Sensing technology-combined sightseeing vehicle chassis fault early warning system

Abstract

The invention provides a sightseeing vehicle chassis fault early warning system combined with a sensing technology, and relates to the technical field of fault early warning, and the system comprises a real-time monitoring module which is used for obtaining multi-dimensional operation state data; the database retrieval module is used for retrieving to obtain a chassis fault sample data set; the clustering analysis module is used for performing clustering analysis to obtain a multi-dimensional chassis fault sample data set; the model construction module is used for constructing a chassis fault positioning model; the multi-dimensional feature extraction module is used for performing multi-dimensional feature extraction and outputting a vibration fault positioning result, a stress fault positioning result, a heat fault positioning result and an atypical fault positioning result through a chassis fault positioning model; and the fault early warning signal generation module is used for generating a fault early warning signal. According to the invention, the technical problem that the precision of fault identification is insufficient because the fault diagnosis of the chassis of the sightseeing vehicle mostly depends on a single sensor for monitoring and the multi-dimensional operation state of the chassis is difficult to comprehensively perceive in the prior art is solved.

Description

Technical Field

[0001] The invention relates to the technical field of fault warning, and in particular to a chassis fault warning system for a sightseeing vehicle combined with a sensing technology. Background Art

[0002] In tourist attractions, sightseeing buses are prone to chassis failures due to the complex operating environment (such as rugged terrain, slippery roads, etc.). However, existing technologies mostly focus on a single data dimension, such as vibration data or stress data, and lack the ability to comprehensively monitor and analyze multi-dimensional operating conditions, which makes it difficult to fully reflect the actual operating conditions of the chassis and may miss key signs of failures. In addition, traditional methods rely on experience or simple algorithms and are unable to accurately locate faults in different areas of the chassis, resulting in vague diagnostic results, low maintenance efficiency, and easy waste of resources or maintenance delays. Summary of the invention

[0003] This application provides a sightseeing car chassis fault warning system combined with sensing technology, aiming to solve the technical problem that the existing technology relies on a single sensor for monitoring the fault diagnosis of the sightseeing car chassis, which makes it difficult to fully perceive the multi-dimensional operating status of the chassis, resulting in insufficient accuracy in fault identification.

[0004] The first aspect disclosed in the present application provides a chassis fault warning system for sightseeing vehicles combined with sensing technology, the system comprising: a real-time monitoring module, used to perform real-time monitoring of the multi-dimensional operating status of a target sightseeing vehicle through a sensor network, and obtain multi-dimensional operating status data, wherein the sensor network is pre-deployed on the chassis of the target sightseeing vehicle, and the multi-dimensional operating status data includes vibration distribution data, stress distribution data and heat distribution data; a database retrieval module, used to obtain the chassis basic data of the target sightseeing vehicle, and use the chassis basic data as a retrieval constraint to retrieve the chassis fault database of the sightseeing vehicle to obtain a chassis fault sample data set; a cluster analysis module, used to perform cluster analysis on the chassis fault sample data set based on the multi-dimensional operating status to obtain a multi-dimensional chassis fault sample data set; a model building module, After extracting features from the multi-dimensional chassis fault sample data set, a chassis fault location model is constructed, wherein the chassis fault location model includes a vibration fault location unit, a stress fault location unit, a heat fault location unit, and an atypical fault location unit; a multi-dimensional feature extraction module is used to extract multi-dimensional features from the multi-dimensional operating status data to obtain a vibration distribution feature matrix, a stress distribution feature matrix, and a heat distribution feature matrix, and output a vibration fault location result, a stress fault location result, a heat fault location result, and an atypical fault location result through the chassis fault location model; a fault warning signal generation module is used to generate a corresponding fault warning signal when any of the vibration fault location result, the stress fault location result, the heat fault location result, and the atypical fault location result is not empty.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages: The sensor network monitors the chassis vibration, stress, heat distribution and other operating conditions in real time, covers the multi-dimensional physical characteristics of the chassis, provides accurate operating data, lays the foundation for fault diagnosis and early warning, and reduces the risk of missing potential faults; retrieves the fault database based on the basic data of the sightseeing car chassis, and generates a chassis fault sample data set containing different dimensions such as vibration, stress, and heat through cluster analysis. The accurate classification of cluster samples reduces the misjudgment rate of the model and ensures the specific analysis capability for different fault types; integrates vibration fault, stress fault, heat fault and atypical fault units into the chassis fault location model, each unit analyzes different fault types in a targeted manner, and accurately outputs the location results of various types of faults. The addition of atypical fault units enables the system to handle unknown or unclassified faults and improves the robustness of the model; through feature matching of the vibration distribution feature matrix, the stress distribution feature matrix and the heat distribution feature matrix, when the matching degree exceeds the threshold, the corresponding fault location result is output in real time. When any fault location result is not empty, a fault warning signal is generated to remind relevant personnel to perform maintenance in time.

[0006] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 A schematic diagram of the structure of a chassis fault warning system for a sightseeing vehicle combined with sensing technology provided in an embodiment of the present application.

[0008] Figure 2 A schematic diagram of the implementation process of the cluster analysis module in the sightseeing vehicle chassis fault warning system combined with sensing technology provided in an embodiment of the present application.

[0009] Explanation of the reference numerals: real-time monitoring module 10 , database retrieval module 20 , cluster analysis module 30 , model building module 40 , multi-dimensional feature extraction module 50 , fault warning signal generation module 60 . DETAILED DESCRIPTION

[0010] The embodiment of the present application solves the technical problem that the fault diagnosis of the chassis of the sightseeing vehicle in the prior art mostly relies on a single sensor for monitoring, making it difficult to fully perceive the multi-dimensional operating status of the chassis, resulting in insufficient accuracy in fault identification, by providing a chassis fault warning system for the sightseeing vehicle combined with sensing technology.

[0011] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0012] like Figure 1 As shown, the embodiment of the present application provides a chassis fault warning system for sightseeing vehicles combined with sensing technology, the system comprising: The real-time monitoring module 10 is used to perform real-time monitoring of the multi-dimensional operating status of the target sightseeing vehicle through a sensor network to obtain multi-dimensional operating status data, wherein the sensor network is pre-deployed on the chassis of the target sightseeing vehicle, and the multi-dimensional operating status data includes vibration distribution data, stress distribution data and heat distribution data.

[0013] According to the structural characteristics of the chassis, the sensor network is deployed in key positions, such as support points, welding points, and areas bearing large loads, for real-time monitoring of the multi-dimensional operating status of the chassis of the target sightseeing vehicle. The sensor network includes: vibration sensors, which are used to detect the vibration amplitude, frequency, and distribution generated by the chassis during operation; strain sensors, such as strain gauges, which are used to monitor the stress changes in various parts of the chassis structure. The collected data include tensile stress, compressive stress, and shear stress; temperature sensors, such as thermocouples or infrared temperature sensors, are used to record the heat distribution on the surface and inside of the chassis.

[0014] The database search module 20 is used to obtain the chassis basic data of the target sightseeing vehicle, and use the chassis basic data as a search constraint to search the chassis fault database of the sightseeing vehicle to obtain a chassis fault sample data set.

[0015] Chassis basic data is the key information that describes the chassis status of the target sightseeing vehicle and its operating conditions, including chassis structure parameters, chassis operating performance parameters, sightseeing vehicle operating environment parameters, and sightseeing vehicle operating record parameters. The chassis basic data is used as a retrieval constraint to ensure that the retrieved chassis fault sample data is consistent with the actual situation of the target sightseeing vehicle. Database indexes or attribute-based fuzzy matching algorithms are used for retrieval. The retrieval results constitute a chassis fault sample data set, which contains fault sample data under similar chassis structures and operating conditions in historical records.

[0016] The cluster analysis module 30 is used to perform cluster analysis on the chassis fault sample data set based on the multi-dimensional operating status to obtain a multi-dimensional chassis fault sample data set.

[0017] The multi-dimensional operating state includes vibration state, stress state, and thermal state. The chassis fault sample data are grouped according to vibration, stress, and thermal characteristics to form a multi-dimensional chassis fault sample data set to reveal the characteristics and laws of different types of faults, and generate vibration fault sample data sets, stress fault sample data sets, thermal fault sample data sets, and atypical fault sample data sets. Among them, the atypical fault sample data sets are samples that cannot be classified into other categories.

[0018] The model building module 40 is used to construct a chassis fault location model after extracting features from the multi-dimensional chassis fault sample data set, wherein the chassis fault location model includes a vibration fault location unit, a stress fault location unit, a thermal fault location unit, and an atypical fault location unit.

[0019] Feature extraction refers to extracting data that can characterize chassis fault characteristics from a multi-dimensional chassis fault sample data set and converting it into a feature vector that can be used for fault location. Based on the extracted features, vibration fault, stress fault, thermal fault, and atypical fault location units are constructed respectively. Each location unit is responsible for locating the fault according to the corresponding fault characteristics. The output of each location unit can be a list containing multiple fault locations and a fault severity score for each location. This provides accurate data support for subsequent fault warnings and maintenance decisions, ensuring that chassis faults can be discovered and located in a timely manner so that appropriate repair measures can be taken.

[0020] The multi-dimensional feature extraction module 50 is used to perform multi-dimensional feature extraction on the multi-dimensional operating status data to obtain a vibration distribution feature matrix, a stress distribution feature matrix and a heat distribution feature matrix, and output a vibration fault location result, a stress fault location result, a heat fault location result and an atypical fault location result through the chassis fault location model.

[0021] Feature extraction is performed on multi-dimensional operating status data, including vibration, stress and heat distribution data, and the eigenvalues ​​of each time window are constructed into a matrix. Each column of the matrix represents the feature of a time window, and each row represents different features. The constructed vibration distribution feature matrix, stress distribution feature matrix and heat distribution feature matrix are used, and the chassis fault location model, including vibration fault location, stress fault location, heat fault location and atypical fault location, is used to perform fault analysis. The model compares the known fault sample data for prediction and outputs the vibration fault location results, stress fault location results, heat fault location results and atypical fault location results, including the fault location and fault severity.

[0022] The fault warning signal generating module 60 is used to generate a corresponding fault warning signal when any of the vibration fault location result, the stress fault location result, the heat fault location result, and the atypical fault location result is not empty.

[0023] Check the vibration fault location results, stress fault location results, heat fault location results and atypical fault location results. If any result is not empty, that is, a fault is detected, then generate a warning signal related to the fault type and location. Specifically, if the vibration fault location unit outputs a fault, generate a vibration abnormality warning; if the stress fault location unit outputs a fault, generate a stress abnormality warning; if the heat fault location unit outputs a fault, generate a heat abnormality warning; if the atypical fault location unit outputs a fault, generate an atypical fault warning. The warning signal contains the fault type, fault location, fault severity and recommended measures, etc., which are used to help operators deal with chassis faults in a timely manner to avoid the expansion and impact of faults.

[0024] Furthermore, the chassis basic data includes chassis structure parameters, chassis operation performance parameters, sightseeing vehicle operation environment parameters, and sightseeing vehicle operation record parameters.

[0025] The basic chassis data includes chassis structure parameters, chassis operation performance parameters, sightseeing car operation environment parameters, and sightseeing car operation record parameters. The chassis structure parameters describe the physical structure of the chassis and the layout of each component, including chassis size, component connection method, axle type and layout, material information, etc.; chassis operation performance parameters are used to describe the performance characteristics of the chassis during actual operation. These parameters reflect the working status of the chassis, including chassis load, vibration amplitude, suspension system performance, steering response, etc.; the sightseeing car operation environment directly affects the working conditions of the chassis, including external environmental factors and road conditions, etc. These parameters have an important impact on the operating status of the chassis, including road condition information, climate conditions, driving mode, vehicle speed, etc.; the sightseeing car operation record parameters record the specific operation data of the vehicle in different time periods and under different road conditions, including the vehicle's usage time, mileage, fault history, maintenance records, etc., which can reveal the aging and wear of chassis components.

[0026] Furthermore, if Figure 2 As shown, the cluster analysis of the chassis fault sample data set based on the multi-dimensional operating state to obtain the multi-dimensional chassis fault sample data set includes: Based on the multi-dimensional operating state, a plurality of cluster labels are constructed, wherein the plurality of cluster labels include a vibration fault label, a stress fault label, and a thermal fault label; based on the plurality of cluster labels, the chassis fault sample data set is clustered to obtain a vibration fault sample data set, a stress fault sample data set, a thermal fault sample data set, and an atypical fault sample data set to form the multi-dimensional chassis fault sample data set.

[0027] Multi-dimensional operating status data includes vibration status, stress status and thermal status, and multiple cluster labels are constructed to identify different types of faults. Each cluster label corresponds to a specific fault type, which can help to more accurately classify and locate faults. Among them, the vibration fault label is used to identify chassis faults caused by abnormal vibration, which may be related to the suspension system, axle, etc.; the stress fault label is used to identify chassis faults that may be caused by stress concentration, which is usually related to damage or deformation of structural components; the thermal fault label is used to identify faults caused by thermal abnormalities, which are common in overheating problems of components such as electrical systems and drive systems; atypical fault labels are used to mark abnormal conditions that do not conform to conventional failure modes, which may involve factors such as material fatigue and external environmental influences.

[0028] The clustering labels constructed above are used to perform cluster analysis on the chassis fault sample data set. Specifically, features are extracted from the original data such as vibration, stress, and heat to form a feature matrix. Common feature extraction methods include Fourier transform, wavelet transform, etc. Clustering is performed based on the characteristics of the data, and the data is divided into corresponding labels to obtain a vibration fault sample data set, a stress fault sample data set, a heat fault sample data set, and an atypical fault sample data set. Among them, the vibration fault sample data set contains sample data related to vibration faults, and these samples may show characteristics of excessive or irregular vibration amplitudes; the stress fault sample data set contains sample data related to stress faults, and these samples usually show characteristics of excessive stress, which may cause deformation or damage to the chassis; the heat fault sample data set contains sample data related to heat faults, which usually show local excessive temperatures, which may be related to friction or electrical overload; the atypical fault sample data set contains sample data that does not conform to conventional fault modes.

[0029] Furthermore, clustering the chassis fault sample data set based on the multiple cluster labels includes: Extracting first chassis fault sample data; when the first chassis fault sample data satisfies any one of the vibration fault label, the stress fault label, and the thermal fault label, adding the first chassis fault sample data to a corresponding fault sample data set; when the first chassis fault sample data satisfies any two or three of the vibration fault label, the stress fault label, and the thermal fault label, copying the first chassis fault sample data, and adding the copying results to the corresponding fault sample data sets respectively; when the first chassis fault sample data does not satisfy any one of the vibration fault label, the stress fault label, and the thermal fault label, adding the first chassis fault sample data to a newly created atypical fault sample data set.

[0030] A sample is randomly selected from the chassis fault sample data set as the first chassis fault sample data, and the features of the first chassis fault sample data are analyzed to determine whether it satisfies any fault label.

[0031] When it is determined that the first chassis fault sample data meets and only meets the condition of a certain fault label, namely, vibration fault or stress fault or thermal fault, the sample data is added to the corresponding fault sample data set. For example, if the sample meets the vibration fault label, it is added to the vibration fault sample data set.

[0032] If the first chassis fault sample data satisfies any two labels at the same time, or satisfies all three labels at the same time, the first chassis fault sample data is completely copied to generate multiple copies, each copy is used for a different fault sample data set, and the copy is added to the corresponding fault sample data set.

[0033] If the first chassis fault sample data does not meet any of the vibration fault, stress fault or thermal fault labels, that is, it does not meet the known label standards at all, then the sample data is considered to belong to an abnormal or unclassified fault type, and an atypical fault sample data set is created for storing unclassified samples. The first chassis fault sample data set is directly added to the atypical fault sample data set to facilitate subsequent in-depth research on unclassified fault types.

[0034] Furthermore, after extracting features from the multi-dimensional chassis fault sample data set, a chassis fault location model is constructed, including: Perform feature extraction on the vibration fault sample data set to obtain a vibration fault sample feature library, perform supervised training to obtain the vibration fault location unit; and so on, obtain the stress fault location unit and the heat fault location unit in sequence; perform feature extraction on the atypical fault sample data set to obtain an atypical fault sample feature library, and embed the atypical fault sample feature library into the atypical fault location unit.

[0035] The vibration fault sample data set is cleaned to remove noise and invalid data. The dimensional differences are eliminated through data standardization or normalization to ensure the consistency of feature dimensions. The key features are extracted from the vibration fault samples using time domain, frequency domain and time-frequency domain analysis methods to form a vibration fault sample feature library for subsequent modeling.

[0036] Input the vibration fault sample feature library into the supervised learning algorithm, select the appropriate model, such as support vector machine, random forest or convolutional neural network, use the labeled data to train the model, adjust the hyperparameters to optimize the model performance, generate the vibration fault location unit, and achieve accurate location of the vibration fault. Use the same method to build the stress fault location unit and the thermal fault location unit to handle a single type of fault respectively.

[0037] Analyze the atypical fault sample data set, extract atypical features, including abnormal features and mixed features, build an atypical fault sample feature library, and directly embed the atypical fault sample feature library into the atypical fault location unit. This unit relies on the atypical fault sample feature library and focuses on processing abnormal situations that are not recognized by standard labels. Integrate the above unit into the chassis fault location model to achieve full coverage diagnosis of different fault types.

[0038] Furthermore, the chassis fault location model is used to output vibration fault location results, stress fault location results, heat fault location results, and atypical fault location results, including: According to the vibration fault location unit, feature matching is performed on the vibration distribution feature matrix, and features with a matching degree greater than a matching degree threshold are output as vibration fault location results, wherein when any matching degree is less than the matching degree threshold, the vibration fault location result is empty; and so on, the stress fault location result, the thermal fault location result, and the atypical fault location result are obtained respectively.

[0039] The vibration distribution feature matrix is ​​matched item by item with the pre-trained vibration fault sample feature library in the vibration fault location unit. The matching degree of each feature vector is evaluated using algorithms such as cosine similarity and Euclidean distance. A matching score is generated for each feature match. The matching range is usually [0,1]. A matching threshold, such as 0.85, is set to determine whether the match is valid. If the matching degree of the feature is greater than or equal to the matching threshold, the match is considered successful and the feature is marked as a vibration fault feature; if the matching degree of all features is less than the matching threshold, the vibration fault location result is empty. Generate the vibration fault location result based on the matched feature position and type.

[0040] By analogy, the stress fault location results, thermal fault location results, and atypical fault location results are obtained respectively.

[0041] Furthermore, the system also includes a fault superposition judgment module, including: When any two or more of the vibration fault location results, stress fault location results, heat fault location results, and atypical fault location results are not empty, a fault superposition judgment is performed according to the fault position to obtain a first superposition fault; and a first superposition fault warning signal is generated based on the first superposition fault.

[0042] Determine whether any two or more of the vibration fault location results, stress fault location results, thermal fault location results, and atypical fault location results are not empty. For multiple non-empty results, extract the fault location of each fault type, and determine the intersection of the fault locations of each fault location result, that is, the overlapping area where multiple fault types appear. If the intersection of the fault locations is not empty, it is considered that there is a superimposed fault at that location. In this case, record the fault types involved in all overlapping locations, and generate a first superimposed fault, such as a vibration-stress superimposed fault, based on the overlapping results of the fault locations, and generate a first superimposed fault warning signal, including the fault location, superimposed type, and fault severity assessment.

[0043] Through the fault superposition judgment method, it is possible to effectively identify the potential threats to vehicle operation safety when multiple faults occur at the same time. The generated first superposition fault warning signal provides clear warning information and operation suggestions based on precise fault analysis and location overlap, greatly improving fault response efficiency and safety.

[0044] Furthermore, the system also includes a sensitive fault learning module, including: The chassis fault diagnosis record of the target sightseeing vehicle is retrieved; based on the chassis fault diagnosis record, sensitive fault learning is performed on the chassis fault location model to complete model optimization.

[0045] The chassis fault data of the target sightseeing vehicle in the past is obtained to provide real reference data for model optimization. The data sources include maintenance record database, on-board diagnostic system, etc. The chassis fault diagnosis records obtained include fault type, fault location, fault time, etc.

[0046] Sensitive fault learning is used to enhance the model’s sensitivity to common historical faults, and improve accuracy and robustness. Specifically, the chassis fault diagnosis records are aligned with the model’s existing multi-dimensional operating status data, common features are extracted, and representative historical fault samples are selected and labeled as sensitive fault labels. Reinforcement learning or transfer learning techniques are used to improve the model’s ability to respond to known fault modes. According to the category distribution of sensitive fault samples, the model’s loss function weights are adjusted to enhance the ability to identify sensitive faults. The optimized model has the ability to respond quickly to common fault modes.

[0047] In summary, the sightseeing vehicle chassis fault warning system combined with sensing technology provided in the embodiment of the present application has the following technical effects: The sensor network monitors the chassis vibration, stress, heat distribution and other operating conditions in real time, covers the multi-dimensional physical characteristics of the chassis, provides accurate operating data, lays the foundation for fault diagnosis and early warning, and reduces the risk of missing potential faults; retrieves the fault database based on the basic data of the sightseeing car chassis, and generates a chassis fault sample data set containing different dimensions such as vibration, stress, and heat through cluster analysis. The accurate classification of cluster samples reduces the misjudgment rate of the model and ensures the specific analysis capability for different fault types; integrates vibration fault, stress fault, heat fault and atypical fault units into the chassis fault location model, each unit analyzes different fault types in a targeted manner, and accurately outputs the location results of various types of faults. The addition of atypical fault units enables the system to handle unknown or unclassified faults and improves the robustness of the model; through feature matching of the vibration distribution feature matrix, the stress distribution feature matrix and the heat distribution feature matrix, when the matching degree exceeds the threshold, the corresponding fault location result is output in real time. When any fault location result is not empty, a fault warning signal is generated to remind relevant personnel to perform maintenance in time.

[0048] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The sightseeing car chassis fault warning system combined with sensor technology is characterized by: The system comprises: A real-time monitoring module is used to perform real-time monitoring of the multi-dimensional operating status of the target sightseeing vehicle through a sensor network to obtain multi-dimensional operating status data, wherein the sensor network is pre-deployed on the chassis of the target sightseeing vehicle, and the multi-dimensional operating status data includes vibration distribution data, stress distribution data and heat distribution data; A database retrieval module is used to obtain chassis basic data of a target sightseeing vehicle, and to retrieve a chassis fault database of a sightseeing vehicle using the chassis basic data as a retrieval constraint to obtain a chassis fault sample data set; A cluster analysis module, configured to perform cluster analysis on the chassis fault sample data set based on the multi-dimensional operating status to obtain a multi-dimensional chassis fault sample data set; A model building module, used to construct a chassis fault location model after extracting features from the multi-dimensional chassis fault sample data set, wherein the chassis fault location model includes a vibration fault location unit, a stress fault location unit, a heat fault location unit, and an atypical fault location unit; A multi-dimensional feature extraction module is used to perform multi-dimensional feature extraction on the multi-dimensional operating status data to obtain a vibration distribution feature matrix, a stress distribution feature matrix and a heat distribution feature matrix, and output a vibration fault location result, a stress fault location result, a heat fault location result and an atypical fault location result through the chassis fault location model; The fault warning signal generating module is used to generate a corresponding fault warning signal when any of the vibration fault location result, the stress fault location result, the heat fault location result, and the atypical fault location result is not empty.

2. The sightseeing vehicle chassis fault warning system combined with sensing technology as claimed in claim 1, characterized in that: The chassis basic data includes chassis structure parameters, chassis operation performance parameters, sightseeing vehicle operation environment parameters, and sightseeing vehicle operation record parameters.

3. The sightseeing vehicle chassis fault warning system combined with sensing technology as claimed in claim 1, characterized in that: The performing cluster analysis on the chassis fault sample data set based on the multi-dimensional operating state to obtain a multi-dimensional chassis fault sample data set includes: Based on the multi-dimensional operating state, construct a plurality of cluster labels, wherein the plurality of cluster labels include a vibration fault label, a stress fault label, and a thermal fault label; Based on the multiple clustering labels, the chassis fault sample data set is clustered to obtain a vibration fault sample data set, a stress fault sample data set, a thermal fault sample data set, and an atypical fault sample data set to form the multi-dimensional chassis fault sample data set.

4. The sightseeing vehicle chassis fault warning system combined with sensing technology as claimed in claim 3 is characterized in that: Clustering the chassis fault sample data set based on the multiple clustering labels includes: Extracting first chassis fault sample data; When the first chassis fault sample data satisfies any one of the vibration fault label, the stress fault label, and the thermal fault label, adding the first chassis fault sample data to a corresponding fault sample data set; When the first chassis fault sample data satisfies any two or three of the vibration fault label, the stress fault label, and the heat fault label, the first chassis fault sample data is copied, and the copy results are added to the corresponding fault sample data sets respectively; When the first chassis fault sample data does not satisfy any one of the vibration fault label, the stress fault label, and the thermal fault label, the first chassis fault sample data is added to a newly created atypical fault sample data set.

5. The sightseeing vehicle chassis fault warning system combined with sensing technology as claimed in claim 3, characterized in that: After extracting features from the multi-dimensional chassis fault sample data set, a chassis fault location model is constructed, including: Extracting features from the vibration fault sample data set to obtain a vibration fault sample feature library, and performing supervised training to obtain the vibration fault location unit; By analogy, the stress fault location unit and the heat fault location unit are obtained in sequence; Feature extraction is performed on the atypical fault sample data set to obtain an atypical fault sample feature library, and the atypical fault sample feature library is embedded in the atypical fault locating unit.

6. The sightseeing vehicle chassis fault warning system combined with sensing technology as claimed in claim 5, characterized in that: The chassis fault location model is used to output vibration fault location results, stress fault location results, heat fault location results, and atypical fault location results, including: According to the vibration fault location unit, feature matching is performed on the vibration distribution feature matrix, and features with a matching degree greater than a matching degree threshold are output as vibration fault location results, wherein when any matching degree is less than the matching degree threshold, the vibration fault location result is empty; By analogy, the stress fault location result, the heat fault location result, and the atypical fault location result are obtained respectively.

7. The sightseeing vehicle chassis fault warning system combined with sensing technology as claimed in claim 1, characterized in that: The system also includes a fault superposition judgment module, including: When any two or more of the vibration fault location results, stress fault location results, heat fault location results, and atypical fault location results are not empty, a fault superposition judgment is performed according to the fault position to obtain a first superposition fault; A first superimposed fault warning signal is generated based on the first superimposed fault.

8. The sightseeing vehicle chassis fault warning system combined with sensing technology as claimed in claim 1, characterized in that: The system also includes a sensitive fault learning module, including: Retrieving chassis fault diagnosis records of the target sightseeing vehicle; Based on the chassis fault diagnosis record, sensitive fault learning is performed on the chassis fault location model to complete model optimization.

Images