tire pressure data and the tire's internal structural feature map. The 3D model is then rasterized to obtain a 3D image of the tire, including: Based on tire pressure data and internal structure feature diagrams, the processor accesses the model library to obtain tire model files and obtains a preliminary tire data model. Obtain a preliminary tire data model, use a graphics library and combine it with internal structural feature map information to construct a three-dimensional tire model, which is used to determine the basic shape of the three-dimensional tire model; Obtain the constructed 3D tire model. If the pressure value exceeds the preset range, determine the tire deformation parameters based on the pressure value, modify the 3D tire model using the graphics library, and obtain the deformed 3D tire model.

6. A tire lifecycle supply chain management system, characterized in that, include: The intelligent sensor module is connected to the edge computing node module to collect tire pressure data and internal structure images in real time; The edge computing node module is connected to the smart sensor module and the distributed computing cluster module. It is used to receive tire pressure data from the smart sensor, process filtering, noise reduction, and feature extraction, and perform pooling and classification through convolutional neural networks to determine the internal structural features of the tire. The distributed computing cluster module is connected to the edge computing node module and the data standardization module. It is used to receive data uploaded by edge nodes, distribute it to different computing nodes for parallel computing, and perform exception handling. If a computing node fails, a dynamic branch prediction mechanism is activated to reallocate the data. The data standardization module is connected to the distributed computing cluster module and the data traceability construction module, and is used to convert data into data with a unified data standard. The data traceability construction module is connected to the data standardization module and the distributed database storage module. It is used to build a tire data traceability map based on the standardized data. When data processing is abnormal, the traceability map is used to locate the problem node. The distributed database storage module is connected to the data traceability construction module and the tire 3D model generation module. It is used to store the processed data in the distributed database, visualize the pressure data through the rendering engine, and render the structural feature map. The tire 3D model generation module is connected to the distributed database storage module and the tire 3D model optimization module. It is used to construct the tire 3D model using graphics library and internal structural feature map information. The tire 3D image after rasterization is output by the visualization server. The tire 3D model optimization module is connected to the tire 3D model generation module. It is used to build the mapping relationship and initial model data of the tire 3D model, create the lighting model, and perform rasterization processing to generate a 2D image.

Images