Digital-twin-based automobile rain test data fusion processing method and system
By constructing a digital twin-based vehicle rain test data fusion system, the problems of data fusion deviation and missing correlation were solved, achieving accurate fusion and anomaly localization of test data, and improving the refined processing capability of rain tests.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHONGQING BOKE ELECTROMECHANICAL EQUIP CO LTD
- Filing Date
- 2026-04-15
- Publication Date
- 2026-06-05
AI Technical Summary
Existing automotive rain test data processing suffers from issues such as core data fusion deviation and missing correlations, making it impossible to accurately reflect the waterproof performance of automobiles and difficult to pinpoint specific defects and their root causes.
Digital twin technology is used to construct a digital twin of the test vehicle with full-dimensional data association, realizing real-time mapping and state synchronization between physical entities and virtual models, collecting associated data throughout the entire life cycle, performing data preprocessing and association alignment, and generating associated test reports through hierarchical association fusion and anomaly verification.
It achieves precise data fusion, enhances the utilization value of test data, provides reliable support for the refined and personalized processing of automobile rain tests, and can accurately locate abnormalities in automobile waterproofing performance.
Smart Images
Figure CN122153802A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of automotive test data processing technology, and in particular to a method and system for fusion processing of automotive rain test data based on digital twins. Background Technology
[0002] The rain test is a core testing step in verifying the waterproof performance of automobiles. Its core purpose is to simulate the usage scenario of a car in natural rainfall, test the waterproof performance of key components such as the car body, doors, windows, and seals, and promptly detect defects such as leaks and seepage, providing a basis for automotive product design optimization and quality control. With the intelligent and digital development of the automotive industry, the complexity of rain tests is constantly increasing. The data generated during the test exhibits characteristics of multi-source and heterogeneity. Among these, the multi-dimensional data of the test vehicle itself (structure, state, defects, etc.) is the core data reflecting the vehicle's waterproof performance. Currently, data processing for rain tests mainly adopts a fusion method centered on test equipment and environmental data, neglecting the correlation between the multi-dimensional data of the test vehicle itself. Existing technologies have the following prominent shortcomings: First, the core of data fusion is deviated; the correlation of test vehicle data is not taken as the core logic, resulting in insufficient strong correlation between the fused data and the vehicle's waterproof performance, failing to accurately reflect the impact of the vehicle's own characteristics on waterproof performance. Second, data correlation is lacking; the structural, state, and defect data of the test vehicle are independent of the test equipment and environmental data, and no effective correlation rules have been established, making it difficult to locate specific defects and their root causes through data fusion.
[0003] Digital twin technology, as an emerging technology that integrates the Internet of Things, big data, and simulation modeling, can build a one-to-one mapping between physical entities and virtual models, realize real-time interaction and linkage between virtual and real data, and provide technical support for the correlation and fusion of test vehicle data.
[0004] Therefore, it is essential to propose a data fusion processing method and system for automobile rain tests that avoids core deviations and missing correlations in data fusion, thereby enhancing the utilization value of test data and providing reliable support for the refined and personalized processing of automobile rain tests. Summary of the Invention
[0005] The purpose of this invention is to provide a method and system for fusion processing of automotive rain test data based on digital twins, aiming to avoid core deviations and missing correlations in data fusion, thereby enhancing the utilization value of test data and providing reliable support for the refined and personalized processing of automotive rain tests.
[0006] To achieve the above objectives, the present invention employs a method for fusing and processing vehicle rain test data based on digital twins, comprising the following steps: Construct a digital twin of the vehicle rain test based on the full-dimensional data association of the test vehicle, establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment, and perform mapping and state synchronization between the physical entity of the test vehicle and the virtual model; Collect relevant data on the entire lifecycle of the test vehicle, test auxiliary data, and virtual simulation data; preprocess the collected data; perform data association and alignment; and output the associated data. Perform hierarchical correlation and fusion on the correlated data, and output fused data related to the waterproof performance of automobiles; Based on the correlation of vehicle data, the system verifies and merges data, locates anomalies, and generates a correlation-based test report.
[0007] Among the steps involved are: constructing a digital twin of the vehicle rain test based on the full-dimensional data association of the test vehicle; establishing real-time communication between the digital twin and the physical rain test equipment; and mapping and synchronizing the physical entity and virtual model of the test vehicle. Using the virtual model of the test vehicle as the core, and linking the test scenario twin module and the rain equipment twin module, a digital twin of the vehicle rain test is constructed; Establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment; The inherent structure, real-time status, and defects of the physical test vehicle are mapped to the virtual model in a comprehensive manner. The state between the physical entity of the test vehicle and the virtual model of the test vehicle.
[0008] Among the steps involving the state between the physical entity of the test vehicle and the virtual model of the test vehicle: Using the multi-source correlated data of the test vehicle itself as the core, the parameters of the vehicle's structural data, status data, defect data and the virtual model of the test vehicle are synchronized, and the parameters of the test scenario and the rain equipment are linked and synchronized with the corresponding status of the virtual model of the test vehicle.
[0009] Among the steps are: collecting full lifecycle related data of the test vehicle, test auxiliary data, and virtual simulation data; preprocessing the collected data; performing data association and alignment; and outputting the associated data. Collect multi-source correlation data of the entire life cycle of the test vehicle; the multi-source correlation data includes vehicle inherent structure data, test process status data, and waterproof defect correlation data. Collect test auxiliary data and virtual simulation data of the digital twin of the vehicle rain test; the test auxiliary data includes rain equipment parameter data and test environment parameter data, and the virtual simulation data includes virtual vehicle correlation data and virtual auxiliary test data; The collected data are preprocessed as follows: missing values, outliers and duplicate data are removed from the data, wavelet denoising algorithm is used to remove data noise, all data are standardized and output as standardized data. Establish data association rules for test vehicles, divide the entire chain of association logic, and align standardized data with test vehicles by aligning timestamps and association identifiers, and output associated data.
[0010] In the steps of establishing data association rules for test vehicles, dividing the entire chain of association logic, aligning standardized data with test vehicles through timestamp alignment and association identifier alignment, and outputting associated data, the process of aligning standardized data with test vehicles through timestamp alignment is as follows: Add a timestamp in a uniform format to each piece of standardized data; the timestamp corresponds to the moment the data was collected and generated. Based on the timeline of the test vehicle testing process, the standardized data are sorted in chronological order to synchronize and align different types of data in the time dimension.
[0011] In the steps of establishing data association rules for test vehicles, dividing the entire chain of association logic, aligning standardized data with test vehicles through timestamp alignment and association identifier alignment, and outputting associated data, the process of aligning standardized data with test vehicles through association identifier alignment is as follows: Based on the established data association rules for test vehicles, a unique global association identifier is assigned to each test vehicle. At the same time, sub-identifiers associated with the global identifier are assigned to different structural parts, different test stages, and various types of data of the test vehicle. Standardized data is bound to corresponding sub-identifiers to divide each data point into the test vehicle, specific parts of the vehicle, and test stage. By using a global association identifier, standardized data of all bound sub-identifiers are associated and matched with the test vehicle, achieving full-dimensional alignment between the data and the test vehicle.
[0012] In the step of performing hierarchical correlation and fusion of correlated data to output fused data related to automotive waterproofing performance: Data layer association and fusion are carried out. The automotive-related data, test auxiliary data, and virtual simulation data in the associated data are initially merged. The automotive-related data is given a higher basic weight, and the weight is dynamically adjusted according to the correlation between the data and the waterproof performance of the vehicle. Feature layer correlation fusion is performed to extract feature parameters that are strongly correlated with the waterproof performance of automobiles after data layer fusion. Principal component analysis algorithm is used to reduce the dimensionality of the feature parameters, remove redundant features, and retain key correlated features. The decision-level association and fusion is carried out. Based on the association rules of automobile data, the key feature parameters extracted from the feature layer are used for decision fusion, the core information of various data are integrated, and standardized fused data is output.
[0013] Among them, in the steps of verifying and fusing data based on the correlation of vehicle data and locating anomalies to generate a correlation test report: Based on the correlation of vehicle data, a deviation analysis algorithm is used to verify the validity of the fused data. Corresponding verification thresholds are set for different structural parts and different waterproof performance indicators of the test vehicle. By comparing the deviation between the fused data and the preset standard data, abnormal situations can be identified. Effective fusion data is selected, and basic information of the test vehicle, multi-dimensional related data, fusion results and anomaly analysis information are integrated to obtain test conclusions and generate a correlation test report.
[0014] In the step of comparing the deviation between the fused data and the preset standard data to determine anomalies: When the deviation exceeds the set threshold, it is judged as abnormal data. Then, the vehicle-related parts, causes of abnormality, and associated test parameters are located through vehicle data association rules.
[0015] This invention also provides a digital twin-based vehicle rain test data fusion processing system, including an entity and virtual mapping module, an associated data acquisition module, a hierarchical data fusion module, and an abnormal state identification module; wherein: The entity and virtual mapping module is used to construct a digital twin of the vehicle rain test based on the full-dimensional data association of the test vehicle, establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment, and perform mapping and state synchronization between the physical entity and the virtual model of the test vehicle. The associated data acquisition module is used to collect associated data of the entire life cycle of the test vehicle, test auxiliary data and virtual simulation data, preprocess the collected data, perform data association and alignment, and output associated data. The hierarchical data fusion module is used to perform hierarchical correlation and fusion of correlated data and output fused data related to the waterproof performance of automobiles. The abnormal state identification module is used to verify and merge data based on the correlation of vehicle data and locate abnormalities, and generate a correlation test report.
[0016] This invention discloses a digital twin-based method and system for fusion processing of automotive rain test data. The system comprises an entity and virtual mapping module, a correlated data acquisition module, a hierarchical data fusion module, and an anomaly identification module, comprising the following steps: constructing a digital twin of the automotive rain test based on the full-dimensional data correlation of the test vehicle; establishing real-time communication between the digital twin and the physical rain test equipment; mapping and synchronizing the physical entity and virtual model of the test vehicle; collecting correlation data throughout the entire lifecycle of the test vehicle, test auxiliary data, and virtual simulation data; preprocessing the collected data; performing data correlation alignment; outputting correlated data; performing hierarchical correlation fusion of the correlated data; outputting fused data correlated with the vehicle's waterproof performance; verifying the fused data based on the vehicle data correlation relationships and locating anomalies; and generating a correlated test report. By combining digital twin technology with the test vehicle data correlation as the core, a multi-source data correlation fusion system is constructed to avoid core deviations and missing correlations in data fusion, thereby enhancing the utilization value of test data and providing reliable support for the refined and personalized processing of automotive rain tests. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a flowchart of the steps of the vehicle rain test data fusion processing method based on digital twin of the present invention.
[0019] Figure 2 This is a flowchart of steps S100 of the present invention.
[0020] Figure 3 This is a flowchart of steps S200 of the present invention.
[0021] Figure 4 This is a flowchart of steps S300 of the present invention.
[0022] Figure 5 This is a flowchart of steps S400 of the present invention.
[0023] Figure 6 This is a schematic diagram of the structural principle of the vehicle rain test data fusion processing system based on digital twins of the present invention.
[0024] Figure 7 This is a schematic diagram of the electronic device of the present invention.
[0025] 501 - Entity and Virtual Mapping Module, 502 - Related Data Acquisition Module, 503 - Hierarchical Data Fusion Module, 504 - Abnormal State Identification Module. Detailed Implementation
[0026] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.
[0027] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.
[0028] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."
[0029] Please see Figures 1-5 This invention provides a method for fusing and processing vehicle rain test data based on digital twins, comprising the following steps: S100: Construct a digital twin of the vehicle rain test based on the full-dimensional data association of the test vehicle, establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment, and perform mapping and state synchronization between the physical entity of the test vehicle and the virtual model.
[0030] In this embodiment, a digital twin of the vehicle rain test is constructed based on the full-dimensional data association of the test vehicle. Real-time communication is established between the digital twin and the physical rain test equipment, and the mapping and state synchronization between the physical entity of the test vehicle and the virtual model are performed. The specific process is as follows: S101: Using the virtual model of the test vehicle as the core, and linking the test scenario twin module and the rain equipment twin module, a digital twin of the vehicle rain test is constructed; S102: Establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment; S103: Map the inherent structure, real-time status, and defect status of the physical entity of the test vehicle to the virtual model. S104: Synchronize the state between the physical entity of the test vehicle and the virtual model of the test vehicle; taking the multi-source correlation data of the test vehicle itself as the core, synchronize the vehicle structure data, state data, defect data and the parameters of the virtual model of the test vehicle, and link the parameters of the test scenario and the rain equipment with the corresponding state of the virtual model of the test vehicle.
[0031] In the aforementioned process, basic data such as the 3D design drawings, inherent structural parameters, and waterproofing process parameters of the test vehicle were acquired. Using Unity3D and Python as joint development tools, a virtual model of the test vehicle was constructed. The model's precision reached the part level, completely replicating the test vehicle's body structure, waterproof seals (such as door seals and A-pillar seals), and key connection points (such as rear door seams and speaker mounting locations). Simultaneously, a built-in vehicle data association and mapping module was incorporated for subsequent synchronization of physical and virtual data. Secondly, a test scenario twin module was constructed, recreating the virtual test scenario 1:1 according to the actual dimensions of the physical rain test site, the division of rain zones, the layout of the drainage system, and site environmental parameters. This ensured that the geometric parameters and environmental conditions of the virtual scenario were consistent with the physical site. Furthermore, this module was linked to the virtual model of the test vehicle, enabling real-time responses to changes in the test vehicle's position and attitude. Finally, a rain equipment twin module is constructed, collecting core parameters of the physical rain equipment (such as the number of spray heads, spray angle adjustment range, spray pressure range, flow rate adjustment capability, etc.) to build a virtual rain equipment model consistent with the physical equipment, simulating the operating state of the physical equipment. This module is also linked with the virtual model of the test vehicle, and can adjust the spray parameters according to the position and attitude of the test vehicle to achieve precise matching between the spray process and the test vehicle. The three modules work together to form a complete digital twin of the vehicle rain test, with the data association of the test vehicle as the core throughout the process, ensuring that the data of each module revolves around the test vehicle.
[0032] A hybrid communication approach combining 5G and industrial Ethernet is employed to establish a real-time communication link between the digital twin of the automotive rain test and the physical rain test equipment. 5G communication is used to transmit real-time data such as vehicle status data and defect detection data, while industrial Ethernet is used to transmit large-capacity data such as test scenario parameters and historical operating data of the rain test equipment, ensuring the stability and real-time performance of the communication link. At the physical rain test equipment end, data acquisition and communication modules are installed to collect and transmit the equipment's operating parameters (spray pressure, flow rate, angle, etc.) and operating status to the automotive rain test digital twin in real time. At the automotive rain test digital twin end, communication receiving and data parsing modules are built to perform real-time parsing and format conversion of the received physical equipment data, ensuring that the data can be recognized and utilized by the automotive rain test digital twin. Meanwhile, a communication delay control mechanism is set up to strictly control the communication delay within 100ms, so as to avoid the asynchrony between virtual and real data due to communication delay, ensure the accuracy of the synchronization between the physical entity of the vehicle and the virtual model in subsequent tests, realize two-way data interaction between the digital twin of the vehicle rain test and the physical rain test equipment, and ensure that the virtual model can replicate the operating status of the physical equipment in real time, and the physical equipment can also respond to the parameter adjustment instructions of the virtual model.
[0033] The test vehicle's physical data was comprehensively analyzed, including inherent structural data (vehicle dimensions, seal types and installation parameters, structural parameters of body connections, waterproofing process parameters, etc.), real-time status data (vehicle posture, surface temperature, seal deformation, stress on key body parts, etc. during the test), and defect data (leakage location, leakage rate, defect level, etc.). This data was categorized and organized into a data classification catalog, clearly defining the attributes, collection methods, and meanings of each data type. Secondly, relying on the data association mapping module built into the test vehicle's virtual model, a one-to-one correspondence was established between physical data and virtual model parameters. The inherent structural data of the physical vehicle was mapped to the corresponding components of the virtual vehicle model, ensuring complete consistency between the virtual model's structure, dimensions, and waterproofing process and the physical entity. Real-time status data of the physical entity was mapped to the corresponding status parameters of the virtual model, enabling the virtual model to replicate the physical entity's posture, temperature, deformation, and other states in real time. Defect data of the physical entity was mapped to the corresponding parts of the virtual model, simulating the location and extent of physical defects, achieving virtual-real synchronization of defects. During the mapping process, a mapping verification mechanism is established to check the consistency between physical data and virtual model mapping data in real time. If mapping deviation occurs, it is corrected in a timely manner to ensure the accuracy and reliability of the mapping and lay the foundation for subsequent state synchronization.
[0034] Using the multi-source correlated data of the test vehicle itself as the core, a state synchronization mechanism is established to achieve real-time synchronization between the physical entity and the virtual model. On one hand, the vehicle structure data, state data, and defect data of the physical entity of the test vehicle are collected in real time and transmitted to the digital twin of the vehicle rain test through a real-time communication link. The data parsing module of the digital twin of the vehicle rain test processes the data and feeds it back to the virtual model of the test vehicle, adjusting the corresponding parameters of the virtual model to ensure that the structure, state, and defect status of the virtual model are completely synchronized with the physical entity. For example, when the seal of the physical entity deforms, the corresponding seal of the virtual model will also deform to the same extent synchronously; when the physical entity detects a water leakage defect, the corresponding part of the virtual model will also display the defect status synchronously. On the other hand, the system achieves synchronized linkage between the test scenario, rain equipment parameters, and the virtual model of the test vehicle. When the environmental parameters (temperature, humidity) of the physical test scenario change, the corresponding parameters of the virtual scenario adjust synchronously to ensure consistency between the virtual and physical scenarios. When the spray parameters (pressure, angle, flow rate) of the physical rain equipment are adjusted, the corresponding parameters of the virtual rain equipment adjust synchronously, and the virtual test vehicle model also responds synchronously to the changes in spray parameters, simulating changes in the vehicle's state under different spray conditions. During the synchronization process, the synchronization accuracy is monitored in real time. If a synchronization deviation occurs, a correction mechanism is triggered promptly to adjust the virtual model parameters or physical equipment parameters, ensuring the real-time performance and accuracy of state synchronization. This achieves full-link linkage and synchronization of "physical entity - virtual model - test scenario - rain equipment," revolving entirely around the data association of the test vehicle and ensuring the consistency of various data linkages.
[0035] S200: Collects relevant data throughout the entire lifecycle of the test vehicle, test auxiliary data, and virtual simulation data; preprocesses the collected data; performs data association and alignment; and outputs associated data.
[0036] In this embodiment, the entire lifecycle data of the test vehicle, test auxiliary data, and virtual simulation data are collected. The collected data is preprocessed, and data association and alignment are performed to output the associated data. The specific process is as follows: S201: Collect multi-source correlation data of the entire life cycle of the test vehicle; the multi-source correlation data includes vehicle inherent structure data, test process status data, and waterproof defect correlation data; S202: Collect test auxiliary data and virtual simulation data of the digital twin of the vehicle rain test; among which, the test auxiliary data includes rain equipment parameter data and test environment parameter data, and the virtual simulation data includes virtual vehicle correlation data and virtual auxiliary test data; S203: Preprocess the collected data of various types. The preprocessing process is as follows: remove missing values, outliers and duplicate data from various types of data, remove data noise using wavelet denoising algorithm, standardize all data and output standardized data. S204: Establish data association rules for test vehicles, divide the entire chain of association logic, and align standardized data with test vehicles by aligning timestamps and association identifiers, and output associated data.
[0037] Furthermore, in the steps of establishing data association rules for test vehicles, dividing the entire chain of association logic, aligning standardized data with test vehicles through timestamp alignment and association identifier alignment, and outputting associated data, the process of aligning standardized data with test vehicles through timestamp alignment is as follows: Add a timestamp in a uniform format to each piece of standardized data; the timestamp corresponds to the moment the data was collected and generated. Based on the timeline of the test vehicle testing process, the standardized data are sorted in chronological order to synchronize and align different types of data in the time dimension.
[0038] Furthermore, in the steps of establishing data association rules for test vehicles, dividing the entire chain of association logic, aligning standardized data with test vehicles through timestamp alignment and association identifier alignment, and outputting associated data, the process of aligning standardized data with test vehicles through association identifier alignment is as follows: Based on the established data association rules for test vehicles, a unique global association identifier is assigned to each test vehicle. At the same time, sub-identifiers associated with the global identifier are assigned to different structural parts, different test stages, and various types of data of the test vehicle. Standardized data is bound to corresponding sub-identifiers to divide each data point into the test vehicle, specific parts of the vehicle, and test stage. By using a global association identifier, standardized data of all bound sub-identifiers are associated and matched with the test vehicle, achieving full-dimensional alignment between the data and the test vehicle.
[0039] In the above process, a "full lifecycle acquisition + multi-device collaborative acquisition" approach is adopted to collect multi-source correlated data throughout the entire lifecycle of the test vehicle. For inherent structural data of the vehicle, before the test begins, static data such as body dimensions, seal type and installation parameters, structural parameters of body connection parts, and waterproofing process parameters (e.g., sealant thickness and application location) are collected by reading the test vehicle's design drawings and manufacturing process documents. This data serves as the basic benchmark for subsequent data correlation, is collected and stored in one go, and a data archive is established for easy retrieval and comparison later. For test process status data, during the test, dynamic data such as the vehicle's attitude, surface temperature, seal deformation, and stress on key body parts are collected in real time using various sensors (attitude sensors, temperature sensors, deformation sensors, pressure sensors, etc.) installed on the test vehicle. The acquisition frequency is set to 1 time / second to ensure complete capture of changes in the vehicle's state during the test. The collected data is transmitted to the data processing module in real time to avoid data loss. For waterproofing defect correlation data, a multi-device collaborative data acquisition approach was adopted during and after the test. Endoscopic inspection equipment was used to collect data on water leakage inside the vehicle body, infrared cameras were used to capture water leakage traces on the vehicle body surface, and visual inspection equipment was used to identify the location of external water leakage. Simultaneously, the leakage rate and defect level were recorded. A one-to-one correspondence was established between the defect data and the vehicle's inherent structural data, clearly defining the corresponding structural part and seal model for each defect, ensuring the correlation and traceability of the defect data. All collected data was initially recorded according to a unified format, labeling basic information such as data type, collection time, and collection location to facilitate subsequent preprocessing.
[0040] On the one hand, auxiliary test data is collected. This auxiliary data supplements the data related to the test vehicle and is used to analyze the impact of external factors on the vehicle's waterproof performance. Specifically, rain equipment parameter data is collected in real time using pressure sensors, flow sensors, and angle sensors installed on the physical rain equipment. Parameters such as spray pressure, spray flow rate, spray time, and spray angle are collected at a frequency consistent with the test vehicle's status data (1 time / second) to ensure data synchronization. On the other hand, environmental parameter data is collected in real time using environmental sensors installed at the test site. Parameters such as temperature, humidity, and atmospheric pressure are collected every 5 minutes to ensure that the environmental data reflects environmental changes throughout the entire test process. On the other hand, virtual simulation data is collected from the digital twin of the vehicle rain test. Simultaneously with the physical rain test, the digital twin runs a virtual simulation, collecting virtual simulation data. This data includes the virtual vehicle's structural parameters, state parameters (attitude, temperature, deformation, etc.), and defect simulation data (leakage location, defect level, etc.), which correspond one-to-one with the data associated with the physical vehicle. Virtual auxiliary test data includes the parameters of the virtual rain equipment (spray pressure, flow rate, etc.) and the environmental parameters of the virtual test scenario (temperature, humidity, etc.), which correspond one-to-one with the physical test auxiliary data. The virtual simulation data is collected and stored in real time, ensuring synchronization with the physical data and providing virtual data support for subsequent data fusion, achieving complementarity between virtual and physical data.
[0041] A step-by-step processing approach is adopted to comprehensively preprocess the collected test vehicle-related data, test auxiliary data, and virtual simulation data to ensure the validity and consistency of the data: The first step is data cleaning, which removes missing values, outliers, and duplicate data from various data sets. For missing values, the mean imputation method (suitable for continuous data such as environmental parameters and spray parameters) or the mode imputation method (suitable for discrete data such as defect levels) is used to fill in the missing values, ensuring data integrity. For outliers, the 3σ criterion is used for identification, and data exceeding the mean ± 3 times the standard deviation are judged as outliers. Based on the actual experimental situation, if the outlier is caused by acquisition error, it is deleted; if it is a real anomaly during the experiment (such as a sudden change in spray pressure caused by equipment failure), it is marked separately and retained for subsequent anomaly analysis. For duplicate data, duplicate records are deleted through data comparison to avoid data redundancy.
[0042] The second step is data denoising. Wavelet denoising algorithms are used to remove data noise, focusing on data such as test vehicle status data and defect data that are susceptible to sensor noise interference. First, the original data is decomposed into low-frequency components (effective signals) and high-frequency components (noise signals). Then, thresholding is performed on the high-frequency components. By setting a reasonable threshold, the high-frequency signals corresponding to noise are removed, while the effective high-frequency signals are retained. Finally, inverse wavelet transform is performed on the processed low-frequency and high-frequency components to reconstruct the denoised clean data, improving the accuracy of the data.
[0043] The third step is data standardization. Due to differences in the format and units of various types of data (e.g., spray pressure is in MPa, temperature is in ℃, and deformation is in mm), the min-max standardization method is used to map all data to the [0,1] interval, unify the data format and units, eliminate data heterogeneity, and ensure that various types of data can be associated, aligned and merged in the subsequent process. After preprocessing, standardized data is output and a standardized data archive is established, with information such as data source and processing time marked.
[0044] Establish data association rules for test vehicles, delineate the entire association logic, and, centered on test vehicle data association, clarify the entire association logic of "vehicle inherent structure data - test process state data - waterproof defect association data - test auxiliary data - virtual simulation data": Using vehicle inherent structure data as the basic association benchmark, test process state data is associated and matched with vehicle inherent structure data to clarify the changing patterns of vehicle state under different structural parameters; waterproof defect association data is bidirectionally associated with vehicle inherent structure data and test process state data to clarify the correspondence between defects and vehicle structure and state; test auxiliary data is linked and matched with vehicle state data and defect association data to clarify the impact of different spray parameters and environmental parameters on vehicle state and defects; virtual simulation data is associated one-to-one with physical data (vehicle association data, auxiliary data) to ensure the correspondence between virtual and physical data. Secondly, standardized data is associated and aligned with test vehicles through timestamp alignment and association identifier alignment, as follows: The process of aligning standardized data with test vehicles using timestamp alignment is as follows: A unified timestamp is added to each piece of standardized data, using the format "year, month, day, hour, minute, second, millisecond" to accurately correspond to the moment of data collection and generation, ensuring that each piece of data has a unique time identifier. Based on the timeline of the test vehicle's testing process, all timestamped standardized data are sorted chronologically. Test vehicle status data, waterproofing defect data, rain equipment parameter data, and virtual simulation data collected at the same time are matched accordingly, achieving synchronous alignment of different types of data in the time dimension. This avoids data misalignment caused by time deviations, ensuring that all types of data at the same moment correspond to the same state of the test vehicle, providing a time synchronization foundation for subsequent data fusion.
[0045] The process of aligning standardized data with test vehicles using association identifiers is as follows: Based on the established test vehicle data association rules, a unique global association identifier (e.g., "Test Vehicle-001") is assigned to each test vehicle. This identifier is used throughout the entire testing process, serving as the unique identity of the test vehicle. Simultaneously, sub-identifiers (e.g., "A-pillar-001-Structural Data" and "Rear Door-002-Defect Data") are assigned to different structural parts of the test vehicle (e.g., A-pillar, tailgate, door seals), different testing stages (e.g., spray test, defect detection), and various types of data (e.g., structural data, status data, defect data) and are associated with the global identifier, clarifying the correspondence between the sub-identifiers and the global identifier. All standardized data are bound to their corresponding sub-identifiers, clearly defining the test vehicle, specific part of the vehicle, and testing stage corresponding to each piece of data to avoid data confusion. Finally, the standardized data bound to sub-identifiers are matched with the test vehicle using the global association identifier, integrating all data under the same global identifier to achieve full-dimensional alignment between data and the test vehicle, ensuring that each piece of data accurately corresponds to a specific test vehicle and related associated objects. After the association alignment is completed, the associated data is output. The associated data includes information such as the data itself, timestamp, and association identifier, which clarifies the association relationship between various types of data and the correspondence with the test vehicle. It can be directly used for subsequent hierarchical association fusion processing.
[0046] S300: Performs hierarchical correlation and fusion on the correlated data, and outputs fused data related to the waterproof performance of automobiles.
[0047] In this embodiment, the correlated data is hierarchically correlated and fused to output fused data related to the vehicle's waterproof performance. The specific process is as follows: S301: Perform data layer association and fusion, initially integrate vehicle-related data, test auxiliary data, and virtual simulation data in the associated data, assign higher basic weight to vehicle-related data, and dynamically adjust the weight according to the correlation between the data and the waterproof performance of the vehicle. S302: Perform feature layer correlation fusion, extract feature parameters that are strongly correlated with the waterproof performance of automobiles after data layer fusion, and use principal component analysis algorithm to reduce the dimensionality of feature parameters, remove redundant features, and retain key correlation features; S303: Perform decision-level association fusion, combine automotive data association rules, perform decision fusion on key feature parameters extracted from the feature layer, integrate the core information of various data, and output standardized fused data.
[0048] In the above process, the core is the association of test vehicle data, highlighting the leading role of vehicle-related data. Vehicle-related data is given a higher basic weight (60%-80%), while test auxiliary data and virtual simulation data are given lower basic weights (20%-40%), ensuring that vehicle-related data plays a dominant role in the fusion process. Secondly, a dynamic weight adjustment mechanism is established. Based on the correlation between various data types and vehicle waterproofing performance, the weight coefficients are dynamically adjusted; the higher the correlation, the larger the weight coefficient; the lower the correlation, the smaller the weight coefficient. The correlation determination is based on the test vehicle data association rules and combined with experimental experience. For example, vehicle seal deformation data and defect data have the highest correlation with vehicle waterproofing performance, so their weight coefficients are set to the highest; rain equipment spray pressure data and environmental humidity data have the next highest correlation with vehicle waterproofing performance, so their weight coefficients are set to medium; virtual simulation data serves as a supplement, so its weight coefficient is set to low. Then, a weighted average fusion algorithm is used to initially fuse the vehicle-related data, test auxiliary data, and virtual simulation data in the associated data. Each type of data is weighted according to its corresponding weight coefficient to obtain the fused data layer. During the fusion process, relying on associated identifiers and timestamps, it is ensured that various types of data from the same test vehicle, at the same time, and from the same location can be accurately fused, avoiding data misalignment. The core information of various types of data is initially integrated, laying the foundation for subsequent feature layer fusion. The fused data still retains its association with the test vehicle, ensuring data relevance.
[0049] Feature extraction was performed on the fused data, focusing on extracting feature parameters strongly correlated with automotive waterproofing performance. Based on the association rules of the test vehicle data, the extracted feature parameters mainly included: features corresponding to the vehicle-related data (deformation amplitude of seals, stress on seals, pressure on key parts of the vehicle body, leakage location characteristics, leakage rate characteristics, defect level characteristics, etc.), features corresponding to the test auxiliary data (spray pressure change characteristics, spray angle characteristics, ambient temperature and humidity change characteristics, etc.), and features corresponding to the virtual simulation data (virtual seal deformation characteristics, virtual defect simulation characteristics, etc.). During the extraction process, feature extraction algorithms (such as convolutional neural network algorithms) were used to mine features from the fused data, ensuring that the extracted features accurately reflect the core influencing factors of automotive waterproofing performance, while preserving the correlation between features and the test vehicle, and clearly defining the corresponding test vehicle part and test stage for each feature. Secondly, principal component analysis (PCA) is used to reduce the dimensionality of the extracted feature parameters. Since the extracted feature parameters have some redundancy (e.g., the deformation amplitude of the seal is correlated with the magnitude of the seal stress), PCA integrates multiple related feature parameters into a few unrelated principal components, retaining key correlated features that reflect the vehicle's waterproof performance, removing redundant features, reducing data volume, and improving the efficiency of subsequent decision-making fusion. During dimensionality reduction, it is ensured that key correlated features are not lost, and the cumulative contribution rate of the principal components is not less than 85%, ensuring that the dimensionality-reduced data accurately reflects the core information of the original fused data. After dimensionality reduction, key correlated feature parameters are output, which still maintain a correlation with the test vehicle, clarifying the vehicle parts corresponding to each principal component and the factors influencing waterproof performance.
[0050] A decision-making fusion model was constructed using a support vector machine (SVM) algorithm. The model's input consisted of key correlation feature parameters extracted from the feature layer, and its output was standardized fusion data related to automotive waterproofing performance. The model generation process was as follows: Historical test data (including test vehicle-related data, auxiliary data, virtual data, and corresponding waterproofing performance test results) was selected as training samples to train the SVM model. Key correlation feature parameters were used as input, and automotive waterproofing performance evaluation results (such as qualified, unqualified, and defect level) were used as output. Through iterative training, model parameters were adjusted to optimize the model's decision accuracy, ensuring that the model could accurately judge automotive waterproofing performance based on key feature parameters. Simultaneously, the correlation rules between feature parameters and vehicle structure and defects were integrated into the model training process, combining them with test vehicle data association rules, enabling the model to identify the correlation logic between different feature parameters and automotive waterproofing performance. After model training, the key correlation feature parameters extracted from the feature layer were input into the trained SVM model. The model, combined with automotive data association rules, performed decision fusion on various feature parameters, integrating the core information of automotive-related data, auxiliary data, and virtual data to comprehensively judge and integrate various data related to automotive waterproofing performance. Finally, standardized fused data is output, which includes core information from the test vehicle's full-dimensional correlation data, test auxiliary data, and virtual simulation data. The correlation between each data and the vehicle's waterproof performance is clearly defined. The data format is unified and the logic is clear. It can be directly used for subsequent fused data verification, anomaly location, and test report generation. At the same time, the association identifier and timestamp with the test vehicle are retained to ensure data traceability.
[0051] S400: Verify and merge data based on the correlation of vehicle data and locate anomalies to generate a correlation test report.
[0052] In this embodiment, based on the correlation relationship of vehicle data, the fused data is verified and anomalies are located to generate a correlation-based test report. The specific process is as follows: S401: Based on the correlation of vehicle data, a deviation analysis algorithm is used to verify the validity of the fused data. Corresponding verification thresholds are set for different structural parts and different waterproof performance indicators of the test vehicle. S402: Compare the deviation between the fused data and the preset standard data to determine abnormal situations; when the deviation exceeds the set threshold, it is determined to be abnormal data, and the corresponding vehicle-related parts, causes of abnormality, and associated test parameters are located through vehicle data association rules; S403: Select effective fusion data, integrate basic information of the test vehicle, multi-dimensional related data, fusion results and anomaly analysis information, obtain test conclusions, and generate a related test report.
[0053] In the above process, based on the data association rules of the test vehicles, the correlation between the fused data and the waterproof performance of the vehicles was clarified. For different structural parts of the test vehicles (such as the A-pillar, tailgate, and door seals) and different waterproof performance indicators (such as sealing performance and leak-proof performance), corresponding verification thresholds were set based on national automotive rain test standards and testing experience. These verification thresholds were divided into absolute and relative thresholds. For example, the absolute threshold for A-pillar seal deformation was set to ±0.05mm, and the relative threshold for leakage rate was set to ±20%. The verification thresholds for different parts and indicators were differentiated according to the inherent structural parameters of the vehicle and the waterproof performance requirements to ensure the relevance and rationality of the verification thresholds. Secondly, a deviation analysis algorithm was used to verify the validity of the fused data, and a deviation analysis model was built. The model generation process is as follows: Based on historical valid test data, a deviation calculation model between the fused data and the preset standard data was established, using absolute and relative deviation calculation methods. Absolute deviation = |fused data - preset standard data|; Relative deviation = (|fused data - preset standard data| / preset standard data) × 100%; The accuracy of deviation calculation is ensured by optimizing the model training. Then, standardized fused data is input into the deviation analysis model to calculate the absolute and relative deviations between the fused data and the preset standard data for the corresponding parts and indicators. Based on the set verification threshold, the validity of the fused data is judged. If the deviation does not exceed the threshold, it is judged as valid fused data; if the deviation exceeds the threshold, it is judged as abnormal data. At the same time, the association identifier, timestamp, and corresponding vehicle part of the abnormal data are marked, laying the foundation for subsequent anomaly localization.
[0054] Each piece of fused data is compared one by one with the corresponding preset standard data. Combined with the deviation analysis results, data anomalies are identified, and the specific values, deviation magnitudes, corresponding time periods, and vehicle parts involved in the anomalies are determined. When the deviation exceeds a set threshold, it is identified as anomaly data, and the association identifier of the anomaly data is recorded. Through the association identifier, the test vehicle, specific structural parts of the vehicle, test procedures, and related data corresponding to the anomaly data can be traced. Secondly, by using vehicle data association rules, the corresponding vehicle parts, causes of anomalies, and associated test parameters are located for abnormal data. Based on the full-chain association logic of "vehicle structure - test state - defect status - auxiliary parameters," the vehicle parts corresponding to abnormal data (such as the middle section of the A-pillar or the rear door seam) are determined according to the association identifiers. Combining the vehicle's inherent structural data and test process state data, the causes of anomalies are analyzed. For example, if the abnormal data indicates excessive deformation of the seals, combining the seal installation parameters in the vehicle's inherent structural data, the cause is determined to be seal installation deviation. Combining the spray pressure data in the test auxiliary data, it is determined that the excessively high spray pressure exacerbated the deformation anomaly. Simultaneously, the test auxiliary parameters (such as spray pressure and ambient humidity) corresponding to the abnormal data are associated with virtual simulation data. Comparing the deviations between the virtual and real data further verifies the cause of the anomaly, clarifying whether the root cause is a physical equipment failure, a defect in the vehicle itself, or a deviation in the virtual model parameters. This ensures the accuracy and comprehensiveness of anomaly location, laying the foundation for subsequent optimization suggestions.
[0055] Based on the validity verification results, all valid fused data were selected, and outlier data was removed (outlier data was separately organized for anomaly analysis). The valid fused data was then categorized and archived according to the test vehicle's association identifier, data type, and timestamp to ensure the completeness and organization of the valid data. Next, various relevant information was integrated, including basic information about the test vehicle (model, inherent structural parameters, waterproofing process parameters, etc.), comprehensive related data of the test vehicle (inherent structural data, test process status data, waterproofing defect related data), data fusion results (standardized fused data, key related characteristic parameters), and anomaly analysis information (outlier data, outlier locations, outlier causes, related test parameters). Combined with vehicle data association rules, the relationships between various types of information were clarified, forming a complete test data system. Then, based on the integrated information, test conclusions were obtained: determining whether the waterproofing performance of the test vehicle was qualified, clarifying the waterproofing performance of various parts of the vehicle, summarizing the anomalies and root causes during the test, and analyzing the impact of the vehicle's own structure and test parameters on waterproofing performance based on the vehicle data association relationships. Finally, a correlation test report is generated. The report adopts a standardized format, highlighting the correlation logic of the test vehicle data and clarifying the relationship between each test result and the vehicle's own data. The report content includes basic test information, test data summary (effectively fused data, correlated data), fusion result analysis, anomaly analysis (anomaly location, anomaly cause), test conclusions, and targeted optimization suggestions (combining the correlation of vehicle data, proposing optimization schemes related to the vehicle's own structure, waterproofing process, and test parameters, such as adjusting the installation parameters of sealing components and optimizing the spray pressure, etc.). This ensures that the test report can provide accurate and practical support for optimizing the waterproofing performance of vehicles and adjusting test schemes, while retaining all correlated and fused data for subsequent traceability and review.
[0056] Corresponding to the aforementioned embodiments of the vehicle rain test data fusion processing method based on digital twins, this application also provides embodiments of a vehicle rain test data fusion processing system based on digital twins.
[0057] Figure 6 This is a schematic diagram illustrating the structural principle of a digital twin-based vehicle rain test data fusion processing system according to an exemplary embodiment. (Refer to...) Figure 6 The system may include: an entity and virtual mapping module 501, an associated data acquisition module 502, a hierarchical data fusion module 503, and an abnormal state identification module 504; wherein: The entity and virtual mapping module 501 is used to construct a digital twin of the vehicle rain test based on the full-dimensional data association of the test vehicle, establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment, and perform mapping and state synchronization between the physical entity and the virtual model of the test vehicle. The associated data acquisition module 502 is used to collect associated data of the entire life cycle of the test vehicle, test auxiliary data and virtual simulation data, preprocess the collected data, perform data association and alignment, and output associated data. The layered data fusion module 503 is used to perform layered association fusion on the associated data and output fused data related to the waterproof performance of the vehicle. The abnormal state identification module 504 is used to verify and merge data based on the correlation of vehicle data and locate abnormalities, and generate a correlation test report.
[0058] In this embodiment, the entity and virtual mapping module 501 constructs a digital twin of the vehicle rain test based on the full-dimensional data association of the test vehicle, establishes real-time communication between the digital twin and the physical rain test equipment, and performs mapping and state synchronization between the physical entity and the virtual model of the test vehicle; the associated data acquisition module 502 collects the full life cycle associated data, test auxiliary data, and virtual simulation data of the test vehicle, preprocesses the collected data, performs data association alignment, and outputs associated data; the hierarchical data fusion module 503 performs hierarchical association fusion of the associated data and outputs fused data associated with the vehicle's waterproof performance; the abnormal state identification module 504 verifies the fused data based on the vehicle data association relationship and locates anomalies, generating an associated test report; by combining digital twin technology, with the test vehicle data association as the core, a multi-source data association fusion system is constructed to avoid deviation of the core data fusion and missing associations, thereby improving the utilization value of test data and providing reliable support for the refined and personalized processing of vehicle rain tests.
[0059] Regarding the system in the above embodiments, the specific manner in which each module performs its operations has been described in detail in the embodiments related to the method, and will not be elaborated upon here.
[0060] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0061] Accordingly, this application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the above-described method for fusing and processing vehicle rain test data based on digital twins. Figure 7 The diagram shown is a hardware structure diagram of any device with data processing capabilities, which is part of a digital twin-based vehicle rain test data fusion processing system provided in an embodiment of the present invention. (Except for...) Figure 7 In addition to the processor, memory, and network interface shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.
[0062] Accordingly, this application also provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the aforementioned digital twin-based vehicle rain test data fusion processing method. The computer-readable storage medium can be an internal storage unit of any data processing-capable device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data processing-capable device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing-capable device, and can also be used to temporarily store data that has been output or will be output.
[0063] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.
[0064] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.
Claims
1. A method for fusing and processing vehicle rain test data based on digital twins, characterized in that, Includes the following steps: A digital twin of the vehicle rain test was constructed based on the full-dimensional data association of the test vehicle. Real-time communication between the digital twin and the physical rain test equipment was established, and the mapping and state synchronization between the physical entity of the test vehicle and the virtual model were performed. The specific process is as follows: Using the virtual model of the test vehicle as the core, and linking the test scenario twin module and the rain equipment twin module, a digital twin of the vehicle rain test is constructed; Establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment; The inherent structure, real-time status, and defects of the physical test vehicle are mapped to the virtual model in a comprehensive manner. The state between the physical entity of the test vehicle and the virtual model of the test vehicle during synchronization; Collect relevant data on the entire lifecycle of the test vehicle, test auxiliary data, and virtual simulation data; preprocess the collected data; perform data association and alignment; and output the associated data. Perform hierarchical correlation and fusion on the correlated data, and output fused data related to the waterproof performance of automobiles; The specific process is as follows: Data layer association and fusion are carried out. The automotive-related data, test auxiliary data, and virtual simulation data in the associated data are initially merged. The automotive-related data is given a higher basic weight, and the weight is dynamically adjusted according to the correlation between the data and the waterproof performance of the vehicle. Feature layer correlation fusion is performed to extract feature parameters that are strongly correlated with the waterproof performance of automobiles after data layer fusion. Principal component analysis algorithm is used to reduce the dimensionality of the feature parameters, remove redundant features, and retain key correlated features. Decision-level correlation and fusion are carried out. Based on the correlation rules of automobile data, the key feature parameters extracted from the feature layer are fused for decision-making, and the core information of various types of data is integrated to output standardized fused data. Based on the correlation of vehicle data, the system verifies and merges data, locates anomalies, and generates a correlation test report.
2. The method for fusing and processing vehicle rain test data based on digital twin as described in claim 1, characterized in that, In the steps of synchronizing the state between the physical entity of the test vehicle and the virtual model of the test vehicle: Using the multi-source correlated data of the test vehicle itself as the core, the parameters of the vehicle's structural data, status data, defect data and the virtual model of the test vehicle are synchronized, and the parameters of the test scenario and the rain equipment are linked and synchronized with the corresponding status of the virtual model of the test vehicle.
3. The method for fusing and processing vehicle rain test data based on digital twin as described in claim 1, characterized in that, In the steps of collecting relevant data throughout the entire lifecycle of the test vehicle, test auxiliary data, and virtual simulation data, preprocessing the collected data, performing data correlation and alignment, and outputting correlated data: Collect multi-source correlation data of the entire life cycle of the test vehicle; the multi-source correlation data includes vehicle inherent structure data, test process status data, and waterproof defect correlation data. Collect test auxiliary data and virtual simulation data of the digital twin of the vehicle rain test; the test auxiliary data includes rain equipment parameter data and test environment parameter data, and the virtual simulation data includes virtual vehicle correlation data and virtual auxiliary test data; The collected data are preprocessed as follows: missing values, outliers and duplicate data are removed from the data, wavelet denoising algorithm is used to remove data noise, all data are standardized and output as standardized data. Establish data association rules for test vehicles, divide the entire chain of association logic, and align standardized data with test vehicles by aligning timestamps and association identifiers, and output associated data.
4. The method for fusing and processing vehicle rain test data based on digital twin as described in claim 3, characterized in that, In the steps of establishing data association rules for test vehicles, dividing the entire chain of association logic, and associating standardized data with test vehicles through timestamp alignment and association identifier alignment, and then outputting the associated data, the process of associating standardized data with test vehicles through timestamp alignment is as follows: Add a timestamp in a uniform format to each piece of standardized data; The timestamp corresponds to the moment when the data was collected and generated. Based on the timeline of the test vehicle testing process, the standardized data are sorted in chronological order to synchronize and align different types of data in the time dimension.
5. The method for fusing and processing vehicle rain test data based on digital twin as described in claim 4, characterized in that, In the steps of establishing data association rules for test vehicles, dividing the entire chain of association logic, and associating standardized data with test vehicles through timestamp alignment and association identifier alignment, and outputting associated data, the process of associating standardized data with test vehicles through association identifier alignment is as follows: Based on the established data association rules for test vehicles, a unique global association identifier is assigned to each test vehicle. At the same time, sub-identifiers associated with the global identifier are assigned to different structural parts, different test stages, and various types of data of the test vehicle. Standardized data is bound to corresponding sub-identifiers to divide each data point into the test vehicle, specific parts of the vehicle, and test stage. By using a global association identifier, standardized data of all bound sub-identifiers are associated and matched with the test vehicle, achieving full-dimensional alignment between the data and the test vehicle.
6. The method for fusing and processing vehicle rain test data based on digital twin as described in claim 3, characterized in that, In the steps of verifying and fusing data based on vehicle data correlation and locating anomalies to generate a correlation test report: Based on the correlation of vehicle data, a deviation analysis algorithm is used to verify the validity of the fused data. Corresponding verification thresholds are set for different structural parts and different waterproof performance indicators of the test vehicle. By comparing the deviation between the fused data and the preset standard data, abnormal situations can be identified. Effective fusion data is selected, and basic information of the test vehicle, multi-dimensional related data, fusion results and anomaly analysis information are integrated to obtain test conclusions and generate a correlation test report.
7. The method for fusing and processing vehicle rain test data based on digital twin as described in claim 6, characterized in that, In the step of comparing the deviation between the fused data and the preset standard data to identify anomalies: When the deviation exceeds the set threshold, it is judged as abnormal data. Then, the vehicle-related parts, causes of abnormality, and associated test parameters are located through vehicle data association rules.
8. A vehicle rain test data fusion processing system based on digital twins, employing the vehicle rain test data fusion processing method based on digital twins as described in claim 1, characterized in that, It includes an entity and virtual mapping module, a relational data acquisition module, a hierarchical data fusion module, and an abnormal state identification module; among which: The entity and virtual mapping module is used to construct a digital twin of the vehicle rain test based on the full-dimensional data association of the test vehicle, establish real-time communication between the digital twin of the vehicle rain test and the physical rain test equipment, and perform mapping and state synchronization between the physical entity and the virtual model of the test vehicle. The associated data acquisition module is used to collect associated data of the entire life cycle of the test vehicle, test auxiliary data and virtual simulation data, preprocess the collected data, perform data association and alignment, and output associated data. The hierarchical data fusion module is used to perform hierarchical correlation and fusion of correlated data and output fused data related to the waterproof performance of automobiles. The abnormal state identification module is used to verify and merge data based on the correlation of vehicle data and locate abnormalities, and generate a correlation test report.