Intelligent Vehicle Maintenance Decision-Making Method Based on Digital Twin

Abstract

This invention discloses a vehicle intelligent maintenance decision-making method based on digital twins, belonging to the field of intelligent maintenance decision-making technology. The method includes acquiring vehicle operating data, calculating crankshaft vibration frequency domain energy and brake pad wear ratio, and constructing a three-dimensional vehicle twin based on the Unity3D engine. It receives fault diagnosis results, maps fault types, instantiates fault description objects based on the mapped fault types, and injects these objects into corresponding components of the vehicle's three-dimensional twin. It then performs multiphysics coupling simulation to generate a multiphysics coupling feature matrix. The obtained multiphysics coupling feature matrix is ​​input into a maintenance decision knowledge graph, which sequentially performs similar case retrieval, predictive maintenance quotation calculation, and workshop resource scheduling, outputting a maintenance decision plan. This invention provides a quantitative and visualized scientific basis for fault mechanism analysis and localization; it significantly improves the pertinence, economy, and workshop resource utilization efficiency of maintenance plans.

Description

Technical Field

[0001] This invention relates to the field of intelligent maintenance decision-making technology, and in particular to a vehicle intelligent maintenance decision-making method based on digital twins. Background Technology

[0002] In recent years, with the rapid development of intelligent connected vehicles, the vehicle operation and maintenance field is transforming from a single experience-driven approach to a data-driven one. On the one hand, the maturity of technologies such as onboard sensors, edge computing, and 5G V2X has made it possible to collect and process multi-source heterogeneous time-series data such as vehicle vibration, current, temperature, and point clouds in real time. On the other hand, the application of digital twin technology in aerospace, intelligent manufacturing, and other fields has demonstrated the methodological advantages of synchronously reflecting the operating status of physical equipment using three-dimensional virtual models. Academia and industry have successively proposed simulation frameworks based on finite element analysis and multiphysics coupling to predict structural stress, heat conduction, and vibration propagation. Simultaneously, the combination of knowledge graphs and deep learning provides end-to-end support for intelligent decision-making, from historical case retrieval to cost prediction. These studies and practices have laid the technical foundation for introducing digital twin technology into vehicle maintenance decision-making scenarios.

[0003] However, existing vehicle repair technologies are limited by the following shortcomings: First, most solutions are limited to diagnostic analysis of a single data source or a single physical field, making it difficult to comprehensively depict the fault evolution process. They also often rely on experience-based time standards and lack visual simulation basis for complex coupled faults. Second, static pricing and manual scheduling modes are difficult to respond to fluctuations in parts prices and dynamic changes in workshop resources, resulting in insufficient decision-making efficiency and economy. Third, repair solutions are mostly presented in a process template format, lacking interactive reasoning and case matching capabilities based on digital twin 3D visualization models. Summary of the Invention

[0004] In view of the problems existing in existing vehicle intelligent maintenance decision-making methods based on digital twins, this invention is proposed. Therefore, the problem to be solved by this invention is how to provide a vehicle intelligent maintenance decision-making method based on digital twins.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] In a first aspect, the present invention provides a vehicle intelligent maintenance decision-making method based on digital twins, comprising:

[0007] Acquire vehicle operating data, calculate crankshaft vibration frequency domain energy and brake pad wear ratio, and construct a three-dimensional twin of the vehicle based on the Unity3D engine;

[0008] The system receives fault diagnosis results, maps fault types, instantiates fault description objects based on the mapped fault types, injects the fault description objects into the corresponding parts of the vehicle's three-dimensional twin, performs multi-physics coupling simulation, and generates a multi-physics coupling feature matrix.

[0009] The obtained multiphysics coupling feature matrix is ​​input into the maintenance decision knowledge graph, which includes a case matching layer, a cost optimization layer, and a resource scheduling layer. Similar case retrieval, predictive maintenance quotation calculation, and workshop resource scheduling are performed in sequence to output a maintenance decision scheme.

[0010] As a preferred embodiment of the vehicle intelligent maintenance decision-making method based on digital twins described in this invention, the vehicle operating data includes crankshaft vibration data, temperature data, and brake pad point cloud data.

[0011] As a preferred embodiment of the vehicle intelligent maintenance decision-making method based on digital twins described in this invention, the calculation of crankshaft vibration frequency domain energy and brake pad wear ratio includes:

[0012] The crankshaft vibration frequency domain energy is calculated using crankshaft vibration data and mapped to a crankshaft stress contour map. The formula for calculating the crankshaft vibration frequency domain energy is as follows:

[0013] ;

[0014] in: frequency band The frequency domain energy of crankshaft vibration The Fourier transform results of the crankshaft vibration data; and For frequency band analysis;

[0015] Brake surface contours are fitted using brake pad point cloud data to calculate brake pad wear ratio. The brake pad wear ratio is then converted into a color heatmap. The formula for calculating the brake pad wear ratio is as follows:

[0016] ;

[0017] in: This refers to the wear ratio of the brake pads. This is the current thickness of the brake pads. This refers to the nominal thickness of the brake pads.

[0018] As a preferred embodiment of the vehicle intelligent maintenance decision-making method based on digital twins described in this invention, the step of performing multi-physics coupled simulation deduction includes:

[0019] In the Unity3D engine, the crankshaft and brake pad models are bound to finite element meshes respectively to obtain crankshaft mesh node sets and brake pad mesh node sets;

[0020] Solving the damped wave equation on the crankshaft mesh node set, it is expressed as:

[0021] ;

[0022] in: For a moment Crankshaft mesh nodes The source of the fault excitation, For displacement field, For material density, For Young's modulus, The damping coefficient is... For gradient;

[0023] The vibration energy of each crankshaft grid node is calculated from the displacement field to identify abnormal vibration regions, as shown below:

[0024] ;

[0025] in: For a moment Vibrational energy;

[0026] Crankshaft mesh nodes whose vibration energy exceeds a predetermined threshold are marked as abnormal nodes. All abnormal nodes are combined into an abnormal node coordinate set and rendered as arrows on the surface of the vehicle's 3D twin crankshaft. The abnormal area is displayed in conjunction with cloud map coloring.

[0027] Solving the steady-state heat conduction equation on the brake pad mesh node set, it can be expressed as:

[0028] ;

[0029] in: As a heat source generated by friction, Thermal conductivity, For real-time temperature field;

[0030] Using the Unity3D engine, historical wear profiles and real-time temperature profiles are mapped onto the same profile geometry to obtain profile data and generate a multiphysics coupling feature matrix containing vibration energy and temperature gradient.

[0031] As a preferred embodiment of the vehicle intelligent maintenance decision-making method based on digital twins described in this invention, the case matching layer is used to perform similar case retrieval on the multi-physics coupling feature matrix of a case, including:

[0032] The multiphysics coupling feature matrix is ​​projected and expressed as:

[0033] ;

[0034] in: This is the projected multiphysics coupling characteristic matrix. This is the multiphysics coupling characteristic matrix. and To separate encoder weights and biases;

[0035] Construct case vectors from the projected multiphysics coupling feature matrix, and calculate the similarity between the new case vectors and all historical case vectors, denoted as:

[0036] ;

[0037] in: For similarity, For historical case vectors, For new case vectors;

[0038] Cases with similarity exceeding a predetermined similarity limit are searched in the historical case database as preliminary candidate solutions.

[0039] As a preferred embodiment of the vehicle intelligent maintenance decision-making method based on digital twins described in this invention, the cost optimization layer is used to calculate the predicted maintenance quote for preliminary candidate solutions, including:

[0040] Historical spare parts prices and raw material indices are input into the LSTM model to predict spare parts prices. The model outputs the predicted spare parts prices for the next predetermined number of days and takes the predicted spare parts prices for the day as the real-time spare parts prices.

[0041] Query the historical maintenance man-hour distribution for similar maintenance faults to obtain the standard deviation and maximum value of historical maintenance man-hours. Calculate the dynamic maintenance difficulty coefficient, expressed as:

[0042] ;

[0043] in: The dynamic difficulty coefficient of maintenance. The standard deviation of historical maintenance man-hours. This represents the maximum historical maintenance man-hours.

[0044] The predicted maintenance cost for each preliminary candidate solution is calculated and expressed as follows:

[0045] ;

[0046] in: To predict repair quotes, Based on the standard hourly rate, This is the real-time price of spare parts.

[0047] As a preferred embodiment of the vehicle intelligent maintenance decision-making method based on digital twins described in this invention, the resource scheduling layer is used for workshop resource scheduling of vehicle maintenance, including:

[0048] Obtain the list of maintenance personnel, equipment status, and the daily working hour limit for maintenance personnel from the intelligent terminal in the workshop;

[0049] A scheduling model is constructed, which breaks down the maintenance process steps into a set of tasks. Each task includes the required skill tags, estimated working hours, and priority weights.

[0050] The constraints are: maintenance personnel skill tag matching, total working hours less than the available working hours for the day, and equipment can only perform one task at a time; the objective is to minimize the completion time.

[0051] Using a heuristic scheduling algorithm, the highest priority task is selected in each round and assigned to the earliest available maintenance personnel and equipment with matching skill tags;

[0052] The output includes a maintenance decision plan, a maintenance procedure sequence, a predicted maintenance quote, and a parts list.

[0053] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a vehicle intelligent maintenance decision-making method based on digital twins.

[0054] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements the steps of a vehicle intelligent maintenance decision-making method based on digital twins.

[0055] The beneficial effects of this invention are as follows: This invention achieves high-fidelity visualization and simulation of the status of key components, providing a quantitative and visual scientific basis for fault mechanism analysis and localization; the maintenance decision-making process driven by knowledge graphs and deep learning, combined with case retrieval, cost prediction and resource scheduling, significantly improves the pertinence, economy and workshop resource utilization efficiency of maintenance solutions. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0057] Figure 1 This is a flowchart of a vehicle intelligent maintenance decision-making method based on digital twins. Detailed Implementation

[0058] To make the above-mentioned objects, features, and advantages of the present invention more readily understood, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0059] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0060] Secondly, the term "one embodiment" or "example" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the invention. An embodiment appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment that selectively excludes other embodiments.

[0061] Reference Figure 1 This is the first embodiment of the present invention, which provides a vehicle intelligent maintenance decision-making method based on digital twins, including:

[0062] S1: Acquire vehicle operating data, calculate crankshaft vibration frequency domain energy and brake pad wear ratio, and construct a three-dimensional twin of the vehicle based on the Unity3D engine;

[0063] Specifically, a parameter-driven 3D twin is built based on the Unity3D engine. The crankshaft stress cloud map module associates vibration sensor data to generate a dynamic gradient shading model, and the brake pad wear module integrates point cloud data from a 3D scanner to generate a thickness attenuation heat map.

[0064] The architecture of an in-vehicle edge computing system includes edge computing devices, vibration sensors, infrared temperature sensing arrays, 3D scanners, and communication modules;

[0065] Vehicle operating data includes crankshaft vibration data collected by vibration sensors, temperature data collected by infrared temperature sensor array, and brake pad point cloud data collected by 3D scanner;

[0066] All sensor synchronization is controlled by a master clock (provided by the edge computing device), and each data point is accompanied by a UTC timestamp.

[0067] The spatial reference system of each sensor is predefined using the vehicle CAD model, and all sensor data is projected onto a unified vehicle coordinate system.

[0068] A 3D twin model was created, and a complete 3D vehicle structure was built based on the Unity3D engine (with a focus on modeling the crankshaft and brake pads). The twin is a parameter-driven body, and the sensor data stream directly controls the model animation / texture rendering.

[0069] The crankshaft vibration frequency domain energy is calculated by collecting crankshaft vibration data from vibration sensors and then mapped to a crankshaft stress cloud map. The formula for calculating the crankshaft vibration frequency domain energy is as follows:

[0070] ;

[0071] in: frequency band The frequency domain energy of crankshaft vibration The Fourier transform results of the crankshaft vibration data; and For frequency band analysis;

[0072] Brake surface contours are fitted by periodically outputting brake pad point cloud data from a 3D scanner. The wear ratio of the brake pads is calculated by comparing the current thickness with the nominal thickness, and is expressed as:

[0073] ;

[0074] in: This refers to the wear ratio of the brake pads. This is the current thickness of the brake pads. This refers to the nominal thickness of the brake pads.

[0075] The wear ratio of the brake pads is converted into a color heat map and then applied to the surface of the brake pad model.

[0076] S2: Receive fault diagnosis results, map fault types, establish fault description objects inside the twin, inject the diagnosed fault types into the three-dimensional twin, perform multi-physics coupling simulation and deduction, and generate multi-physics coupling feature matrix.

[0077] Specifically, the system performs fault type mapping, receives fault diagnosis results, establishes a fault description object within the twin, attaches the object to the corresponding module (crankshaft or brake pad), and triggers the simulation process.

[0078] In the Unity3D engine, the crankshaft and brake pad models are bound to finite element (FEM) meshes respectively to obtain the crankshaft mesh node set and the brake pad mesh node set;

[0079] Solving the damped wave equation on the crankshaft mesh node set, it is expressed as:

[0080] ;

[0081] in: For a moment Crankshaft mesh nodes The source of the fault excitation, For displacement field, For material density, For Young's modulus, The damping coefficient is... For gradient;

[0082] The vibration energy of each crankshaft grid node is calculated from the displacement field to identify abnormal vibration regions, as shown below:

[0083] ;

[0084] in: For a moment Vibrational energy;

[0085] Crankshaft mesh nodes whose vibration energy exceeds a predetermined threshold are marked as abnormal nodes. All abnormal nodes are combined into an abnormal node coordinate set, which is then rendered as arrows on the surface of the vehicle's 3D twin crankshaft. The abnormal areas are displayed in conjunction with cloud map coloring.

[0086] Obtain historical wear gradients and real-time temperature fields. Historical wear gradients: Read multiple point cloud fitting results from the database and calculate the brake pad thickness change rate at the brake pad mesh nodes. Real-time temperature field: The temperature distribution after array interpolation to the brake pad mesh nodes.

[0087] Solving the steady-state heat conduction equation on the brake pad mesh node set, it can be expressed as:

[0088] ;

[0089] in: As a heat source generated by friction, Thermal conductivity, This represents the real-time temperature field.

[0090] Using the Unity3D engine, historical wear profiles and real-time temperature profiles are mapped onto the same profile geometry to obtain profile data and generate a multiphysics coupling feature matrix containing vibration energy and temperature gradient (each row corresponds to a mesh node, and each column corresponds to a type of physical field feature, such as crankshaft vibration frequency domain energy, brake pad wear ratio, etc.).

[0091] S3: Input the obtained multiphysics coupling feature matrix into the maintenance decision knowledge graph, and sequentially perform similar case retrieval, predictive maintenance quotation calculation and workshop resource scheduling to output maintenance decision scheme.

[0092] Specifically, the obtained multiphysics coupling feature matrix is ​​input into the maintenance decision knowledge graph, which includes a case matching layer, a cost optimization layer, and a resource scheduling layer. The case matching layer is used for similar case retrieval; similarity retrieval is performed on the cases, and the multiphysics coupling feature matrix is ​​projected, represented as:

[0093] ;

[0094] in: This is the projected multiphysics coupling characteristic matrix. This is the multiphysics coupling characteristic matrix. and To separate encoder weights and biases;

[0095] Construct case vectors from the projected multiphysics coupling feature matrix, and calculate the similarity between the new case vectors and all historical case vectors, denoted as:

[0096] ;

[0097] in: For similarity, For historical case vectors, For new case vectors;

[0098] Cases with similarity exceeding a predetermined similarity limit are searched in the historical case database as preliminary candidate solutions.

[0099] The cost optimization layer is used to calculate predictive maintenance quotes. Historical parts prices and raw material indices are input into the LSTM model to predict parts prices, and the predicted prices of each part within the next predetermined number of days are output. The predicted prices of parts on the current day are taken as the real-time parts prices.

[0100] Query the historical maintenance man-hour distribution corresponding to similar maintenance faults to obtain the standard deviation and maximum value of historical maintenance man-hours, and calculate the dynamic maintenance difficulty coefficient, which is expressed as:

[0101] ;

[0102] in: The dynamic difficulty coefficient of maintenance. The standard deviation of historical maintenance man-hours. This is the maximum historical maintenance man-hours, the maximum value after removing outliers.

[0103] The predicted maintenance cost for each preliminary candidate solution is calculated and expressed as follows:

[0104] ;

[0105] in: To predict repair quotes, Based on the standard hourly rate, Real-time parts prices;

[0106] The resource scheduling layer is used to schedule workshop resources for vehicle maintenance, and obtains the list of maintenance personnel, equipment status (idle / occupied) and the daily working hour limit of maintenance personnel from the workshop intelligent terminal;

[0107] A scheduling model is constructed, which breaks down the maintenance process steps into a set of tasks. Each task includes the required skill label, estimated working time, and priority weight (derived from the sorting of preliminary candidate solutions by the case matching layer).

[0108] The constraints are: maintenance personnel skill tag matching, total working hours less than the available working hours for the day, and equipment can only perform one task at a time; the objective is to minimize the completion time.

[0109] Using a heuristic scheduling algorithm, the highest priority task is selected in each round and assigned to the earliest available maintenance personnel and equipment with matching skill tags;

[0110] The output includes a repair procedure sequence, a predicted repair quote, and a parts list for repair decision-making.

[0111] This embodiment also provides a computer device applicable to the vehicle intelligent maintenance decision-making method based on digital twins, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement all or part of the steps of the method described in the above embodiments of the present invention.

[0112] This embodiment also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, it performs the method in any optional implementation of the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0113] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0114] In summary, this invention achieves high-fidelity visualization and simulation of the status of key components, providing a quantitative and visual scientific basis for fault mechanism analysis and localization. The maintenance decision-making process driven by knowledge graphs and deep learning, combined with case retrieval, cost prediction, and resource scheduling, significantly improves the pertinence, economy, and workshop resource utilization efficiency of maintenance solutions.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A vehicle intelligent maintenance decision-making method based on digital twins, characterized in that: include, Acquire vehicle operating data, calculate crankshaft vibration frequency domain energy and brake pad wear ratio, and construct a three-dimensional twin of the vehicle based on the Unity3D engine; The vehicle operation data includes crankshaft vibration data, temperature data, and brake pad point cloud data; The system receives fault diagnosis results, maps fault types, instantiates fault description objects based on the mapped fault types, injects the fault description objects into the corresponding parts of the vehicle's three-dimensional twin, performs multi-physics coupling simulation, and generates a multi-physics coupling feature matrix. The execution of the multiphysics coupling simulation includes: In the Unity3D engine, the crankshaft and brake pad models are bound to finite element meshes respectively to obtain crankshaft mesh node sets and brake pad mesh node sets; Solving the damped wave equation on the crankshaft mesh node set, it is expressed as: ; in: For a moment Crankshaft mesh nodes The source of the fault excitation, For displacement field, For material density, For Young's modulus, The damping coefficient is... For gradient; The vibration energy of each crankshaft grid node is calculated from the displacement field to identify abnormal vibration regions, as shown below: ; in: For a moment Vibrational energy; Crankshaft mesh nodes whose vibration energy exceeds a predetermined threshold are marked as abnormal nodes. All abnormal nodes are combined into an abnormal node coordinate set and rendered as arrows on the surface of the vehicle's 3D twin crankshaft. The abnormal area is displayed in conjunction with cloud map coloring. Solving the steady-state heat conduction equation on the brake pad mesh node set, it can be expressed as: ; in: As a heat source generated by friction, Thermal conductivity, For real-time temperature field; Using the Unity3D engine, historical wear profiles and real-time temperature profiles are mapped onto the same profile geometry to obtain profile data and generate a multiphysics coupling feature matrix containing vibration energy and temperature gradient. The obtained multiphysics coupling feature matrix is ​​input into a maintenance decision knowledge graph, which includes a case matching layer, a cost optimization layer, and a resource scheduling layer. The graph sequentially performs similar case retrieval, predictive maintenance quote calculation, and workshop resource scheduling, outputting a maintenance decision plan. The case matching layer is used to perform similar case retrieval on the multiphysics coupling feature matrix of the cases, including: The multiphysics coupling feature matrix is ​​projected and expressed as: ; in: This is the projected multiphysics coupling characteristic matrix. This is the multiphysics coupling characteristic matrix. and To separate encoder weights and biases; Construct case vectors from the projected multiphysics coupling feature matrix, and calculate the similarity between the new case vectors and all historical case vectors, denoted as: ; in: For similarity, For historical case vectors, For new case vectors; Cases with similarity exceeding a predetermined similarity limit are retrieved from the historical case database as preliminary candidate solutions; The cost optimization layer is used to calculate predicted maintenance quotes for preliminary candidate solutions, including: Historical spare parts prices and raw material indices are input into the LSTM model to predict spare parts prices. The model outputs the predicted spare parts prices for the next predetermined number of days and takes the predicted spare parts prices for the day as the real-time spare parts prices. Query the historical maintenance man-hour distribution for similar maintenance faults to obtain the standard deviation and maximum value of historical maintenance man-hours. Calculate the dynamic maintenance difficulty coefficient, expressed as: ; in: The dynamic difficulty coefficient of maintenance. The standard deviation of historical maintenance man-hours. This represents the maximum historical maintenance man-hours. The predicted maintenance cost for each preliminary candidate solution is calculated and expressed as follows: ; in: To predict repair quotes, Based on the standard hourly rate, Real-time parts prices; The resource scheduling layer is used for workshop resource scheduling of vehicle maintenance, including: Obtain the list of maintenance personnel, equipment status, and the daily working hour limit for maintenance personnel from the intelligent terminal in the workshop; A scheduling model is constructed, which breaks down the maintenance process steps into a set of tasks. Each task includes the required skill tags, estimated working hours, and priority weights. The constraints are: maintenance personnel skill tag matching, total working hours less than the available working hours for the day, and equipment can only perform one task at a time; the objective is to minimize the completion time. Using a heuristic scheduling algorithm, the highest priority task is selected in each round and assigned to the earliest available maintenance personnel and equipment with matching skill tags; The output includes a maintenance decision plan, a maintenance procedure sequence, a predicted maintenance quote, and a parts list.

2. The vehicle intelligent maintenance decision-making method based on digital twin as described in claim 1, characterized in that: The calculation of crankshaft vibration frequency domain energy and brake pad wear ratio includes: The crankshaft vibration frequency domain energy is calculated using crankshaft vibration data and mapped to a crankshaft stress contour map. The formula for calculating the crankshaft vibration frequency domain energy is as follows: ; in: frequency band The frequency domain energy of crankshaft vibration The Fourier transform results of the crankshaft vibration data; and For frequency band analysis; Brake surface contours are fitted using brake pad point cloud data to calculate brake pad wear ratio. The brake pad wear ratio is then converted into a color heatmap. The formula for calculating the brake pad wear ratio is as follows: ; in: This refers to the wear ratio of the brake pads. This is the current thickness of the brake pads. This refers to the nominal thickness of the brake pads.

3. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the vehicle intelligent maintenance decision-making method based on digital twins as described in any one of claims 1 to 2.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the vehicle intelligent maintenance decision-making method based on digital twins as described in any one of claims 1 to 2.

Images