Intelligent Diagnosis Method and System for New Energy Vehicle Chassis Faults Based on Knowledge Graph

Abstract

The present invention provides an intelligent diagnosis method and system for new energy vehicle chassis faults based on a knowledge graph, which relates to the field of data processing. The method includes: determining an initial chassis monitoring plan according to the three-dimensional model of the new energy vehicle chassis, and installing a first chassis monitoring component on the physical model corresponding to the new energy vehicle chassis; obtaining the chassis state information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes through the first chassis monitoring component, determining the optimal chassis monitoring plan, and establishing a chassis fault diagnosis knowledge graph; installing a second chassis monitoring component on the new energy vehicle chassis according to the optimal chassis monitoring plan, and obtaining the real-time chassis state information of the new energy vehicle chassis; generating fault diagnosis information for the new energy vehicle chassis according to the real-time chassis state information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph, which has the advantage of improving the accuracy of fault diagnosis for the new energy vehicle chassis.

Description

Technical Field

[0001] The present invention relates to the field of automotive chassis fault diagnosis, and particularly to an intelligent fault diagnosis method and system for new energy vehicle chassis based on a knowledge graph. Background Art

[0002] The chassis is one of the important components of an automobile. It supports and connects various components such as the body, engine, transmission system, and suspension system. Faults in chassis components may cause many problems, such as unstable driving, difficult steering, poor braking, loose suspension, etc., and may even seriously affect driving safety. Therefore, it is very necessary and important to conduct chassis fault diagnosis. Chassis fault diagnosis can improve driving safety; the chassis is an important support component during the driving process of an automobile. If the chassis fails or is damaged, it may cause the vehicle to be unstable, the driving to be unsmooth, and even serious consequences such as brake failure and difficult steering. Therefore, timely detection and diagnosis of chassis faults are of great significance for improving driving safety. The chassis components are interrelated. When a component fails, it may affect the normal operation of other components and even cause further damage. Timely chassis fault diagnosis and repair can effectively avoid the expansion of faults and secondary damage; chassis faults may cause the driving condition of the vehicle to deteriorate and the wear to increase, which will affect the life of the automobile. If chassis faults are detected and diagnosed in a timely manner and appropriate repairs and maintenance are carried out, the long-term normal operation of the vehicle can be ensured and the service life of the automobile can be extended.

[0003] In the prior art, chassis fault diagnosis mainly determines whether a fault occurs by simply comparing the collected data with a preset threshold. The false alarm rate of the fault alarm method is high and the safety factor is low.

[0004] Therefore, there is a need to provide an intelligent fault diagnosis method and system for new energy vehicle chassis based on a knowledge graph to improve the accuracy of new energy vehicle chassis fault diagnosis. Summary of the Invention

[0005] The present invention provides an intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph, including: obtaining a three-dimensional model of the new energy vehicle chassis; determining an initial chassis monitoring plan according to the three-dimensional model of the new energy vehicle chassis, where the initial chassis monitoring plan includes a plurality of initial chassis monitoring positions and the monitoring feature types corresponding to each initial chassis monitoring position; obtaining a physical model corresponding to the new energy vehicle chassis; installing a first chassis monitoring component on the physical model corresponding to the new energy vehicle chassis according to the initial chassis monitoring plan, where the first chassis monitoring component includes a plurality of sensors installed according to the initial chassis monitoring plan; obtaining chassis state information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes through the first chassis monitoring component; determining an optimal chassis monitoring plan according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes, where the optimal chassis monitoring plan includes a plurality of target chassis monitoring positions and the monitoring feature types corresponding to each target chassis monitoring position; establishing a chassis fault diagnosis knowledge graph according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes; installing a second chassis monitoring component on the new energy vehicle chassis according to the optimal chassis monitoring plan, where the second chassis monitoring component includes a plurality of sensors installed according to the optimal chassis monitoring plan; obtaining real-time chassis state information of the new energy vehicle chassis through the second chassis monitoring component; and generating fault diagnosis information of the new energy vehicle chassis according to the real-time chassis state information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph.

[0006] Further, determining the initial chassis monitoring plan according to the three-dimensional model of the new energy vehicle chassis includes: establishing a finite element model of the new energy vehicle chassis according to the three-dimensional model of the new energy vehicle chassis; and determining the initial chassis monitoring plan according to the finite element model of the new energy vehicle chassis.

[0007] Further, determining the initial chassis monitoring plan according to the finite element model of the new energy vehicle chassis includes: performing stress analysis according to the finite element model of the new energy vehicle chassis to determine the initial vibration monitoring positions; performing heat conduction analysis according to the finite element model of the new energy vehicle chassis to determine the initial temperature monitoring positions; determining candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions; and performing electromagnetic field analysis according to the finite element model of the new energy vehicle chassis to determine the initial current monitoring positions, initial voltage monitoring positions, and initial sound monitoring positions from the candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions, where the plurality of initial chassis monitoring positions at least include the initial vibration monitoring positions, initial temperature monitoring positions, initial current monitoring positions, initial voltage monitoring positions, and initial sound monitoring positions.

[0008] Further, based on the test chassis status information of the physical model corresponding to the new energy vehicle chassis under multiple fault modes, an optimal chassis monitoring scheme is determined, including: for each fault mode, based on the test chassis status information of the physical model corresponding to the new energy vehicle chassis under the fault mode, determine the abnormal parameters corresponding to each initial chassis monitoring position for the fault mode; based on the abnormal parameters corresponding to each initial chassis monitoring position for each fault mode, screen the multiple initial chassis monitoring positions to determine multiple intermediate chassis monitoring positions; for any two intermediate chassis monitoring positions, based on the test chassis status information of the physical model corresponding to the new energy vehicle chassis under each fault mode, determine the single-mode correlation parameters of the two intermediate chassis monitoring positions corresponding to each fault mode, and based on the single-mode correlation parameters of the two intermediate chassis monitoring positions corresponding to each fault mode, determine the multi-mode correlation parameters of the two intermediate chassis monitoring positions; group the multiple intermediate chassis monitoring positions to determine multiple intermediate chassis monitoring position groups; for each intermediate chassis monitoring position group, based on the multi-mode correlation parameters of any two of the intermediate chassis monitoring positions included in the intermediate chassis monitoring position group, screen the multiple intermediate chassis monitoring positions included in the intermediate chassis monitoring position group to determine the multiple target chassis monitoring positions corresponding to the intermediate chassis monitoring position group and the monitoring feature type corresponding to each target chassis monitoring position; based on the multiple target chassis monitoring positions corresponding to each intermediate chassis monitoring position group and the monitoring feature type corresponding to each target chassis monitoring position, determine the optimal chassis monitoring scheme.

[0009] Further, grouping the multiple intermediate chassis monitoring positions to determine multiple intermediate chassis monitoring position groups includes: based on the abnormal parameters corresponding to each initial chassis monitoring position for each fault mode, determine the abnormal parameters corresponding to each intermediate chassis monitoring position for each fault mode; for any two intermediate chassis monitoring positions, calculate the abnormal similarity of the two intermediate chassis monitoring positions based on the abnormal parameters corresponding to each intermediate chassis monitoring position for each fault mode; through the K-means algorithm, cluster the multiple intermediate chassis monitoring positions based on the abnormal similarity of any two intermediate chassis monitoring positions to determine multiple intermediate chassis monitoring position groups.

[0010] Furthermore, based on the test chassis status information of the new energy vehicle chassis corresponding to the physical model under multiple fault modes, a chassis fault diagnosis knowledge graph is established, including: determining the abnormal parameters corresponding to each target chassis monitoring position for each fault mode according to the abnormal parameters corresponding to each initial chassis monitoring position for each fault mode; for each target chassis monitoring position, determining the fault modes associated with the target chassis monitoring position according to the abnormal parameters corresponding to the target chassis monitoring position for each fault mode; for each fault mode, determining the target chassis monitoring positions associated with the fault mode according to the fault modes associated with each target chassis monitoring position; for each fault mode, extracting the abnormal features corresponding to the fault mode according to the target chassis monitoring positions associated with the fault mode and the test chassis status information of the new energy vehicle chassis corresponding to the physical model under the fault mode; establishing a chassis fault diagnosis knowledge graph according to the association relationship between the target chassis monitoring position and the fault mode and the abnormal features corresponding to the fault mode, wherein the chassis fault diagnosis knowledge graph includes target chassis monitoring position nodes, fault mode nodes and abnormal feature nodes.

[0011] Furthermore, extracting the abnormal features corresponding to the fault mode according to the target chassis monitoring positions associated with the fault mode and the test chassis status information of the new energy vehicle chassis corresponding to the physical model under the fault mode includes: for each target chassis monitoring position associated with the fault mode, extracting the test data corresponding to the target chassis monitoring position from the test chassis status information of the new energy vehicle chassis corresponding to the physical model under the fault mode, performing variational mode decomposition on the test data, and extracting the position abnormal features corresponding to the target chassis monitoring position; generating the abnormal features corresponding to the fault mode according to the position abnormal features corresponding to each target chassis monitoring position associated with the fault mode.

[0012] Furthermore, generating the fault diagnosis information of the new energy vehicle chassis according to the real-time chassis status information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph includes: determining the abnormal target chassis monitoring position according to the real-time chassis status information of the new energy vehicle chassis; determining the candidate fault modes according to the chassis fault diagnosis knowledge graph and the abnormal target chassis monitoring position; for each candidate fault mode, determining the real-time abnormal features of the new energy vehicle chassis corresponding to the candidate fault mode according to the real-time chassis status information of the new energy vehicle chassis and the target chassis monitoring positions associated with the candidate fault mode, and determining the matching degree of the candidate fault mode according to the real-time abnormal features of the new energy vehicle chassis corresponding to the candidate fault mode and the abnormal features corresponding to the fault mode; generating the fault diagnosis information of the new energy vehicle chassis according to the matching degree of each candidate fault mode.

[0013] Further, based on the real-time chassis status information of the new energy vehicle chassis and the target chassis monitoring positions associated with the candidate fault modes, determine the real-time abnormal features of the new energy vehicle chassis corresponding to the candidate fault modes. Based on the real-time abnormal features of the new energy vehicle chassis corresponding to the candidate fault modes and the abnormal features corresponding to the fault modes, determine the matching degree of the candidate fault modes, including: extracting the real-time data corresponding to the target chassis monitoring positions from the real-time chassis status information of the new energy vehicle chassis according to the target chassis monitoring positions associated with the candidate fault modes, performing variational mode decomposition on the test data, and extracting the position real-time features corresponding to the target chassis monitoring positions; generating the real-time features corresponding to the candidate fault modes according to the position real-time features corresponding to each target chassis monitoring position associated with the candidate fault modes; and determining the matching degree of the candidate fault modes according to the real-time features and abnormal features corresponding to the candidate fault modes.

[0014] The present invention provides an intelligent diagnostic system for new energy vehicle chassis faults based on a knowledge graph, which applies the above-mentioned intelligent diagnostic method for new energy vehicle chassis faults based on a knowledge graph, and includes: a data acquisition module for acquiring a three-dimensional model of the new energy vehicle chassis; a scheme optimization module for determining an initial chassis monitoring scheme according to the three-dimensional model of the new energy vehicle chassis, wherein the initial chassis monitoring scheme includes a plurality of initial chassis monitoring positions and the monitoring feature types corresponding to each initial chassis monitoring position; the scheme optimization module is further configured to acquire the physical model corresponding to the new energy vehicle chassis; the scheme optimization module is further configured to install a first chassis monitoring component on the physical model corresponding to the new energy vehicle chassis according to the initial chassis monitoring scheme, wherein the first chassis monitoring component includes a plurality of sensors installed according to the initial chassis monitoring scheme; the scheme optimization module is further configured to obtain the chassis state information of the physical model corresponding to the new energy vehicle chassis in a plurality of fault modes through the first chassis monitoring component; the scheme optimization module is further configured to determine an optimal chassis monitoring scheme according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in a plurality of fault modes, wherein the optimal chassis monitoring scheme includes a plurality of target chassis monitoring positions and the monitoring feature types corresponding to each target chassis monitoring position; a graph construction module for constructing a chassis fault diagnosis knowledge graph according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in a plurality of fault modes; a state monitoring module for installing a second chassis monitoring component on the new energy vehicle chassis according to the optimal chassis monitoring scheme, wherein the second chassis monitoring component includes a plurality of sensors installed according to the optimal chassis monitoring scheme; the state monitoring module is further configured to obtain the real-time chassis state information of the new energy vehicle chassis through the second chassis monitoring component; a fault diagnosis module for generating fault diagnosis information of the new energy vehicle chassis according to the real-time chassis state information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph.

[0015] Compared with the prior art, the intelligent diagnostic method and system for new energy vehicle chassis faults based on a knowledge graph provided by the present invention at least have the following beneficial effects:

[0016] 1. By formulating the initial chassis monitoring plan and the optimal chassis monitoring plan, it is possible to specifically monitor the key positions of the new energy vehicle chassis, reduce ineffective monitoring, and improve the diagnostic efficiency. By utilizing the chassis fault diagnosis knowledge graph, it is possible to quickly match fault characteristics, provide accurate fault diagnosis information, and reduce the probabilities of misdiagnosis and missed diagnosis. The intelligent diagnosis method can promptly detect chassis faults, avoid greater losses caused by the deterioration of faults, and reduce the maintenance cost. Through precise monitoring and diagnosis, the maintenance plan can be optimized, unnecessary maintenance operations can be reduced, and the maintenance cost can be further lowered. By continuously monitoring the chassis status in real time, potential safety hazards can be promptly detected and warned, enhancing the driving safety of new energy vehicles. Accurate fault diagnosis information helps to promptly repair chassis faults, ensure that new energy vehicles operate in good condition, and reduce the accident risk. The intelligent diagnosis method can quickly and accurately provide users with chassis fault information, reduce the waiting time for users, and improve user satisfaction. By integrating advanced technologies such as knowledge graphs and the Internet of Things, it promotes technological innovation and industrial upgrading in the field of new energy vehicles. The application of intelligent diagnosis technology helps to enhance the overall performance and market competitiveness of new energy vehicles, and promotes the sustainable development of the new energy vehicle industry.

[0017] 2. By establishing a finite element model of the new energy vehicle chassis, precise stress analysis, heat conduction analysis, and electromagnetic field analysis can be carried out, thereby determining more accurate initial chassis monitoring positions. The application of the finite element model makes the determination of monitoring positions more scientific and reasonable, reduces blindness, and improves the diagnostic accuracy. By analyzing the test chassis status information of the new energy vehicle chassis under multiple fault modes, the abnormal parameters corresponding to different fault modes for each initial chassis monitoring position can be determined. Further, through the analysis of multi-mode correlation parameters, more critical target chassis monitoring positions can be screened out, making the diagnosis more efficient and accurate. By using the K-means algorithm to cluster and group the intermediate chassis monitoring positions, similar monitoring positions can be grouped together according to the anomaly similarity for subsequent analysis and processing. Intelligent screening and grouping reduce redundant information and improve the efficiency and accuracy of diagnosis. By integrating advanced technologies such as finite element analysis, knowledge graphs, and the K-means algorithm, it promotes technological innovation and industrial upgrading in the field of new energy vehicles.

[0018] 3. By establishing a knowledge graph for chassis fault diagnosis, the target chassis monitoring positions, fault modes, and abnormal characteristics are associated, making the diagnosis process more systematic and organized. The introduction of the knowledge graph enables fault diagnosis to be based on a large amount of historical data and experience, improving the accuracy and reliability of diagnosis. Through technologies such as variational mode decomposition, the position abnormal characteristics of the target chassis monitoring positions are extracted from the test data, making the abnormal characteristics more accurate and prominent. The precise extraction of abnormal characteristics helps to quickly locate faults and improves the efficiency of diagnosis. By comparing the real-time abnormal characteristics of the candidate fault modes of the new energy vehicle chassis with the abnormal characteristics corresponding to the fault modes, the matching degree of the candidate fault modes is determined, thus achieving fast and accurate fault identification. Since the knowledge graph contains the association relationship between the target chassis monitoring positions and fault modes and the abnormal characteristics corresponding to the fault modes, the fault position and type can be accurately located, reducing the possibility of false alarms and missed alarms. Through the double verification of the knowledge graph and abnormal characteristics, the diagnosis results are more stable and reliable, reducing diagnosis errors caused by single factors. It can handle multiple fault modes and make dynamic adjustments according to the real-time chassis status information, so it has strong adaptability. The knowledge graph can be continuously updated and improved with the addition of new data, enabling the diagnosis system to continuously maintain a high performance level. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described in the form of exemplary embodiments, and these exemplary embodiments will be described in detail through the drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where:

[0020] Figure 1 is a schematic flowchart of an intelligent fault diagnosis method for a new energy vehicle chassis based on a knowledge graph according to some embodiments of this specification;

[0021] Figure 2 is a schematic flowchart of determining an optimal chassis monitoring scheme according to some embodiments of this specification;

[0022] Figure 3 is a schematic diagram of a knowledge graph for chassis fault diagnosis according to some embodiments of this specification;

[0023] Figure 4 is a schematic diagram of modules of an intelligent fault diagnosis system for a new energy vehicle chassis based on a knowledge graph according to some embodiments of this specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] To more clearly illustrate the technical solutions of the embodiments of this specification, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some examples or embodiments of this specification. For those of ordinary skill in the art, without creative efforts, this specification can also be applied to other similar scenarios based on these drawings. Unless obvious from the language context or otherwise stated, the same reference numerals in the drawings represent the same structure or operation.

[0025] Figure 1 is a schematic flow chart of an intelligent diagnostic method for new energy vehicle chassis faults based on a knowledge graph as shown in Figure 1 shown, the intelligent diagnostic method for new energy vehicle chassis faults based on a knowledge graph may include the following steps.

[0026] Step S110, obtain a three-dimensional model of the new energy vehicle chassis.

[0027] Step S120, determine an initial chassis monitoring plan according to the three-dimensional model of the new energy vehicle chassis.

[0028] Among them, the initial chassis monitoring plan includes multiple initial chassis monitoring positions and the monitoring feature types corresponding to each initial chassis monitoring position. For example, temperature features are collected at the initial chassis monitoring position A1, vibration features are collected at the initial chassis monitoring position A2, etc.

[0029] In some embodiments, step S120 specifically includes:

[0030] Establish a finite element model of the new energy vehicle chassis according to the three-dimensional model of the new energy vehicle chassis;

[0031] Determine an initial chassis monitoring plan according to the finite element model of the new energy vehicle chassis.

[0032] Specifically, the finite element model of the new energy vehicle chassis can be established through the following process:

[0033] First, simplify and geometrically clean the three-dimensional model of the new energy vehicle chassis, removing unnecessary details and redundant information to facilitate better mesh generation.

[0034] Then, determine the mesh size according to the size and accuracy requirements of the three-dimensional model of the new energy vehicle chassis, and use quadrilateral or triangular meshes to divide the three-dimensional model of the new energy vehicle chassis.

[0035] In the finite element model of the new energy vehicle chassis, assign real material properties to each component of the chassis, such as density, Poisson's ratio, elastic modulus, etc. These properties will directly affect the calculation results and accuracy of the model.

[0036] Set reasonable boundary conditions for the finite element model of the new energy vehicle chassis. For example, the displacements of certain nodes can be fixed or certain loads and constraint conditions can be applied to simulate the stress conditions of the chassis during actual operation.

[0037] After completing the mesh generation, check and adjust the mesh quality. This includes checking issues such as the continuity, integrity, and element shape of the mesh to ensure that there are no errors or abnormalities in the finite element model of the new energy vehicle chassis during the calculation process.

[0038] In some embodiments, according to the finite element model of the new energy vehicle chassis, determine an initial chassis monitoring plan, including:

[0039] Based on the finite element model of the new energy vehicle chassis, conduct a stress analysis to determine the initial vibration monitoring positions;

[0040] Based on the finite element model of the new energy vehicle chassis, conduct a heat conduction analysis to determine the initial temperature monitoring positions;

[0041] Determine candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions. By way of example only, the candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions can be determined through manual or big data analysis techniques;

[0042] Based on the finite element model of the new energy vehicle chassis, conduct an electromagnetic field analysis to determine the initial current monitoring positions, initial voltage monitoring positions, and initial sound monitoring positions from the candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions. Among them, the multiple initial chassis monitoring positions at least include the initial vibration monitoring positions, initial temperature monitoring positions, initial current monitoring positions, initial voltage monitoring positions, and initial sound monitoring positions.

[0043] Specifically, through the stress analysis of the finite element model of the new energy vehicle chassis, the stress distribution of the new energy vehicle chassis under various working conditions can be understood. According to the stress analysis results, the stress concentration areas and potential dangerous points can be determined, thereby determining the initial vibration monitoring positions. These positions are usually areas on the new energy vehicle chassis that are easily affected by vibration and prone to problems.

[0044] During the operation of the new energy vehicle chassis, certain heat will be generated, so heat conduction analysis is required to understand the temperature distribution of the new energy vehicle chassis. Through the heat conduction analysis of the finite element model of the new energy vehicle chassis, the high-temperature areas and potential thermal failure points on the new energy vehicle chassis can be determined, thereby determining the initial temperature monitoring positions. These positions are usually areas on the new energy vehicle chassis that are prone to heat accumulation and sensitive to temperature.

[0045] For new energy vehicles, electromagnetic fields are an important consideration. By performing electromagnetic field analysis on the finite element model of the new energy vehicle chassis, the electromagnetic field distribution on the new energy vehicle chassis can be understood, and candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions with strong electromagnetic field interference can be excluded. The remaining candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions are used as the initial current monitoring position, the initial voltage monitoring position, and the initial sound monitoring position.

[0046] Step 130: Obtain the physical model corresponding to the new energy vehicle chassis.

[0047] Specifically, the physical model corresponding to the new energy vehicle chassis can be a model that is consistent with the structure, dimensions, etc. of the new energy vehicle chassis.

[0048] Step 140: Install the first chassis monitoring component on the physical model corresponding to the new energy vehicle chassis according to the initial chassis monitoring plan.

[0049] Among them, the first chassis monitoring component has multiple sensors installed according to the initial chassis monitoring plan. Specifically, corresponding sensors can be installed at the corresponding positions on the physical model corresponding to the new energy vehicle chassis according to multiple initial chassis monitoring positions and the monitoring feature types corresponding to each initial chassis monitoring position. The first chassis monitoring component can include sensors set at each initial chassis monitoring position.

[0050] Step 150: Obtain the chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes through the first chassis monitoring component.

[0051] Only as an example, multiple fault modes can include transmission faults (such as transmission fluid leakage, transmission abnormal noise, difficult shifting, etc.), steering system faults (such as loose fixing nut between the steering wheel and the steering shaft, excessive meshing clearance between the driving and driven parts of the steering gear, bent and deformed steering shaft, lack of oil in the steering gear, deformation of the front axle or frame, different degrees of wear or inconsistent tire pressures of the front wheels, etc.), braking system faults (such as the position of the brake pipe leakage point, low brake fluid level, too large or too small brake clearance, uneven brake clearances on both sides of the brake, contact area difference, out-of-round brake drum, too smooth brake friction surface or foreign objects inserted, rust of the cable or outer sleeve, broken or fallen traction spring, etc.), driving system faults (such as uneven tire wear, suspension deformation, shock absorber failure, suspension deformation, abnormal shock absorber, etc.), etc.

[0052] Step 160: Determine the optimal chassis monitoring plan according to the test chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes.

[0053] Among them, the optimal chassis monitoring solution includes multiple target chassis monitoring positions and the monitoring feature types corresponding to each target chassis monitoring position.

[0054] Figure 2 It is a schematic flowchart of determining the optimal chassis monitoring solution shown in some embodiments of this specification. As Figure 2 shown, in some embodiments, step 160 specifically includes:

[0055] For each failure mode, according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in the failure mode, determine the abnormal parameters corresponding to each initial chassis monitoring position for the failure mode;

[0056] According to the abnormal parameters corresponding to each initial chassis monitoring position for each failure mode, screen multiple initial chassis monitoring positions to determine multiple intermediate chassis monitoring positions;

[0057] For any two intermediate chassis monitoring positions, according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in each failure mode, determine the single-mode correlation parameters corresponding to the two intermediate chassis monitoring positions for each failure mode, and according to the single-mode correlation parameters corresponding to the two intermediate chassis monitoring positions for each failure mode, determine the multi-mode correlation parameters of the two intermediate chassis monitoring positions;

[0058] Group multiple intermediate chassis monitoring positions to determine multiple intermediate chassis monitoring position groups;

[0059] For each intermediate chassis monitoring position group, according to the multi-mode correlation parameters of any two intermediate chassis monitoring positions included in the intermediate chassis monitoring position group, screen the multiple intermediate chassis monitoring positions included in the intermediate chassis monitoring position group to determine the multiple target chassis monitoring positions corresponding to the intermediate chassis monitoring position group and the monitoring feature types corresponding to each target chassis monitoring position;

[0060] According to the multiple target chassis monitoring positions corresponding to each intermediate chassis monitoring position group and the monitoring feature types corresponding to each target chassis monitoring position, determine the optimal chassis monitoring solution.

[0061] Specifically, the test data of each initial chassis monitoring position can be obtained in advance under the non-fault state. According to the test data of the initial chassis monitoring position in the non-fault state, the position normal characteristics corresponding to the initial chassis monitoring position in the non-fault state are determined. Specifically, the variational mode decomposition can be performed on the test data of the initial chassis monitoring position in the non-fault state to extract the position normal characteristics corresponding to the initial chassis monitoring position, where the position normal characteristics may include the eigenvectors of each intrinsic mode component obtained after performing variational mode decomposition on the test data of the initial chassis monitoring position (for example, modal center frequency, amplitude, phase, bandwidth, energy, etc.).

[0062] For each fault mode, according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in the fault mode, the test data of each initial chassis monitoring position corresponding to the fault mode is determined. According to the test data of the initial chassis monitoring position corresponding to the fault mode, the position abnormal characteristics of the initial chassis monitoring position corresponding to the fault mode are determined. Specifically, the variational mode decomposition can be performed on the test data of the initial chassis monitoring position corresponding to the fault mode to extract the position abnormal characteristics of the initial chassis monitoring position corresponding to the fault mode, where the position abnormal characteristics of the initial chassis monitoring position corresponding to the fault mode may include the eigenvectors of each intrinsic mode component obtained after performing variational mode decomposition on the test data of the initial chassis monitoring position corresponding to the fault mode (for example, modal center frequency, amplitude, phase, bandwidth, energy, etc.).

[0063] The abnormal parameter of the initial chassis monitoring position corresponding to the fault mode can be calculated according to the following formula:

[0064]

[0065] where η (i,g) is the abnormal parameter of the i-th initial chassis monitoring position corresponding to the g-th fault mode, ψ1 is a preset parameter, ψ1 > 0, F ((normal,k),i,g) is the eigenvector of the k-th intrinsic mode component obtained after performing variational mode decomposition on the test data of the i-th initial chassis monitoring position corresponding to the g-th fault mode, F ((fault,k),i,g) is the eigenvector of the k-th intrinsic mode component obtained after performing variational mode decomposition on the test data corresponding to the fault mode of the test data of the j-th initial chassis monitoring position corresponding to the g-th fault mode, cos(F ((normal,k),i,g) ,F ((fault,k),i,g)) is the cosine similarity between the eigenvector of the k-th intrinsic mode component obtained by variational mode decomposition of the test data corresponding to the i-th initial chassis monitoring position and the g-th fault mode and the eigenvector of the k-th intrinsic mode component obtained by variational mode decomposition of the test data corresponding to the g-th fault mode of the test data corresponding to the j-th initial chassis monitoring position. K is the total number of preset intrinsic mode components.

[0066] For each initial chassis monitoring position, the mean value and standard deviation of the abnormal parameters corresponding to each fault mode of the initial chassis monitoring position can be calculated. The initial chassis monitoring position with the mean value of the abnormal parameters greater than the abnormal parameter mean threshold, the standard deviation of the abnormal parameters less than the maximum abnormal parameter standard deviation threshold and greater than the minimum abnormal parameter standard deviation threshold is used as the intermediate chassis monitoring position.

[0067] The single-mode correlation parameter corresponding to the fault mode of two intermediate chassis monitoring positions can be calculated according to the following formula:

[0068]

[0069] where γ ((i,j),g) is the single-mode correlation parameter corresponding to the i-th intermediate chassis monitoring position and the j-th intermediate chassis monitoring position and the g-th fault mode, V ((i,t),g) is the data value at the t-th test time point corresponding to the g-th fault mode of the i-th intermediate chassis monitoring position, V (j,t),g) is the data value at the t-th test time point corresponding to the g-th fault mode of the j-th intermediate chassis monitoring position, and T is the total number of sampled test time points.

[0070] For any two intermediate chassis monitoring positions, the single-mode correlation parameters corresponding to each fault mode of the two intermediate chassis monitoring positions can be weighted and summed to obtain the multi-mode correlation parameter of the two intermediate chassis monitoring positions.

[0071] In some embodiments, grouping multiple intermediate chassis monitoring positions to determine multiple groups of intermediate chassis monitoring positions includes:

[0072] Determine the abnormal parameters corresponding to each fault mode of each intermediate chassis monitoring position according to the abnormal parameters corresponding to each fault mode of each initial chassis monitoring position;

[0073] For any two intermediate chassis monitoring positions, calculate the abnormal similarity of the two intermediate chassis monitoring positions according to the abnormal parameters corresponding to each fault mode of each intermediate chassis monitoring position;

[0074] By using the K-means algorithm, multiple intermediate chassis monitoring positions are clustered according to the anomaly similarity between any two intermediate chassis monitoring positions to determine multiple groups of intermediate chassis monitoring positions.

[0075] Specifically, the anomaly similarity between two intermediate chassis monitoring positions can be calculated according to the following formula:

[0076]

[0077] where S (i,j) is the anomaly similarity between the i-th intermediate chassis monitoring position and the j-th intermediate chassis monitoring position, η (j,g) is the anomaly parameter corresponding to the g-th fault mode of the j-th initial chassis monitoring position, and ψ2 is a preset parameter, and ψ2 is greater than 0.

[0078] The multiple intermediate chassis monitoring positions included in the intermediate chassis monitoring position group can be screened according to the multi-mode correlation parameters between any two intermediate chassis monitoring positions included in the intermediate chassis monitoring position group through the following process:

[0079] S11. For any one intermediate chassis monitoring position included in the intermediate chassis monitoring position group, calculate the mean value of the multi-mode correlation parameters of the multi-mode correlation parameters between this intermediate chassis monitoring position and any other intermediate chassis monitoring position included in this intermediate chassis monitoring position group;

[0080] S12. Judge whether the mean value of the multi-mode correlation parameters of any one intermediate chassis monitoring position included in the intermediate chassis monitoring position group is greater than the multi-mode correlation parameter mean threshold. If so, execute S15. If not, execute S13;

[0081] S13. Judge whether the number of intermediate chassis monitoring positions included in the intermediate chassis monitoring position group is greater than the minimum threshold of the number of intermediate chassis monitoring positions. If so, execute S14. If not, complete the screening;

[0082] S14. Screen out the intermediate chassis monitoring position with the minimum mean value of the multi-mode correlation parameters from the intermediate chassis monitoring position group, update the intermediate chassis monitoring position group, and execute S11;

[0083] S15. Judge whether the number of intermediate chassis monitoring positions included in the intermediate chassis monitoring position group is less than the maximum threshold of the number of intermediate chassis monitoring positions. If so, complete the screening. If not, execute S14.

[0084] Step 170, establish a chassis fault diagnosis knowledge graph according to the test chassis state information of the new energy vehicle chassis corresponding to the physical model under multiple fault modes.

[0085] In some embodiments, step 160 specifically includes:

[0086] Determine the abnormal parameters corresponding to each target chassis monitoring position for each fault mode according to the abnormal parameters corresponding to each initial chassis monitoring position for each fault mode;

[0087] For each target chassis monitoring position, determine the fault modes associated with the target chassis monitoring position according to the abnormal parameters corresponding to the target chassis monitoring position for each fault mode. For example, regard the fault modes with abnormal parameters greater than the abnormal parameter threshold as the fault modes associated with the target chassis monitoring position;

[0088] For each fault mode, determine the target chassis monitoring positions associated with the fault mode according to the fault modes associated with each target chassis monitoring position. For example, if the fault modes associated with the target chassis monitoring position A1 include fault modes B1 and B2, the fault modes associated with the target chassis monitoring position A2 include fault modes B3 and B2, and the fault modes associated with the target chassis monitoring position A3 include fault modes B1, B2, and B4, then the target chassis monitoring positions associated with the fault mode B2 may include the target chassis monitoring position A1, the target chassis monitoring position A2, and the target chassis monitoring position A3;

[0089] For each fault mode, extract the abnormal features corresponding to the fault mode according to the target chassis monitoring positions associated with the fault mode and the test chassis state information of the physical model corresponding to the new energy vehicle chassis in the fault mode;

[0090] Establish a chassis fault diagnosis knowledge graph according to the association relationship between the target chassis monitoring position and the fault mode and the abnormal features corresponding to the fault mode, where Figure 3 is a schematic diagram of the chassis fault diagnosis knowledge graph shown in some embodiments of this specification, such as Figure 3 shown, the chassis fault diagnosis knowledge graph includes target chassis monitoring position nodes, fault mode nodes, and abnormal feature nodes.

[0091] In some embodiments, extracting the abnormal features corresponding to the fault mode according to the target chassis monitoring positions associated with the fault mode and the test chassis state information of the physical model corresponding to the new energy vehicle chassis in the fault mode includes:

[0092] For each target chassis monitoring position associated with the fault mode, extract the test data corresponding to the target chassis monitoring position from the test chassis state information of the physical model corresponding to the new energy vehicle chassis in the fault mode, perform variational mode decomposition on the test data, and extract the position abnormal features corresponding to the target chassis monitoring position;

[0093] Generate the abnormal features corresponding to the fault mode according to the position abnormal features corresponding to each target chassis monitoring position associated with the fault mode.

[0094] Specifically, the abnormal features corresponding to the failure mode may include the position abnormal features corresponding to each target chassis monitoring position associated with the failure mode. For more descriptions on variational mode decomposition of the test data and extraction of the position abnormal features corresponding to the target chassis monitoring position, reference can be made to the foregoing, which will not be elaborated here.

[0095] Step 180: Install a second chassis monitoring component on the new energy vehicle chassis according to the optimal chassis monitoring scheme.

[0096] Among them, the second chassis monitoring component is a plurality of sensors installed according to the optimal chassis monitoring scheme.

[0097] Specifically, according to the multiple target chassis monitoring positions included in the optimal chassis monitoring scheme and the monitoring feature types corresponding to each target chassis monitoring position, corresponding sensors can be installed at the corresponding positions on the new energy vehicle chassis.

[0098] Step 190: Obtain the real-time chassis status information of the new energy vehicle chassis through the second chassis monitoring component.

[0099] Among them, the real-time chassis status information of the new energy vehicle chassis may include the data collected by the sensors installed at each target chassis monitoring position at multiple consecutive time points.

[0100] Step 200: Generate the fault diagnosis information of the new energy vehicle chassis according to the real-time chassis status information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph.

[0101] In some embodiments, step 190 specifically includes:

[0102] Determine the abnormal target chassis monitoring position according to the real-time chassis status information of the new energy vehicle chassis;

[0103] Determine the candidate failure modes according to the chassis fault diagnosis knowledge graph and the abnormal target chassis monitoring position;

[0104] For each candidate failure mode, determine the real-time abnormal features of the new energy vehicle chassis corresponding to the candidate failure mode according to the real-time chassis status information of the new energy vehicle chassis and the target chassis monitoring position associated with the candidate failure mode, and determine the matching degree of the candidate failure mode according to the real-time abnormal features of the new energy vehicle chassis corresponding to the candidate failure mode and the abnormal features corresponding to the failure mode;

[0105] Generate the fault diagnosis information of the new energy vehicle chassis according to the matching degree of each candidate failure mode.

[0106] Specifically, for each fault mode, according to the chassis fault diagnosis knowledge graph, determine the target chassis monitoring positions associated with the fault mode. Based on the determined abnormal target chassis monitoring positions, judge whether each target chassis monitoring position associated with the fault mode is an abnormal target chassis monitoring position. If so, determine that the fault mode is a candidate fault mode.

[0107] In some embodiments, according to the real-time chassis state information of the new energy vehicle chassis and the target chassis monitoring positions associated with the candidate fault mode, determine the real-time abnormal characteristics of the new energy vehicle chassis corresponding to the candidate fault mode. According to the real-time abnormal characteristics of the new energy vehicle chassis corresponding to the candidate fault mode and the abnormal characteristics corresponding to the fault mode, determine the matching degree of the candidate fault mode, including:

[0108] According to the target chassis monitoring positions associated with the candidate fault mode, extract the real-time data corresponding to the target chassis monitoring positions from the real-time chassis state information of the new energy vehicle chassis. Perform variational mode decomposition on the test data to extract the position real-time characteristics corresponding to the target chassis monitoring positions. For more descriptions, reference can be made to the foregoing, which will not be elaborated here;

[0109] Generate the real-time characteristics corresponding to the candidate fault mode according to the position real-time characteristics corresponding to each target chassis monitoring position associated with the candidate fault mode;

[0110] Determine the matching degree of the candidate fault mode according to the real-time characteristics and abnormal characteristics corresponding to the candidate fault mode.

[0111] Specifically, the matching degree of the candidate fault mode can be determined according to the following formula:

[0112]

[0113] where M g is the matching degree of the g-th candidate fault mode, F ((fault,k),h,g) is the eigenvector of the k-th intrinsic mode component in the abnormal characteristics corresponding to the h-th target chassis monitoring position associated with the g-th candidate fault mode, F ((current,k),h,g) is the eigenvector of the k-th intrinsic mode component in the real-time abnormal characteristics corresponding to the h-th target chassis monitoring position associated with the g-th candidate fault mode, cos(F ((fault,k),h,g) , F ((current,k),h,g) ) is the cosine similarity between the eigenvector of the k-th intrinsic mode component in the abnormal characteristics corresponding to the h-th target chassis monitoring position associated with the g-th candidate fault mode and the eigenvector of the k-th intrinsic mode component in the real-time abnormal characteristics corresponding to the h-th target chassis monitoring position associated with the g-th candidate fault mode, H is the total number of target chassis monitoring positions associated with the g-th candidate fault mode, and ψ3 is a preset parameter, and ψ3 is greater than 0.

[0114] The fault diagnosis information of the new energy vehicle chassis can be generated by the fault diagnosis model according to the matching degree of each candidate fault mode, wherein the fault diagnosis model can be a convolutional neural network model.

[0115] Figure 4 It is a schematic diagram of the modules of the intelligent fault diagnosis system for the new energy vehicle chassis based on the knowledge graph shown in some embodiments of this specification. As Figure 4 shown, the intelligent fault diagnosis system for the new energy vehicle chassis based on the knowledge graph can include a data acquisition module, a scheme optimization module, a graph construction module, a state monitoring module and a fault diagnosis module.

[0116] The data acquisition module is used to acquire the three-dimensional model of the new energy vehicle chassis;

[0117] The scheme optimization module is used to determine an initial chassis monitoring scheme according to the three-dimensional model of the new energy vehicle chassis, wherein the initial chassis monitoring scheme includes a plurality of initial chassis monitoring positions and the monitoring feature types corresponding to each initial chassis monitoring position;

[0118] The scheme optimization module is also used to acquire the physical model corresponding to the new energy vehicle chassis;

[0119] The scheme optimization module is also used to install a first chassis monitoring component on the physical model corresponding to the new energy vehicle chassis according to the initial chassis monitoring scheme, wherein the first chassis monitoring component is a plurality of sensors installed according to the initial chassis monitoring scheme;

[0120] The scheme optimization module is also used to acquire the chassis state information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes through the first chassis monitoring component;

[0121] The scheme optimization module is also used to determine an optimal chassis monitoring scheme according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes, wherein the optimal chassis monitoring scheme includes a plurality of target chassis monitoring positions and the monitoring feature types corresponding to each target chassis monitoring position;

[0122] [[ID=!28]]The graph construction module is used to establish a chassis fault diagnosis knowledge graph according to the test chassis state information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes;

[0123] The state monitoring module is used to install a second chassis monitoring component on the new energy vehicle chassis according to the optimal chassis monitoring scheme, wherein the second chassis monitoring component is a plurality of sensors installed according to the optimal chassis monitoring scheme;

[0124] The state monitoring module is also used to acquire the real-time chassis state information of the new energy vehicle chassis through the second chassis monitoring component;

[0125] A fault diagnosis module, which is used to generate fault diagnosis information of the new energy vehicle chassis according to the real-time chassis state information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph.

[0126] The intelligent fault diagnosis system for the new energy vehicle chassis based on the knowledge graph can be used to execute the intelligent fault diagnosis method for the new energy vehicle chassis based on the knowledge graph, which will not be elaborated here.

[0127] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other deformations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be regarded as consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. An intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph, characterized in that, Including: Obtain the three-dimensional model of the new energy vehicle chassis; According to the three-dimensional model of the new energy vehicle chassis, determine the initial chassis monitoring plan, where the initial chassis monitoring plan includes multiple initial chassis monitoring positions and the monitoring feature types corresponding to each initial chassis monitoring position; Obtain the physical model corresponding to the new energy vehicle chassis; According to the initial chassis monitoring plan, install the first chassis monitoring component on the physical model corresponding to the new energy vehicle chassis, where the first chassis monitoring component includes multiple sensors installed according to the initial chassis monitoring plan; Obtain the chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes through the first chassis monitoring component; According to the test chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes, determine the optimal chassis monitoring plan, where the optimal chassis monitoring plan includes multiple target chassis monitoring positions and the monitoring feature types corresponding to each target chassis monitoring position; According to the test chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes, establish a chassis fault diagnosis knowledge graph; According to the optimal chassis monitoring plan, install the second chassis monitoring component on the new energy vehicle chassis, where the second chassis monitoring component includes multiple sensors installed according to the optimal chassis monitoring plan; Obtain the real-time chassis status information of the new energy vehicle chassis through the second chassis monitoring component; Generate the fault diagnosis information of the new energy vehicle chassis according to the real-time chassis status information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph; According to the test chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes, establish a chassis fault diagnosis knowledge graph, including: According to the abnormal parameters corresponding to each initial chassis monitoring position in each fault mode, determine the abnormal parameters corresponding to each target chassis monitoring position in each fault mode; For each target chassis monitoring position, according to the abnormal parameters corresponding to the target chassis monitoring position in each fault mode, determine the fault mode associated with the target chassis monitoring position; For each fault mode, according to the fault mode associated with each target chassis monitoring position, determine the target chassis monitoring positions associated with the fault mode; For each fault mode, according to the target chassis monitoring positions associated with the fault mode and the test chassis status information of the physical model corresponding to the new energy vehicle chassis in the fault mode, extract the abnormal features corresponding to the fault mode; According to the association relationship between the target chassis monitoring positions and the fault modes and the abnormal features corresponding to the fault modes, establish a chassis fault diagnosis knowledge graph, where the chassis fault diagnosis knowledge graph includes target chassis monitoring position nodes, fault mode nodes, and abnormal feature nodes.

2. The intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph according to claim 1, characterized in that, According to the three-dimensional model of the new energy vehicle chassis, determine the initial chassis monitoring plan, including: According to the three-dimensional model of the new energy vehicle chassis, establish a finite element model of the new energy vehicle chassis; According to the finite element model of the new energy vehicle chassis, determine the initial chassis monitoring plan.

3. The intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph according to claim 2, wherein, Based on the finite element model of the new energy vehicle chassis, determine the initial chassis monitoring plan, including: Based on the finite element model of the new energy vehicle chassis, conduct stress analysis to determine the initial vibration monitoring positions; Based on the finite element model of the new energy vehicle chassis, conduct heat conduction analysis to determine the initial temperature monitoring positions; Determine the candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions; Based on the finite element model of the new energy vehicle chassis, conduct electromagnetic field analysis to determine the initial current monitoring positions, initial voltage monitoring positions, and initial sound monitoring positions from the candidate current monitoring positions, candidate voltage monitoring positions, and candidate sound monitoring positions. Among them, the multiple initial chassis monitoring positions at least include the initial vibration monitoring positions, initial temperature monitoring positions, initial current monitoring positions, initial voltage monitoring positions, and initial sound monitoring positions.

4. The intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph according to claim 3, characterized in that, Based on the test chassis state information of the physical model corresponding to the new energy vehicle chassis under multiple fault modes, determine the optimal chassis monitoring plan, including: For each fault mode, based on the test chassis state information of the physical model corresponding to the new energy vehicle chassis under the fault mode, determine the abnormal parameters corresponding to each initial chassis monitoring position for the fault mode; Based on the abnormal parameters corresponding to each initial chassis monitoring position for each fault mode, screen the multiple initial chassis monitoring positions to determine multiple intermediate chassis monitoring positions; For any two intermediate chassis monitoring positions, based on the test chassis state information of the physical model corresponding to the new energy vehicle chassis under each fault mode, determine the single-mode correlation parameters corresponding to the two intermediate chassis monitoring positions for each fault mode. Based on the single-mode correlation parameters corresponding to the two intermediate chassis monitoring positions for each fault mode, determine the multi-mode correlation parameters of the two intermediate chassis monitoring positions; Group the multiple intermediate chassis monitoring positions to determine multiple intermediate chassis monitoring position groups; For each intermediate chassis monitoring position group, based on the multi-mode correlation parameters of any two of the intermediate chassis monitoring positions included in the intermediate chassis monitoring position group, screen the multiple intermediate chassis monitoring positions included in the intermediate chassis monitoring position group to determine the multiple target chassis monitoring positions corresponding to the intermediate chassis monitoring position group and the monitoring feature type corresponding to each target chassis monitoring position; Based on the multiple target chassis monitoring positions corresponding to each intermediate chassis monitoring position group and the monitoring feature type corresponding to each target chassis monitoring position, determine the optimal chassis monitoring plan.

5. The intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph according to claim 4, characterized in that Group the multiple intermediate chassis monitoring positions to determine multiple intermediate chassis monitoring position groups, including: Based on the abnormal parameters corresponding to each initial chassis monitoring position for each fault mode, determine the abnormal parameters corresponding to each intermediate chassis monitoring position for each fault mode; For any two intermediate chassis monitoring positions, calculate the abnormal similarity of the two intermediate chassis monitoring positions based on the abnormal parameters corresponding to each intermediate chassis monitoring position for each fault mode; By using the K-means algorithm, cluster the multiple intermediate chassis monitoring positions according to the anomaly similarity between any two intermediate chassis monitoring positions, and determine multiple groups of intermediate chassis monitoring positions.

6. The intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph according to claim 1, wherein According to the target chassis monitoring positions associated with the fault mode and the test chassis state information of the new energy vehicle chassis corresponding to the physical model under the fault mode, extract the anomaly features corresponding to the fault mode, including: For each target chassis monitoring position associated with the fault mode, extract the test data corresponding to the target chassis monitoring position from the test chassis state information of the new energy vehicle chassis corresponding to the physical model under the fault mode, perform variational mode decomposition on the test data, and extract the position anomaly features corresponding to the target chassis monitoring position; Generate the anomaly features corresponding to the fault mode according to the position anomaly features corresponding to each target chassis monitoring position associated with the fault mode.

7. The intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph according to claim 6, characterized in that, Generate the fault diagnosis information of the new energy vehicle chassis according to the real-time chassis state information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph, including: Determine the abnormal target chassis monitoring positions according to the real-time chassis state information of the new energy vehicle chassis. Determine the candidate fault modes according to the chassis fault diagnosis knowledge graph and the abnormal target chassis monitoring positions. For each candidate fault mode, determine the real-time anomaly features of the new energy vehicle chassis corresponding to the candidate fault mode according to the real-time chassis state information of the new energy vehicle chassis and the target chassis monitoring positions associated with the candidate fault mode, and determine the matching degree of the candidate fault mode according to the real-time anomaly features of the new energy vehicle chassis corresponding to the candidate fault mode and the anomaly features corresponding to the fault mode; Generate the fault diagnosis information of the new energy vehicle chassis according to the matching degree of each candidate fault mode.

8. The intelligent diagnosis method for new energy vehicle chassis faults based on a knowledge graph according to claim 7, characterized in that, Determine the real-time anomaly features of the new energy vehicle chassis corresponding to the candidate fault mode according to the real-time chassis state information of the new energy vehicle chassis and the target chassis monitoring positions associated with the candidate fault mode, and determine the matching degree of the candidate fault mode according to the real-time anomaly features of the new energy vehicle chassis corresponding to the candidate fault mode and the anomaly features corresponding to the fault mode, including: According to the target chassis monitoring positions associated with the candidate fault mode, extract the real-time data corresponding to the target chassis monitoring position from the real-time chassis state information of the new energy vehicle chassis, perform variational mode decomposition on the test data, and extract the position real-time features corresponding to the target chassis monitoring position; Generate the real-time features corresponding to the candidate fault mode according to the position real-time features corresponding to each target chassis monitoring position associated with the candidate fault mode. Determine the matching degree of the candidate fault mode according to the real-time features and anomaly features corresponding to the candidate fault mode.

9. An intelligent diagnosis system for new energy vehicle chassis faults based on a knowledge graph, characterized in that, Applying the knowledge graph-based intelligent fault diagnosis method for new energy vehicle chassis according to any one of claims 1-8, including: A data acquisition module for acquiring a three-dimensional model of the new energy vehicle chassis. A solution optimization module for determining an initial chassis monitoring solution based on a 3D model of a new energy vehicle chassis, wherein the initial chassis monitoring solution includes multiple initial chassis monitoring positions and the monitoring feature types corresponding to each initial chassis monitoring position; The solution optimization module is further configured to obtain a physical model corresponding to the new energy vehicle chassis; The solution optimization module is further configured to install a first chassis monitoring component on the physical model corresponding to the new energy vehicle chassis according to the initial chassis monitoring solution, wherein the first chassis monitoring component includes multiple sensors installed according to the initial chassis monitoring solution; The solution optimization module is further configured to obtain the chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes through the first chassis monitoring component; The solution optimization module is further configured to determine an optimal chassis monitoring solution based on the test chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes, wherein the optimal chassis monitoring solution includes multiple target chassis monitoring positions and the monitoring feature types corresponding to each target chassis monitoring position; A knowledge graph building module for building a chassis fault diagnosis knowledge graph based on the test chassis status information of the physical model corresponding to the new energy vehicle chassis in multiple fault modes; A status monitoring module for installing a second chassis monitoring component on the new energy vehicle chassis according to the optimal chassis monitoring solution, wherein the second chassis monitoring component includes multiple sensors installed according to the optimal chassis monitoring solution; The status monitoring module is further configured to obtain the real-time chassis status information of the new energy vehicle chassis through the second chassis monitoring component; A fault diagnosis module for generating fault diagnosis information of the new energy vehicle chassis based on the real-time chassis status information of the new energy vehicle chassis and the chassis fault diagnosis knowledge graph.

Images