Vehicle body bending and torsion modal data hierarchical matching and double-layer normalization processing method and system

Abstract

The application provides a vehicle body bending and torsion modal data hierarchical matching and double-layer normalization processing method and system. A standardized mode reconstruction and double-path parallel MAC matching coupling algorithm is designed to unify the mode topological dimension of batch PCH simulation files, automatically identify the vehicle body bending and torsion mode, and completely retain the mode core characteristics. The application avoids the defects of manual mode identification error and low efficiency of traditional single-path comparison. The application solves the dynamic boundary of different dimensions based on pure simulation data and forms a constraint closed loop with double-layer normalization. The application does not need to manually set the scaling interval, completely eliminates the numerical overflow problem caused by incremental simulation and floating point operation, and outputs a unified standard data set to adapt to the needs of NVH analysis and AI model training.

Description

Technical Field

[0001] This application belongs to the field of automotive intelligent data preprocessing technology, and in particular relates to a method and system for hierarchical matching and dual-layer normalization processing of vehicle body bending and torsional modal data. Background Technology

[0002] The entire vehicle, body, and chassis components are all mechanical structures. Modal characteristics are a general term for the inherent vibration characteristics of mechanical structures. With the widespread adoption of CAE simulation technology in automotive R&D, a large amount of modal simulation data and structural quality data for the entire vehicle and its components can be generated. Among these, modal frequencies, modal orders, and structural quality are the core foundational data for automotive NVH (noise, vibration, and harshness) analysis and model training. Traditional methods rely on manual modal identification, data association, anomaly removal, and normalization, which suffers from low efficiency, large errors, and inconsistent standards. Summary of the Invention

[0003] In view of this, this application aims to propose a method and system for hierarchical matching and dual-level normalization of vehicle body bending and torsional modal data, in order to solve the problems of low processing efficiency and difficulty in unifying data processing standards in the existing technology, which relies on manual processing of joint vehicle modal and mass data.

[0004] To achieve the above objectives, the technical solution of this application is implemented as follows: Firstly, this application provides a method for hierarchical matching and dual-level normalization processing of vehicle body bending and torsional modal data, including: Traverse the initial simulation PCH data file, use a preset field as the starting anchor point of the single-order modal data block, read the modal feature values ​​and modal order, determine the validity of the modal feature values ​​through the modal validity judgment condition, and determine the natural frequency based on the valid mode. By performing six-degree-of-freedom vibration component coupling extraction and performing six-degree-of-freedom mode shape ordered reconstruction, generate the current order global normalized mode shape vector. The bending and torsional modes to be identified are obtained. Based on a preset dual-path parallel similarity scoring mechanism, the bending mode similarity and torsional mode similarity calculation are performed simultaneously. The validity of the current sample mode identification is determined by a preset global extreme mode optimization mechanism and dual-mode joint judgment conditions, so as to output a modal structured data table. The dynamic boundary adaptive normalization mapping algorithm is used to establish multi-dimensional independent dynamic boundaries for the cleaned dataset. By differentially scaling the features of each dimension, a preliminary normalized dataset is formed. The multi-dimensional features include bending frequency, torsion frequency and structural quality. The preliminary normalized dataset is segmented and closed-loop corrected using a pre-defined double-layer nested normalization interval closed-loop correction algorithm to output a standardized dataset.

[0005] Secondly, based on the same inventive concept, this application also provides a vehicle body bending and torsional modal data hierarchical matching and dual-layer normalization processing system, including: The mode shape ordered reconstruction module is configured to traverse the initial simulation PCH data file, use a preset field as the starting anchor point of the single-order modal data block, read the modal feature values ​​and modal order, determine the validity of the modal feature values ​​through the modal validity judgment condition, determine the natural frequency based on the valid mode, and generate the current order global normalized mode shape vector by performing six-degree-of-freedom vibration component coupling extraction and performing six-degree-of-freedom mode shape ordered reconstruction. The modal recognition module is configured to acquire the bending and torsional modes to be identified, and simultaneously perform bending and torsional mode similarity calculations based on a preset dual-path parallel similarity scoring mechanism. The module determines the validity of the current sample modal recognition through a preset global extreme mode optimization mechanism and dual-mode joint judgment conditions, and outputs a modal structured data table. The normalization processing module is configured to establish multi-dimensional independent dynamic boundaries for the cleaned pure dataset based on the dynamic boundary adaptive normalization mapping algorithm. By differentially scaling the features of each dimension, it integrates them to form a preliminary normalized dataset. The multi-dimensional features include bending frequency, torsional frequency, and structural quality. The standardized dataset output module is configured to perform segmented closed-loop correction on the preliminary normalized dataset using a preset double-layer nested normalization interval closed-loop correction algorithm to output a standardized dataset.

[0006] Thirdly, based on the same inventive concept, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in the first aspect.

[0007] Fourthly, based on the same inventive concept, this application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions for causing the computer to perform the method as described in the first aspect.

[0008] Compared with existing technologies, the vehicle body bending and torsional modal data hierarchical matching and dual-layer normalization processing method and system described in this application have the following advantages: This application designs a standardized mode shape reconstruction and dual-path parallel MAC matching coupling algorithm, unifies the mode shape topology dimension of batch PCH simulation files, automatically identifies the bending and torsional modes of the vehicle body and completely retains the core features of the modes, avoids the defects of manual mode identification error and low efficiency of traditional single-path comparison, and finally solves the multidimensional dynamic boundary based on pure simulation data and forms a constraint closed loop with double-layer normalization, without the need to manually set the scaling interval, and completely eliminates the numerical out-of-bounds problem caused by incremental simulation and floating-point operation, and outputs a unified and standardized dataset to meet the needs of NVH analysis and AI model training. Attached Figure Description

[0009] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of a method for hierarchical matching and dual-layer normalization of vehicle body bending and torsional modal data according to an embodiment of this application; Figure 2 This is a schematic diagram of the original PCH data described in the embodiments of this application; Figure 3 This is a schematic diagram of a vehicle body bending and torsional modal data hierarchical matching and dual-layer normalization processing system according to an embodiment of this application; Figure 4 This is a schematic diagram of the hardware structure of the electronic device described in an embodiment of this application. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0011] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0012] The embodiments of this application are described in detail below with reference to the accompanying drawings.

[0013] Please see Figure 1 As shown, this embodiment provides a method for hierarchical matching and dual-layer normalization processing of vehicle body bending and torsional modal data, specifically including the following steps: Step S10: Traverse the initial simulation PCH data file, use the preset field as the starting anchor point of the single-order modal data block, read the modal characteristic value and modal order, determine the validity of the modal characteristic value through the modal validity judgment condition, and determine the natural frequency based on the valid mode. By performing six-degree-of-freedom vibration component coupling extraction and performing six-degree-of-freedom mode shape ordered reconstruction, a global normalized mode shape vector of the current order is generated.

[0014] Specifically, in this embodiment, this step is the core algorithm preceding the entire scheme. Addressing the characteristics of disordered PCH text nodes, segmented storage of displacement / rotation, and mixed invalid modes, a four-order reconstruction mechanism—modal filtering, component analysis, node rearrangement, and vector serialization—is designed to achieve strict alignment of mode shape vector dimensions. The specific steps are as follows: Step S101: Modal validity determination and frequency calculation.

[0015] Batch traverse the initial data file, using the $EIGENVALUE field as the starting anchor point for a single-order modal data block, and read the modal feature values. With modal order .

[0016] Construct the criteria for determining modal validity: ; A mode is considered valid only if its eigenvalue is greater than 0; if The current order data is discarded and not used in mode construction.

[0017] Solve for the natural frequencies of all effective modes: ; Finally, the obtained inherent frequencies are written into a structured CSV data table for storage. It participates in subsequent adaptive normalization and closed-loop correction algorithm calculations.

[0018] Step S102: Extraction of six-degree-of-freedom vibration components.

[0019] Extracting the 3D translation components of nodes 3D rotational components Construct a complete six-degree-of-freedom mode shape vector at a single node: ; Step S103: Correction of node sequence number ascending order.

[0020] Raw PCH data file (see diagram of raw PCH data) Figure 2 (As shown) The nodes are output in an unordered manner, and direct splicing will cause the mode shape vector dimension to be disordered. Therefore, this embodiment designs a node numbering sorting correction mechanism: that is, sorting all nodes of the current mode by node number. The keywords are rearranged in ascending order to ensure that the node topology order of each mode splicing is completely consistent.

[0021] Step S104: Construct a global normalized mode shape vector sequence.

[0022] After sorting, concatenate the six-degree-of-freedom vectors one by one according to the node order to generate the global normalized mode shape column vector for the current order: ; All simulation files output mode shape vector dimensions and topological order are completely unified, completing the structured and standardized output of batch modal data.

[0023] Step S20: Obtain the bending mode and torsional mode shape to be identified. Simultaneously perform bending mode similarity and torsional mode similarity calculation based on the preset dual-path parallel similarity scoring mechanism. Determine the validity of the current sample mode identification through the preset global extreme value mode optimization mechanism and dual-mode joint judgment conditions, so as to output the modal structured data table.

[0024] Specifically, in this embodiment, this step is the core algorithm for modal recognition. It constructs a dedicated modal splitting and recognition mechanism based on dual-path synchronous scoring, global extreme value optimization, and joint legality determination, enabling automatic differentiation of bending and torsion target modes in a single file. The specific steps are as follows: Step S201: Curing of the dual-modal reference template.

[0025] Based on experience in whole-vehicle NVH simulation engineering, two types of typical target modes are manually located in the benchmark modal cache library, and the benchmark parameters of typical body modes are fixed: for the first-order bending mode of the body, the corresponding global order is defined within the benchmark file. The mode shape vector of this order is read and denoted as For the first-order torsional mode of the vehicle body, the corresponding global order within the baseline file. The mode shape vector of this order is read and denoted as .

[0026] Step S202: Establish a dual-path parallel similarity scoring mechanism.

[0027] For any mode shape to be identified Simultaneously perform parallel calculations of bending mode similarity and torsional mode similarity: ; Step S203: Establish a global extreme value mode optimization mechanism.

[0028] Perform global optimal filtering on the scoring results of all orders of modes in a single file: ; ; The optimal filtering results are used to automatically match the most suitable first-order bending mode and first-order torsional mode in the current file, thereby achieving automatic mode sorting and classification.

[0029] Step S204: Joint legality determination of dual modes.

[0030] This embodiment sets a matching threshold. Construct a dual-condition joint decision formula: ; If two modalities simultaneously meet the threshold requirements, the current sample modality recognition is deemed valid; otherwise, the modality matching is marked as abnormal.

[0031] After a sample is determined to be a valid match, the optimal bending order, bending natural frequency, optimal torsional order, and torsional natural frequency are written into a modal structured CSV data table and stored as shown in Table 1. This data is then directly fed into the dynamic boundary adaptive normalization mapping algorithm for computation.

[0032] Table 1. Data after MAC matching and dual-modal joint determination Step S30: Based on the dynamic boundary adaptive normalization mapping algorithm, establish a multi-dimensional independent dynamic boundary for the cleaned pure dataset. By differentially scaling the features of each dimension, a preliminary normalized dataset is formed. The multi-dimensional features include bending frequency, torsion frequency, and structural quality.

[0033] Specifically, in this embodiment, this step is the core algorithm for data standardization. It abandons the manual method of fixing intervals and uses the global statistical extreme values ​​of clean, anomaly-free data after cleaning as dynamic boundaries to achieve adaptive standardization mapping of multiple batches of simulation data. The specific steps are as follows: Step S301: Perform multidimensional dynamic boundary solution.

[0034] For the three core features of bending frequency, torsional frequency, and structural mass, the boundary conditions are solved independently based on the cleaned dataset: ; In the formula, It represents three feature dimensions, and each dimension independently generates its own dynamic upper and lower boundaries.

[0035] Step S302: Construct a dimension-adaptive linear mapping algorithm.

[0036] Unified setting of global standardized target range To address the NVH modal characteristics of the vehicle body, a dedicated mapping model is constructed using modal sensitivity weighted coefficients based on three core features. The specific coefficients are as follows: These are bending mode sensitive weighting coefficients, used to amplify subtle frequency differences corresponding to body bending stiffness fluctuations. These are torsional modal sensitivity weighting coefficients, used to retain modal characteristic shifts caused by changes in the vehicle body's torsional structure. It is the steady-state weighting coefficient for the vehicle mass, used to suppress invalid noise in static mass parameters and retain effective structural differences.

[0037] Construct a weighted modal offset adaptive scaling formula for vehicle body bending frequency: ; In the formula, This is the bending resonance correction factor.

[0038] Construct a weighted modal offset adaptive scaling formula oriented towards vehicle body torsional frequency: ; In the formula, This is the torsional stiffness attenuation compensation coefficient.

[0039] Construct a weighted modal offset adaptive scaling formula for vehicle mass: ; In the formula, , , These represent the bending frequency, torsional frequency, and total vehicle mass, respectively. , , These represent the steady-state mean values ​​of bending frequency, torsional frequency, and vehicle mass in the clean dataset, respectively, used to simulate the baseline state of the vehicle structure under batch simulation conditions. , , These represent the initial normalized eigenvalues ​​after scaling.

[0040] After performing differential weighted mapping operations on a sample-by-sample basis, a normalized dataset with physical feature discrimination and adapted to the vehicle body bending and torsional NVH scenario is generated. This dataset is then used as input data for the double-layer nested normalized interval closed-loop correction algorithm to perform piecewise constraint correction.

[0041] Step S40: Perform segmented closed-loop correction on the preliminary normalized dataset using a preset double-layer nested normalization interval closed-loop correction algorithm to output a standardized dataset.

[0042] Specifically, in this embodiment, traditional industry data preprocessing typically employs single-layer normalization. This scheme constructs a two-layer mathematical constraint closed loop consisting of a linear mapping layer and a strong pruning layer to address numerical bounds issues caused by incremental simulation data and floating-point calculation errors. Existing clean data after dynamic boundary mapping can guarantee range compliance, but new simulation samples and floating-point iterative calculations may introduce... Therefore, this embodiment designs an independent segmented correction mechanism. The specific steps are as follows: Step S401: Design a hierarchical determination algorithm.

[0043] First, the preliminary normalized eigenvalues ​​are... Perform effective interval classification to achieve automatic region division, feature values Include , and .set up =-1.0, =1.0; =-0.9, =0.9. The core algorithm formula for this step is as follows: ; Using the above stratification formula, all normalized samples are divided into three categories: 0 (stable and valid interval), 1 (buffered slightly out-of-bounds interval), and 2 (limited out-of-bounds interval). The stable interval data retains the original mapping characteristics; the buffered interval data exhibits slight out-of-bounds behavior and is smoothed to preserve the modal gradient; the limited interval data represents invalid drift values, and hard constraint locking is applied to achieve differentiated and precise correction.

[0044] Step S402: Segmented smooth closed-loop correction algorithm.

[0045] For the three types of determination areas in step S401 The interval values ​​are obtained by applying point-by-point constraints to all normalized results and then performing a piecewise smoothing closed-loop correction algorithm. , These are the standard eigenvalues ​​after correction. The core algorithm formula for this step is shown below: ; In the formula, The modal smoothing correction coefficient for the vehicle body ranges from 0 to 0.15 and is used to perform slight smoothing of the values ​​of slightly out-of-bounds modalities. The final output is a standardized dataset that strictly falls within [-1, 1] without any out-of-bounds conditions, as shown in Table 2. This dataset can be directly used for vehicle NVH resonance verification and AI proxy model training.

[0046] Table 2 Final data after interval closed-loop correction This application designs a standardized mode shape reconstruction and dual-path parallel MAC matching coupling algorithm, unifies the mode shape topology dimension of batch PCH simulation files, automatically identifies the vehicle body bending and torsional modes and completely retains the core modal features, avoiding the defects of manual modal identification errors and the low efficiency of traditional single-path comparison. Finally, it solves the multidimensional dynamic boundary based on pure simulation data and forms a constraint closed loop with double-layer normalization, eliminating the need for manual scaling interval setting, and completely eliminating the numerical out-of-bounds problem caused by incremental simulation and floating-point operations. It outputs a unified and standardized dataset to meet the needs of NVH analysis and AI model training.

[0047] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, the embodiments of this application also provide a vehicle body bending and torsional modal data hierarchical matching and dual-layer normalization processing system.

[0049] like Figure 3 As shown, the vehicle body bending and torsional modal data hierarchical matching and dual-layer normalization processing system includes: The mode shape ordered reconstruction module 11 is configured to traverse the initial simulation PCH data file, use a preset field as the starting anchor point of the single-order modal data block, read the modal feature values ​​and modal order, determine the validity of the modal feature values ​​through the modal validity judgment condition, determine the natural frequency based on the valid mode, and generate the current order global normalized mode shape vector by performing six-degree-of-freedom vibration component coupling extraction and performing six-degree-of-freedom mode shape ordered reconstruction. The modal recognition module 12 is configured to acquire the bending mode and torsional mode shape to be identified, and simultaneously perform bending mode similarity and torsional mode similarity calculation based on a preset dual-path parallel similarity scoring mechanism. The validity of the current sample modal recognition is determined by a preset global extreme value mode optimization mechanism and dual-mode joint judgment conditions, so as to output a modal structured data table. The normalization processing module 13 is configured to establish multi-dimensional independent dynamic boundaries for the cleaned pure dataset based on the dynamic boundary adaptive normalization mapping algorithm, and to integrate the features of each dimension into a preliminary normalized dataset by differential scaling; wherein, the multi-dimensional features include bending frequency, torsion frequency and structural quality. The standardized dataset output module 14 is configured to perform segmented closed-loop correction on the preliminary normalized dataset using a preset double-layer nested normalization interval closed-loop correction algorithm to output a standardized dataset.

[0050] For ease of description, the above system is described by dividing it into various modules based on their functions. Of course, in implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware.

[0051] The system described in the above embodiments is used to implement the corresponding method in any of the foregoing embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0052] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the methods described in any of the above embodiments.

[0053] Figure 4 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0054] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0055] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0056] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0057] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0058] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0059] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0060] The electronic devices described above are used to implement the corresponding methods in any of the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0061] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to perform the methods described in any of the above embodiments.

[0062] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0063] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to perform the methods described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0064] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0065] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the claims of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A method for hierarchical matching and dual-level normalization processing of vehicle body bending and torsional modal data, characterized in that, include: Traverse the initial simulation PCH data file, use a preset field as the starting anchor point of the single-order modal data block, read the modal feature values ​​and modal order, determine the validity of the modal feature values ​​through the modal validity judgment condition, and determine the natural frequency based on the valid mode. By performing six-degree-of-freedom vibration component coupling extraction and performing six-degree-of-freedom mode shape ordered reconstruction, generate the current order global normalized mode shape vector. The bending and torsional modes to be identified are obtained. Based on a preset dual-path parallel similarity scoring mechanism, the bending mode similarity and torsional mode similarity calculation are performed simultaneously. The validity of the current sample mode identification is determined by a preset global extreme mode optimization mechanism and dual-mode joint judgment conditions, so as to output a modal structured data table. The dynamic boundary adaptive normalization mapping algorithm is used to establish multi-dimensional independent dynamic boundaries for the cleaned dataset. By differentially scaling the features of each dimension, a preliminary normalized dataset is formed. The multi-dimensional features include bending frequency, torsion frequency and structural quality. The preliminary normalized dataset is segmented and closed-loop corrected using a pre-defined double-layer nested normalization interval closed-loop correction algorithm to output a standardized dataset.

2. The method according to claim 1, characterized in that, The process of extracting six-degree-of-freedom vibration components by coupling and performing ordered reconstruction of six-degree-of-freedom mode shapes to generate a globally normalized mode shape vector of the current order includes: Extract the three-dimensional translational and rotational components of the nodes and construct the single-node six-degree-of-freedom mode shape vector; The nodes are rearranged and corrected in ascending order based on their node numbers. The six-degree-of-freedom mode shape vectors are then connected in series according to the node order to generate the global normalized mode shape vector for the current order.

3. The method according to claim 1, characterized in that: Based on experience in whole-vehicle NVH simulation engineering, the target typical modes are manually located in the benchmark modal cache library, and the benchmark parameters of the typical body modes are fixed, including: For the first-order bending mode of the vehicle body, the corresponding global order within the reference file is used to read the bending mode shape vector of the corresponding order. For the first torsional mode of the vehicle body, the corresponding global order within the reference file is used to read the torsional mode shape vector of the corresponding order. For any modal shape vector to be identified, the similarity of bending mode and torsional mode is calculated in parallel in two dimensions simultaneously, and the global optimal selection is performed on the scoring results of all orders of modes in a single file; By setting a matching threshold and determining the sample mode as a valid match based on the constructed dual-modal joint judgment conditions, the optimal bending order, bending natural frequency, optimal torsional order, and torsional natural frequency are written into the modal structured data table.

4. The method according to claim 1, characterized in that: Based on three feature dimensions—bending frequency, torsional frequency, and structural quality—dynamic upper and lower boundaries are independently generated using the cleaned dataset. A unified global standardized target range is set, and modal-sensitive weighting coefficients under corresponding features are introduced to construct a mapping model. After completing the differential weighted mapping operation for each sample, a normalized dataset with physical feature discrimination and adapted to the vehicle body bending and torsional modal NVH scenario is generated.

5. The method according to claim 4, characterized in that: Construct a weighted modal offset adaptive scaling formula for vehicle body bending frequency: ； Construct a weighted modal offset adaptive scaling formula oriented towards vehicle body torsional frequency: ； Construct a weighted modal offset adaptive scaling formula for vehicle mass: ； In the formula, This represents the bending resonance correction factor. This represents the torsional stiffness attenuation compensation coefficient. , , Let represent the steady-state mean values ​​of bending frequency, torsional frequency, and total vehicle mass in the clean dataset, respectively. , , These represent the bending frequency, torsional frequency, and total vehicle mass, respectively. , , These represent the bending mode sensitivity weighting coefficient, the torsional mode sensitivity weighting coefficient, and the vehicle mass steady-state weighting coefficient, respectively. , Represents the globally standardized target range. , Indicates the upper and lower boundaries. , , These represent the initial normalized eigenvalues ​​after scaling.

6. The method according to claim 5, characterized in that: A hierarchical decision-making algorithm is used to classify the preliminary normalized feature values ​​into regions, dividing all normalized samples into valid intervals, buffered slightly out-of-bounds intervals, and extreme out-of-bounds intervals; wherein, the formula of the hierarchical decision-making algorithm is: ； In the formula, There are three types of decision regions, and the feature values ​​are... Include , and ,set up =-0.9, =0.

9.

7. The method according to claim 6, characterized in that: Based on the interval values ​​of the three types of decision regions, point-by-point constraints are applied to all normalized results, and the corrected standard feature values ​​are obtained by executing a piecewise smoothing closed-loop correction algorithm; wherein, the formula of the piecewise smoothing closed-loop correction algorithm is as follows: ； In the formula, This is the vehicle body modal smoothing correction coefficient, with a value ranging from 0 to 0.

15.

8. A system for hierarchical matching and dual-level normalization processing of vehicle body bending and torsional modal data, characterized in that, include: The mode shape ordered reconstruction module is configured to traverse the initial simulation PCH data file, use a preset field as the starting anchor point of the single-order modal data block, read the modal feature values ​​and modal order, determine the validity of the modal feature values ​​through the modal validity judgment condition, determine the natural frequency based on the valid mode, and generate the current order global normalized mode shape vector by performing six-degree-of-freedom vibration component coupling extraction and performing six-degree-of-freedom mode shape ordered reconstruction. The modal recognition module is configured to acquire the bending and torsional modes to be identified, and simultaneously perform bending and torsional mode similarity calculations based on a preset dual-path parallel similarity scoring mechanism. The module determines the validity of the current sample modal recognition through a preset global extreme mode optimization mechanism and dual-mode joint judgment conditions, and outputs a modal structured data table. The normalization processing module is configured to establish multi-dimensional independent dynamic boundaries for the cleaned pure dataset based on the dynamic boundary adaptive normalization mapping algorithm. By differentially scaling the features of each dimension, it integrates them to form a preliminary normalized dataset. The multi-dimensional features include bending frequency, torsional frequency, and structural quality. The standardized dataset output module is configured to perform segmented closed-loop correction on the preliminary normalized dataset using a preset double-layer nested normalization interval closed-loop correction algorithm to output a standardized dataset.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as claimed in any one of claims 1-7.

10. A non-transitory computer-readable storage medium, characterized in that, in, The non-transitory computer-readable storage medium stores computer instructions for causing a computer to perform the method described in any one of claims 1-7.

Images