Rapid diagnosis method and system for fluid-structure interaction abnormal vibration noise of high-speed train

Abstract

The invention provides a rapid diagnosis method and system for fluid-structure interaction abnormal vibration noise of a high-speed train. Real-time sound pressure level spectrum data of a monitoring point in the high-speed train to be diagnosed are acquired; comparing the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to judge whether abnormal noise exists or not; if it is judged that abnormal noise exists, according to the mapping relation between the in-vehicle noise monitoring points and the multiple aerodynamic sound source areas outside the vehicle, the noise spectrum, corresponding to the real-time sound pressure level spectrum data, of the aerodynamic sound source areas outside the vehicle is calculated through inversion; and carrying out contrastive analysis on the noise spectrum of the aerodynamic sound source area outside the vehicle calculated by inversion and the abnormal noise spectrum database so as to locate the specific sound source position of the abnormal noise and / or determine the abnormal noise mode of the abnormal noise. According to the method, the fault source area outside the train can be inverted and positioned in an extremely short time, the fault diagnosis time is shortened, the maintenance efficiency is improved, and the train operation safety and the riding comfort are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of high-speed train fault diagnosis and noise control technology, and in particular to a high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method and system. BACKGROUND

[0002] With the continuous improvement of the running speed of high-speed trains, especially when the speed exceeds 350 km / h, aerodynamic noise has replaced wheel-rail noise as the main noise source, seriously affecting passenger comfort. At the same time, the interaction between airflow and vehicle structure (i.e. fluid-structure coupling effect) becomes more and more significant under high-speed operation, often inducing local abnormal vibration and noise, such as low-frequency howling or booming sound around certain parts (such as pantograph, air deflector, door, etc.). These abnormal noises not only reduce the comfort of passengers, but also may be a sign of potential structural failure, posing a threat to train safety.

[0003] Currently, the existing technologies for high-speed train noise problems mainly focus on two aspects: one is to passively reduce noise by optimizing the shape of the vehicle body, using sound insulation materials, etc. during the research and design stage; the other is to monitor abnormal noise by arranging a large number of acoustic sensors or vibration sensors inside or outside the vehicle body during the operation and maintenance stage.

[0004] However, these existing technologies have obvious limitations: Insufficient prediction accuracy: Traditional noise prediction models are mostly one-way coupled or purely acoustic simulation, which fails to fully consider the fluid-structure coupling effect at high speed, leading to distortion in the simulation of the turbulent boundary layer on the vehicle body surface and geometry-induced noise, making it difficult to accurately predict the generation of abnormal noise.

[0005] Diagnosis and positioning difficulties: Fault detection methods relying on physical sensors have slow response speed, high deployment cost, and cannot quickly and accurately locate the physical position of the abnormal sound source in complex operating environments. Operation and maintenance personnel can usually only perceive that there is abnormal noise in the vehicle, but it is difficult to quickly determine which specific component outside the vehicle is the source, resulting in low repair efficiency and high maintenance cost. SUMMARY

[0006] The present application provides a high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method and system to solve the problems of existing technologies, such as difficult positioning of abnormal noise of high-speed trains, low diagnosis efficiency, and inability to effectively deal with fluid-structure coupling abnormalities, and to realize rapid diagnosis of high-speed train fluid-structure coupling abnormal vibration noise.

[0007] The high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method of the present application comprises: Obtaining real-time sound pressure level spectrum data of at least one monitoring point in the vehicle of the high-speed train to be diagnosed; comparing the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether abnormal noise exists; If it is determined that abnormal noise exists, a noise spectrum of an aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data is inversely calculated by using a preset mapping relationship between the in-vehicle noise monitoring points and the multiple aerodynamic sound source areas outside the vehicle; and The noise spectrum of the aerodynamic sound source area outside the vehicle inversely calculated is compared and analyzed with the abnormal noise spectrum database to locate a specific sound source position of the abnormal noise and / or determine an abnormal noise mode thereof.

[0008] According to the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the application, before the noise spectrum of the aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data is inversely calculated by using a preset mapping relationship between the in-vehicle noise monitoring points and the multiple aerodynamic sound source areas outside the vehicle, the method further comprises: A full-speed-level fluid-structure coupling aerodynamic noise prediction model is constructed, and a mapping relationship between the in-vehicle noise monitoring points and the multiple aerodynamic sound source areas outside the vehicle is established based on the model.

[0009] According to the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the application, the full-speed-level fluid-structure coupling aerodynamic noise prediction model is constructed, and a mapping relationship between the in-vehicle noise monitoring points and the multiple aerodynamic sound source areas outside the vehicle is established based on the model, specifically comprising: Based on a train geometric model, a fluid simulation model covering a speed range of 200 km / h to 400 km / h is established, and a large eddy simulation or a detached eddy simulation method is used for transient flow field calculation to obtain pressure field and velocity field data of a volume domain around the train; An acoustic calculation model is established, and the pressure field and velocity field data, wheel-rail noise source data and vehicle body vibration data measured by experiments are used as composite excitation sources to simulate and calculate sound pressure level spectrums of each in-vehicle noise monitoring point and each aerodynamic sound source area outside the vehicle under normal operating conditions; Based on the simulated sound pressure level spectrum data, a weight coefficient matrix is determined by a mathematical regression method, and a mapping relationship between the in-vehicle noise monitoring points and the multiple aerodynamic sound source areas outside the vehicle is established based on the weight coefficient matrix.

[0010] According to the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the application, the abnormal noise spectrum database is established by the following steps: Real-measured aerodynamic noise spectrum data inside and outside the high-speed train under multiple operating conditions are collected; A machine learning algorithm is used to analyze the real-measured spectrum data to distinguish normal noise spectrums from abnormal noise spectrums; The abnormal noise spectrum is classified according to the sound source position and spectral characteristics, and a structured abnormal noise spectrum database is formed.

[0011] According to the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the application, the real-time sound pressure level spectrum data is compared with the preset abnormal noise spectrum database to determine whether there is abnormal noise, and specifically includes: The characteristics of the real-time sound pressure level spectrum data are extracted, and compared with the characteristic range of the normal noise spectrum in the abnormal noise spectrum database; And / or, the similarity between the real-time sound pressure level spectrum data and the typical abnormal noise mode stored in the abnormal noise spectrum database is calculated; When the characteristics of the real-time sound pressure level spectrum data deviate from the characteristic range of the normal noise spectrum or the similarity with a certain abnormal noise mode exceeds a preset threshold, it is determined that there is abnormal noise.

[0012] According to the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the application, the method further includes: The low-frequency band with a frequency not higher than 500Hz in the real-time sound pressure level spectrum data is preferentially analyzed to identify specific noise peaks caused by abnormal vibration of the car body structure.

[0013] According to the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the application, the plurality of aerodynamic sound source regions outside the car include at least one of the pantograph, the fender, the antenna and the door region.

[0014] According to the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the application, the method further includes: The specific sound source position of the diagnosed abnormal noise is highlighted in the corresponding local area of the three-dimensional car body model.

[0015] The application also provides a high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis system, which includes: A data acquisition module is configured to acquire real-time sound pressure level spectrum data of at least one monitoring point in a high-speed train to be diagnosed; An abnormality judgment module is configured to compare the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether there is abnormal noise; A sound source inversion module is configured to, if it is determined that there is abnormal noise, use a preset mapping relationship between the in-car noise monitoring points and the plurality of aerodynamic sound source regions outside the car to inversely calculate the noise spectrum of the aerodynamic sound source region outside the car corresponding to the real-time sound pressure level spectrum data; and A positioning and identifying module is configured to compare and analyze the noise spectrum of the vehicle exterior aerodynamic sound source region calculated by inversion with the abnormal noise spectrum database to locate the specific sound source position of the abnormal noise and / or determine the abnormal mode thereof.

[0016] The application further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method according to any one of the above when executing the computer program.

[0017] The application further provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program implements the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method according to any one of the above when executed by a processor.

[0018] The application further provides a computer program product, which includes a computer program, and the computer program implements the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method according to any one of the above when executed by a processor.

[0019] The high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method and system provided by the application can realize reverse diagnosis from inside to outside by establishing a direct mapping relationship between the inside and outside noises, can invert and lock the vehicle exterior fault source region in a very short time once the abnormal noise in the vehicle is detected, greatly shortens the fault diagnosis time, and significantly improves the repair efficiency; can automatically and accurately judge whether the noise is abnormal and identify the fault mode by introducing the abnormal noise database, reduces the dependence on the personal experience of repair personnel, makes the diagnosis result more objective and reliable, can timely find and locate the abnormal vibration noise caused by the loosening of components, structure fatigue and the like, eliminates or reduces the potential structural safety hidden danger, and effectively guarantees the operation safety of the train. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0021] Figure 1 is the flow chart of the high-speed train fluid-solid coupling abnormal vibration noise rapid diagnosis method provided by the present application.

[0022] Figure 2 is the overall process of constructing a full-speed-level aerodynamic noise prediction model and calculating the in-vehicle and out-vehicle noise spectrum provided by the present application.

[0023] Figure 3 is the process of fluid-solid coupling sound characteristic analysis of the key area provided by the present application.

[0024] Figure 4 is the process of in-vehicle and out-vehicle noise analysis considering the fluid-solid coupling effect provided by the present application.

[0025] Figure 5 is the overall process of abnormal noise rapid identification and evaluation provided by the present application.

[0026] Figure 6 is the structural schematic diagram of the high-speed train fluid-solid coupling abnormal vibration noise rapid diagnosis system provided by the present application.

[0027] Figure 7 is the structural schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be described clearly and completely in the following combined with the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.

[0029] The present invention will now be described in detail with reference to the accompanying drawings. The specific operation methods in the method embodiments can also be applied to the device embodiments or system embodiments. In the description of the present invention, unless otherwise stated, "at least one" includes one or more. "Multiple" refers to two or more. For example, at least one of A, B, and C includes: A existing alone, B existing alone, A and B existing simultaneously, A and C existing simultaneously, B and C existing simultaneously, and A, B, and C existing simultaneously. In the present invention, " / " means "or". For example, A / B can mean A or B. "And / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0030] The present invention will now be described in detail with reference to specific embodiments.

[0031] As high-speed trains operate at speeds exceeding 350 km / h, aerodynamic noise becomes increasingly prominent, severely impacting passenger comfort. Localized fluid-structure interaction (FSI) excitation effects frequently lead to abnormal noise phenomena. An investigation of existing technologies and analysis of competitor patents reveals that while some patented technologies address high-speed train noise, most focus on single-faceted noise reduction, such as optimizing vehicle shape and materials to reduce air resistance and noise propagation. However, these methods often require high R&D and production costs and have limited effectiveness. Fault detection often relies on the installation of acoustic sensors. However, these solutions suffer from slow response times, poor adaptability, and high maintenance costs. They cannot effectively address noise issues under varying operating conditions of high-speed trains in real time, nor can they comprehensively consider the impact of fluid-structure interaction anomalies, resulting in unsatisfactory performance in practical applications. These methods not only have slow response times but also fail to promptly identify abnormal noise in the rapidly changing environment of high-speed train operation.

[0032] To address this issue, this invention constructs a full-speed-level aerodynamic noise prediction model based on actual train acoustic measurement data and LES simulation analysis, establishing a rapid data mapping relationship from excitation sources on the car body surface / near-wall surface to a series of noise monitoring points inside and outside the car. For the actual data from these noise monitoring points, the abnormal sound source is located by comparing and analyzing the noise spectrum under normal operating conditions and using the rapid data mapping relationship. This process not only helps to identify potential problems in a timely manner but also provides important basis for train design optimization and safety improvement.

[0033] In some specific embodiments of the present invention, such as Figure 1 As shown, this solution provides a rapid diagnostic method for abnormal vibration and noise caused by fluid-structure interaction in high-speed trains, including: Step 100, acquiring real-time sound pressure level spectrum data of at least one monitoring point in a high-speed train to be diagnosed; Step 200, comparing the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether there is abnormal noise; Step 300, if it is determined that there is abnormal noise, using a preset mapping relationship between the in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle to inversely calculate the noise spectrum of the aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data; and Step 400, comparing and analyzing the inversely calculated noise spectrum of the aerodynamic sound source area outside the vehicle with the abnormal noise spectrum database to locate the specific sound source position of the abnormal noise and / or determine the abnormal noise mode thereof.

[0034] It should be noted that the existing abnormal vibration noise diagnosis scheme has limitations in high-speed train aerodynamic noise prediction and control, cannot accurately simulate the characteristics of the turbulent boundary layer and geometric induced noise on the surface of the vehicle body at high speed, leads to difficulty in noise source identification and suppression, affects passenger comfort, and increases maintenance difficulty.

[0035] Therefore, the present application acquires real-time sound pressure level spectrum data of at least one monitoring point in a high-speed train to be diagnosed, determines whether there is abnormal noise based on the comparison result of the real-time sound pressure level spectrum data and the preset abnormal noise spectrum database. In the case where it is determined that there is abnormal noise, the inverse calculation is performed according to the mapping relationship between the in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle, the noise spectrum of the aerodynamic sound source area outside the vehicle obtained by the inverse calculation is compared and analyzed in depth with the abnormal noise spectrum database again, the aerodynamic sound source area outside the vehicle generating the abnormal noise is accurately located according to the result of the in-depth comparison and analysis, and the abnormal mode of the aerodynamic sound source area outside the vehicle can be further identified. A technical paradigm of reverse diagnosis from inside to outside is established, the dependence on the sensor outside the vehicle is eliminated, and rapid diagnosis of abnormal noise is realized. Moreover, through the above settings of the present application, it can not only be determined whether there is abnormal noise, but also be located to the specific sound source area and the mode thereof can be identified, which provides clear guidance for accurate repair. In addition, the implementation of the present application only needs to be distributed in the vehicle, which greatly reduces the hardware and maintenance costs.

[0036] In some possible embodiments of the present application, before inversely calculating the noise spectrum of the aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data using the preset mapping relationship between the in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle, the method further comprises: building a full-speed fluid-structure coupling aerodynamic noise prediction model, and establishing a preset mapping relationship between the in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle based on the model.

[0037] Specifically, the embodiment provides an implementation of establishing a mapping relationship between an in-vehicle noise monitoring point and a plurality of aerodynamic sound source areas outside the vehicle, a full-speed-stage fluid-structure coupling aerodynamic noise prediction model is constructed, and the mapping relationship is generated based on the model.

[0038] Further, in some possible embodiments of the present application, the full-speed-stage fluid-structure coupling aerodynamic noise prediction model is constructed, and the mapping relationship between the preset in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle is established based on the model, and specifically includes: Based on a train geometric model, a fluid simulation model covering a speed range of 200 km / h to 400 km / h is established, and a large eddy simulation or detached eddy simulation method is used to perform transient flow field calculation to obtain pressure field and velocity field data of a volume domain around the train; An acoustic calculation model is established, the pressure field and velocity field data, the wheel-rail noise source data and the train body vibration data measured through experiments are used as composite excitation sources, and sound pressure level spectrum of each noise monitoring point in the vehicle and each aerodynamic sound source area outside the vehicle under normal operation conditions is obtained through simulation calculation; Based on the sound pressure level spectrum data obtained through the simulation calculation, a weight coefficient matrix is determined through a mathematical regression method, and the mapping relationship between the preset in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle is established based on the weight coefficient matrix.

[0039] Specifically, the embodiment provides another implementation of establishing a mapping relationship between an in-vehicle noise monitoring point and a plurality of aerodynamic sound source areas outside the vehicle, through fluid simulation and acoustic simulation, a high-precision simulation method and a composite excitation source are used, so that the model can very realistically simulate the fluid-structure coupling noise characteristics of the real train, and thus an accurate weight coefficient matrix is calculated, which provides the most fundamental guarantee for the accuracy of the final diagnosis.

[0040] In some possible embodiments of the present application, the abnormal noise spectrum database is established through the following steps: Real measurement spectrum data of in-vehicle and out-vehicle aerodynamic noise of a high-speed train under a plurality of operation conditions are collected; A machine learning algorithm is used to analyze the real measurement spectrum data, so as to distinguish normal noise spectrum from abnormal noise spectrum; The abnormal noise spectrum is classified according to a sound source position and a spectrum characteristic, and a structured abnormal noise spectrum database is formed.

[0041] Specifically, the embodiment provides a construction method of an abnormal noise spectrum database, real and measured in-vehicle and external noise spectrum data are collected, machine learning algorithms are used to automatically analyze the data, the algorithms can automatically distinguish normal and abnormal data, and classify and structure the abnormal data, so that a comprehensive and accurate abnormal noise fault sample library is efficiently and intelligently constructed. The application of machine learning enables the establishment and expansion of the database to be automatically performed, the system has self-learning ability, manual intervention is greatly reduced, the coverage and accuracy of the database are improved, the fault mode recognition is more reliable, and the intelligentization and automation of the diagnosis system are realized.

[0042] Specifically, based on in-vehicle and external noise simulation, a mapping relationship between in-vehicle pneumatic noise and main external pneumatic noise source areas (pantograph, obstacle remover, antenna and door) is established, and a data corresponding relationship between sound pressure level spectrums of different position points in the vehicle and sound pressure level spectrums of external pneumatic sound source areas is given, for example: (1), wherein, is a sound pressure level spectrum column vector of different position points in the vehicle, is a sound pressure level spectrum column vector of different position points in the external pneumatic sound source areas, is a weight coefficient matrix.

[0043] Further, according to the measured data of the pneumatic noise of different position points in the vehicle and the external vehicle, a machine learning method is used to analyze the data, to distinguish normal data from abnormal data, and to classify the mode of abnormal noise generated by the external data, to form an abnormal noise spectrum database.

[0044] In some possible embodiments of the present application, the real-time sound pressure level spectrum data is compared with the preset abnormal noise spectrum database to determine whether there is abnormal noise, specifically including: extracting features of the real-time sound pressure level spectrum data, and comparing the features with a feature range of normal noise spectrum in the abnormal noise spectrum database; and / or, calculating a similarity between the real-time sound pressure level spectrum data and a typical abnormal noise mode stored in the abnormal noise spectrum database; when the features of the real-time sound pressure level spectrum data deviate from the feature range of the normal noise spectrum or the similarity with a certain abnormal noise mode exceeds a preset threshold, it is determined that there is abnormal noise.

[0045] Specifically, the embodiment provides an implementation for judging whether abnormal noise exists, by comparing whether the characteristics of real-time data exceed the normal range, or whether the similarity between the real-time data and a known abnormal pattern is high enough, when any condition is met, it is determined to be abnormal. Through the setting, the preliminary judgment process of abnormal noise is more sensitive and reliable, avoiding false negatives (insensitivity) and false positives (over-sensitivity). Moreover, by combining the two kinds of judgment logic, the system can not only find unknown abnormalities that have never been seen but are obviously abnormal, but also quickly identify known abnormalities that have been learned, greatly enhancing the comprehensive judgment ability of the diagnosis system and improving the sensitivity and robustness of abnormal judgment.

[0046] That is, in a specific application, according to the sound pressure level spectrum data obtained by testing in the vehicle, whether it is abnormal noise data is determined by comparing it with the data of the abnormal noise spectrum database through threshold judgment, if it is determined to be abnormal noise data, according to the corresponding relationship between the sound pressure level spectrum of different positions in the vehicle and the sound pressure level spectrum of the external aerodynamic sound source area , through the analysis of the weight coefficient matrix A, the specific external aerodynamic sound source area is located, and according to the inverse obtained noise sound pressure spectrum data of the external aerodynamic sound source area, the abnormal noise pattern is obtained by comparing and analyzing with the abnormal noise spectrum database.

[0047] In some possible embodiments of the present application, the method further comprises: Preferentially analyzing the low-frequency band with a frequency not higher than 500Hz in the real-time sound pressure level spectrum data to identify specific noise peaks caused by abnormal vibration of the vehicle body structure.

[0048] Specifically, the embodiment provides an implementation for processing the low-frequency band.

[0049] It is worth noting that the fluid-structure coupling abnormal noise targeted by the present application has a significant low-frequency characteristic, by preferentially analyzing the most valuable low-frequency band, the system can capture the characteristics of abnormal signals more quickly, reducing unnecessary calculation, speeding up the diagnosis process and improving the diagnosis efficiency, truly achieving rapid diagnosis.

[0050] Through the above setting of the present application, when abnormal noise occurs in the train, the abnormal noise spectrum database and the correlation between the inside and outside noise can be used to quickly locate the corresponding noise source area outside the vehicle, and the abnormal noise pattern is given. Especially, the train fluid-structure coupling abnormal noise often has a low-frequency peak characteristic below 500Hz, so the diagnosis process can be further divided into frequency bands to speed up the diagnosis process.

[0051] In some possible embodiments of the present application, the plurality of external aerodynamic sound source areas include at least one of a pantograph, a sanding device, an antenna, and a door area.

[0052] Specifically, the embodiment provides an implementation of an out-of-vehicle aerodynamic sound source area.

[0053] It should be noted that the above-mentioned out-of-vehicle aerodynamic sound source area is taken as an example for description, but the rapid diagnosis method provided by the present application is not limited to the above-mentioned out-of-vehicle aerodynamic sound source area, and is also applicable to other out-of-vehicle aerodynamic sound source areas.

[0054] In some possible embodiments of the present application, the method further comprises: The specific sound source position of the diagnosed abnormal noise is highlighted in the corresponding local area of the three-dimensional vehicle model.

[0055] Specifically, the embodiment provides an implementation of displaying the specific sound source position of the abnormal noise. The specific sound source position of the abnormal noise is displayed in a visual and highlighted manner, so that the maintenance personnel can intuitively understand the fault position, reduce the communication and confirmation cost, improve the overall work efficiency from diagnosis to maintenance, and greatly improve the user friendliness and practicality of the diagnosis system.

[0056] The rapid diagnosis method for abnormal vibration noise of high-speed train fluid-structure coupling provided by the present application builds an aerodynamic noise prediction model based on fluid-structure coupling, analyzes the generation and suppression method of aerodynamic abnormal noise by combining the typical regional noise characteristics in the running process of the high-speed train. The above process is essentially to build a digital twin model connecting the in-vehicle sound field and the out-of-vehicle sound source, and the rapid tracing of abnormal noise is realized through the model.

[0057] In a specific embodiment, the implementation of the rapid diagnosis method for abnormal vibration noise of high-speed train fluid-structure coupling provided by the present application can be divided into two stages: preparation stage and application stage.

[0058] In the preparation stage, the model and the database are constructed, that is, a full-speed-level fluid-structure coupling aerodynamic noise prediction model is constructed, and a mapping relationship is established; that is, an abnormal noise spectrum database is established.

[0059] Firstly, a digital model capable of accurately predicting the acoustic characteristics of a high-speed train in the full-speed domain is constructed, and a mapping relationship is established.

[0060] In some specific embodiments of the present application, Figure 2 The overall process of constructing a full-speed-level aerodynamic noise prediction model and calculating the in-vehicle and out-vehicle noise spectrum is shown in FIG. 1. Figure 2 As shown in FIG. 1, the implementation process of the full-speed-level (200km / h-400km / h) aerodynamic noise prediction model provided by the present embodiment specifically comprises: Step one, obtain a train body simulation model and perform fluid dynamics (CFD) simulation: Based on the precise three-dimensional geometric model of a certain type of high-speed train, a simulation calculation model for computational fluid dynamics (CFD) is established.

[0061] Specifically, according to the train geometric model, a fluid simulation calculation model is established, and a high-precision turbulence simulation method such as large eddy simulation (LES) or detached eddy simulation (DES) method is used to calculate the complex turbulent flow field. After completing the transient flow field calculation, the pressure field data and velocity field data surrounding the train volume domain and other flow field calculation data are obtained as source data for subsequent calculation of aerodynamic noise.

[0062] Further, after obtaining the flow field calculation data, the wheel-rail noise source data and train floor vibration acceleration data obtained through experimental testing are combined to calculate the noise inside and outside the train. The calculation of train noise is carried out in acoustic calculation software (such as Actran, LMS Virtual.Lab, etc.). After establishing the train noise calculation model, aerodynamic noise excitation data is extracted from the aforementioned fluid calculation result data, body sound source data inside the train volume domain is obtained from the velocity field data, and turbulent fluctuating pressure data on the surface of the train body is obtained from the pressure field data. The body sound source, wheel-rail noise, and train vibration-induced noise propagate simultaneously to the inside and outside of the train, and the noise caused by turbulent fluctuating pressure directly propagates into the train. After separate calculation, the noise spectrum at different measurement points inside and outside the train is obtained.

[0063] Specifically, a high-precision turbulence simulation method such as large eddy simulation or detached eddy simulation method can be used to perform large-scale parallel calculation of the unsteady external flow field of the train at multiple typical speeds from 200 km / h to 400 km / h. The goal is to accurately capture the complex turbulent vortex structure, separation, reattachment, and other flow phenomena around the train. After calculation, the time-varying pressure field and velocity field data surrounding the train volume domain are obtained.

[0064] Step two, acoustic simulation and composite excitation source setting: An acoustic calculation model containing the train body structure, the interior sound cavity, and the external sound field is established in professional acoustic calculation software (such as Actran, LMS Virtual.Lab, etc.). The unsteady flow field data obtained in step one are used as the input of the aerodynamic noise source; at the same time, the wheel-rail noise source data and the vibration acceleration data of the train floor (representing the noise transmitted by structural vibration) measured through line experiments and other methods are also loaded into the acoustic model. This enables the model to comprehensively consider the three sources of aerodynamic noise, wheel-rail noise, and structural vibration noise and their coupling effects.

[0065] Step three, establishing a mapping relationship: Acoustic simulations were run to calculate the sound pressure level spectra at multiple preset monitoring points inside the vehicle (such as near windows and in the center of the aisle) and at multiple key aerodynamic sound source areas outside the vehicle (such as pantographs, obstacle clearers, antennas, and door areas) under normal operating conditions. Then, the sound pressure level spectrum data from the n monitoring points inside the vehicle were combined into a column vector I, and the sound pressure level spectrum data from the m sound source areas outside the vehicle were combined into a column vector O. An n×m weighting coefficient matrix A was then solved using mathematical regression methods (such as least squares) to ensure that... Matrix A is the mapping relationship we are looking for, which quantifies the contribution and transmission characteristics of each external sound source to the noise at each internal measurement point.

[0066] While performing global simulation, special attention should be paid to key areas where fluid-structure interaction effects are significant.

[0067] In some specific embodiments of the present invention, such as Figure 3 As shown, Figure 3 The process of analyzing the fluid-structure interaction acoustic characteristics of key areas is demonstrated, specifically including: Step 1: Identify key areas: Through simulation analysis and engineering experience, it was determined that areas such as pantograph base, antenna, obstacle clearer, window, and door edge are the areas where high-speed airflow impact excitation and structural elastic response are most likely to couple.

[0068] Step 2: Perform localized refined analysis: More refined fluid-structure interaction analysis models were established for these key areas to analyze the vibration response and acoustic radiation characteristics of the structure under aerodynamic load excitation under abnormal conditions (such as decreased bolt preload and local aging of sealing strips).

[0069] Step 3: Revealing low-frequency characteristics: For example, the analysis results show that the energy of this type of abnormal noise caused by fluid-structure interaction is often concentrated in a specific low-frequency band, and usually has significant low-frequency peak characteristics below 500 Hz.

[0070] At the same time, an abnormal noise spectrum database needs to be established.

[0071] The goal of this step is to establish a database of fault samples for comparison and identification.

[0072] First, data acquisition is conducted. On the test track or main line, microphones are placed inside and outside the high-speed train to collect a large amount of measured aerodynamic noise spectrum data. The collected operating conditions should be as comprehensive as possible, including normal operating data at various speeds, as well as some known or artificially simulated abnormal operating conditions (such as improper installation of a component, partial damage to the sealing strip, etc.).

[0073] Intelligent analysis and classification are then performed, and machine learning algorithms (such as clustering analysis, support vector machines, deep neural networks, etc.) are used to analyze the massive data collected. The algorithm can first autonomously learn and distinguish the patterns of normal noise spectrum and the patterns of abnormal noise spectrum. Then, for all abnormal data, according to their corresponding physical sources (i.e., sound source positions) and unique spectral characteristics (such as peak values of specific frequencies, harmonic components, etc.), automatic classification and labeling are performed, and finally a structured, rapidly queryable abnormal noise spectrum database is formed.

[0074] To more accurately establish the mapping relationship of the sound field inside and outside the vehicle, in some specific embodiments of the present application, as shown in Figure 4 , Figure 4 a more in-depth analysis process considering the fluid-structure coupling effect is shown, which specifically includes: Step one: unsteady flow field calculation; Step two: fluid-structure coupling interface data transfer; The unsteady fluctuating pressure acting on the surface of the vehicle body calculated in step one is applied as a load to the structural finite element model of the vehicle body.

[0075] Step three: structural dynamics response calculation; The structural dynamics equation is solved to calculate the vibration velocity response of the outer surface panels of the vehicle body.

[0076] Step four: sound radiation calculation; The panel vibration velocity obtained in step three is used as a sound source (structure-borne noise), and combined with the extracted pure aerodynamic sound source (fluid-borne noise), which is input into the acoustic model.

[0077] Step five: obtain the sound field inside and outside the vehicle under the coupling effect.

[0078] Through acoustic simulation, the sound pressure level spectrum inside and outside the vehicle considering the fluid-borne noise and structure-borne noise and their coupling effect is obtained.

[0079] Through the above preparation stage, the mapping relationship (weight coefficient matrix A) containing complex physical processes and the abnormal noise spectrum database established by machine learning methods are obtained.

[0080] When the above preparation work is completed, the diagnostic method of the present application can be used in the daily operation or maintenance of the train, i.e., the application stage.

[0081] In one specific embodiment of the present application, referring to Figure 5 , Figure 5 the overall process of the abnormal noise rapid identification and evaluation provided by the present application is shown, which specifically includes: Step one, obtain real-time data.

[0082] During the operation of the train to be diagnosed, sound pressure level spectrum data at at least one monitoring point is collected in real time using acoustic sensors deployed inside the train (corresponding to the data acquisition module in the system).

[0083] Step 2: Determine if abnormal noise is present. The system rapidly compares the real-time acquired spectral data with the normal noise baseline in a pre-set abnormal noise spectral database. Based on pre-set rules (such as threshold judgment or similarity calculation), the system determines whether the current noise is abnormal. If not, the process ends or monitoring continues; if so, a diagnostic process is triggered.

[0084] For example, the acquired real-time spectrum data can be transmitted to a vehicle-mounted or ground-based processing terminal. The anomaly detection module within the terminal compares this data with a pre-set database of abnormal noise spectra. The detection method can take various forms, such as: Feature comparison is employed: key features of the real-time spectrum (such as total sound pressure level, specific frequency band energy, etc.) are extracted to determine whether they exceed the feature range of the normal pattern in the database (i.e., exceed the preset threshold).

[0085] Pattern matching is employed: the similarity between the real-time spectrum and various typical anomalous noise patterns stored in the database is calculated. When a feature deviates from the normal range or has a high similarity to a certain anomalous pattern, the system determines that anomalous noise exists and proceeds to the next step.

[0086] It is worth noting that fluid-structure interaction (FSI) anomalous noise often exhibits significant low-frequency peaks below 500 Hz. Therefore, during the diagnostic process, priority should be given to analyzing and processing the low-frequency band below 500 Hz to focus on key information and accelerate the judgment process.

[0087] Step 3: Perform sound source inversion: Invoke the mapping relationship (weight coefficient matrix) established during the preparation phase. ), for in-vehicle spectrum data that has been identified as abnormal By performing inversion calculations, the solution is obtained as follows: (2), This allows us to calculate the sound source spectrum of each sound source region outside the vehicle at that moment. .

[0088] Step 4: Perform sound source localization and pattern recognition: The inverted external sound source spectrum The system performs a final match with various classified anomaly patterns in the anomalous noise spectrum database. The match with the highest degree of accuracy determines the specific location of the anomalous noise source and its anomaly pattern. For example, if... The spectrum displayed in the pantograph area is highly consistent with the mode marked as "loose pantograph fairing" in the database, and the system can make a diagnostic conclusion that the abnormal noise is caused by loose pantograph fairing. In this way, the location of the abnormal sound source and the identification of the abnormal mode are realized.

[0089] Step five, result visualization and output: In order to facilitate the maintenance personnel to quickly understand and handle, the diagnostic result can be output in an intuitive way. For example, in a three-dimensional vehicle body model visualization interface, the local area where the located abnormal sound source is located is highlighted or flickered, and a text description is attached, such as "warning: the pantograph area finds suspected low-frequency abnormal noise caused by loose bolt".

[0090] The present application provides a high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method, which realizes rapid data mapping from observation point to excitation source by studying the characteristics of turbulent boundary layer and geometric induced noise of the surface of the typical area vehicle body, combining the wheel-rail noise source model, and constructing a full-speed-level aerodynamic noise prediction model. Through comparative analysis with abnormal sound source spectrum, the location of abnormal sound source is quickly located, and the diagnostic result is highlighted on the corresponding local area of the three-dimensional vehicle body model, which improves the maintenance efficiency. The present application can effectively improve the noise control ability of high-speed train, reduce noise interference, optimize fault diagnosis efficiency, and reduce maintenance cost.

[0091] In some specific embodiments of the present application, as shown in Figure 6 The present application provides a high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis system, which comprises: A data acquisition module 10 is used to acquire real-time sound pressure level spectrum data of at least one monitoring point in the high-speed train to be diagnosed; An abnormality judgment module 20 is used to compare the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether there is abnormal noise; A sound source inversion module 30 is used to calculate the noise spectrum of the aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data by using the mapping relationship between the preset in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle if it is determined that there is abnormal noise; and A positioning and identification module 40 is used to compare and analyze the noise spectrum of the aerodynamic sound source area outside the vehicle calculated by inversion with the abnormal noise spectrum database to locate the specific sound source position of the abnormal noise and / or determine the abnormal mode thereof.

[0092] The present application overcomes the deficiencies of the prior art in noise source identification and control by studying the characteristics of the turbulent boundary layer on the surface of the car body and the geometric induced noise, and using an aerodynamic noise prediction model based on fluid-structure coupling. The present application achieves the goals of improving train operation comfort, optimizing noise source identification and suppression efficiency, and reducing maintenance costs by quickly identifying aerodynamic coupling abnormal noise and locating the abnormal position.

[0093] The high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method and system provided by the present application can accurately simulate the generation and propagation of high-speed train external aerodynamic noise, especially the near-wall body source characteristics of the turbulent boundary layer on the surface of the typical region and the geometric induction, by constructing an aerodynamic noise prediction model covering the full speed range. The present application significantly improves the prediction accuracy of aerodynamic noise. The present application can quickly identify aerodynamic abnormal noise occurring during high-speed train operation, and can combine local and overall noise correlation evaluation to achieve accurate positioning of noise. The present application has significant advantages in the maintenance and detection of high-speed train abnormal noise, and has good scalability and flexibility.

[0094] Figure 7 An example of a schematic diagram of the physical structure of an electronic device is shown in Figure 7 As shown, the electronic device can include a processor 710, a communications interface 720, a memory 730, and a communications bus 740, wherein the processor 710, the communications interface 720, and the memory 730 communicate with each other through the communications bus 740. The processor 710 can invoke the logical instructions in the memory 730 to execute the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the above-mentioned methods, which includes: obtaining real-time sound pressure level spectrum data of at least one monitoring point in the high-speed train to be diagnosed; comparing the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether there is abnormal noise; if it is determined that there is abnormal noise, using the mapping relationship between the in-vehicle noise monitoring point and the plurality of aerodynamic sound source areas outside the vehicle to inversely calculate the noise spectrum of the aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data; and comparing and analyzing the inversely calculated noise spectrum of the aerodynamic sound source area outside the vehicle with the abnormal noise spectrum database to locate the specific sound source position of the abnormal noise and / or determine the abnormal noise mode.

[0095] In addition, the logic instructions in the memory 730 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0096] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to perform the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the above-mentioned methods. The method comprises: obtaining real-time sound pressure level spectrum data of at least one monitoring point in a high-speed train to be diagnosed; comparing the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether there is abnormal noise; if it is determined that there is abnormal noise, using a preset mapping relationship between the in-vehicle noise monitoring point and the multiple aerodynamic sound source areas outside the vehicle to inversely calculate the noise spectrum of the aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data; and comparing and analyzing the inversely calculated noise spectrum of the aerodynamic sound source area outside the vehicle with the abnormal noise spectrum database to locate the specific sound source position of the abnormal noise and / or determine the abnormal noise mode thereof.

[0097] In another aspect, the present application also provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the high-speed train fluid-structure coupling abnormal vibration noise rapid diagnosis method provided by the above-mentioned methods. The method comprises: obtaining real-time sound pressure level spectrum data of at least one monitoring point in a high-speed train to be diagnosed; comparing the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether there is abnormal noise; if it is determined that there is abnormal noise, using a preset mapping relationship between the in-vehicle noise monitoring point and the multiple aerodynamic sound source areas outside the vehicle to inversely calculate the noise spectrum of the aerodynamic sound source area outside the vehicle corresponding to the real-time sound pressure level spectrum data; and comparing and analyzing the inversely calculated noise spectrum of the aerodynamic sound source area outside the vehicle with the abnormal noise spectrum database to locate the specific sound source position of the abnormal noise and / or determine the abnormal noise mode thereof.

[0098] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0099] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0100] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A rapid diagnostic method for abnormal vibration and noise caused by fluid-structure interaction in high-speed trains, characterized in that, include: Acquire real-time sound pressure level spectrum data from at least one monitoring point inside the high-speed train to be diagnosed; The real-time sound pressure level spectrum data is compared with a preset abnormal noise spectrum database to determine whether abnormal noise exists. If abnormal noise is detected, the noise spectrum of the external aerodynamic sound source area corresponding to the real-time sound pressure level spectrum data is calculated by using the preset mapping relationship between in-vehicle noise monitoring points and multiple external aerodynamic sound source areas; and The noise spectrum of the external aerodynamic sound source area calculated by inversion is compared and analyzed with the abnormal noise spectrum database to locate the specific sound source location of the abnormal noise and / or determine its abnormal noise mode.

2. The rapid diagnosis method for abnormal vibration and noise of high-speed train fluid-structure interaction according to claim 1, characterized in that, Before using the pre-defined mapping relationship between in-vehicle noise monitoring points and multiple aerodynamic sound source areas outside the vehicle to invert and calculate the noise spectrum of the external aerodynamic sound source area corresponding to the real-time sound pressure level spectrum data, the method further includes: A full-speed-level fluid-structure interaction aerodynamic noise prediction model was constructed, and based on this model, a mapping relationship was established between preset in-vehicle noise monitoring points and multiple aerodynamic sound source areas outside the vehicle.

3. The rapid diagnostic method for abnormal vibration and noise of high-speed train fluid-structure interaction according to claim 2, characterized in that, The construction of a full-velocity-level fluid-structure interaction aerodynamic noise prediction model, and the establishment of a mapping relationship between preset in-vehicle noise monitoring points and multiple aerodynamic sound source areas outside the vehicle based on this model, specifically includes: Based on the train geometry model, a fluid simulation model covering a speed range of 200km / h to 400km / h was established, and transient flow field calculations were performed using large eddy simulation or separated eddy simulation methods to obtain pressure and velocity field data of the volume domain around the train. An acoustic calculation model was established, using the pressure field and velocity field data, the experimentally measured wheel-rail noise source data and vehicle vibration data as composite excitation sources, and the sound pressure level spectrum of each noise monitoring point inside the vehicle and each aerodynamic sound source area outside the vehicle under normal operating conditions was obtained through simulation calculation. Based on the sound pressure level spectrum data obtained from the simulation calculation, a weighting coefficient matrix is ​​determined by mathematical regression method, and a mapping relationship between preset in-vehicle noise monitoring points and multiple aerodynamic sound source areas outside the vehicle is established based on the weighting coefficient matrix.

4. The rapid diagnosis method for abnormal vibration and noise of high-speed train fluid-structure interaction according to claim 1, characterized in that, The abnormal noise spectrum database is established through the following steps: Collect measured spectrum data of aerodynamic noise inside and outside the high-speed train under various operating conditions; Machine learning algorithms are used to analyze the measured spectrum data to distinguish between normal noise spectrum and abnormal noise spectrum; The abnormal noise spectrum is classified according to the sound source location and spectral characteristics to form a structured abnormal noise spectrum database.

5. The rapid diagnostic method for abnormal vibration and noise of high-speed train fluid-structure interaction according to claim 1, characterized in that, The step of comparing the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether abnormal noise exists specifically includes: The features of the real-time sound pressure level spectrum data are extracted and compared with the feature range of the normal noise spectrum in the abnormal noise spectrum database; And / or, calculate the similarity between the real-time sound pressure level spectrum data and the typical anomalous noise patterns stored in the anomalous noise spectrum database; When the characteristics of the real-time sound pressure level spectrum data deviate from the characteristic range of the normal noise spectrum or the similarity with a certain abnormal noise pattern exceeds a preset threshold, it is determined that there is abnormal noise.

6. The rapid diagnostic method for abnormal vibration and noise of high-speed train fluid-structure interaction according to claim 1, characterized in that, The method further includes: Priority is given to analyzing the low-frequency band with a frequency not higher than 500Hz in the real-time sound pressure level spectrum data to identify specific noise peaks caused by abnormal vibrations of the vehicle body structure.

7. The rapid diagnostic method for abnormal vibration and noise of high-speed train fluid-structure interaction according to claim 1, characterized in that, The multiple aerodynamic sound source areas outside the vehicle include at least one of the pantograph, obstacle clearer, antenna, and door areas.

8. The rapid diagnostic method for abnormal vibration and noise of high-speed train fluid-structure interaction according to any one of claims 1-7, characterized in that, The method further includes: The specific location of the abnormal noise source diagnosed will be highlighted in the corresponding local area of ​​the 3D vehicle model.

9. A rapid diagnostic system for abnormal vibration and noise caused by fluid-structure interaction in high-speed trains, characterized in that, include: The data acquisition module is used to acquire real-time sound pressure level spectrum data at at least one monitoring point inside the high-speed train to be diagnosed; An anomaly detection module is used to compare the real-time sound pressure level spectrum data with a preset abnormal noise spectrum database to determine whether abnormal noise exists. The sound source inversion module is used to calculate the noise spectrum of the external aerodynamic sound source area corresponding to the real-time sound pressure level spectrum data if abnormal noise is detected, by using the preset mapping relationship between the in-vehicle noise monitoring points and multiple external aerodynamic sound source areas. as well as The positioning and identification module is used to compare and analyze the noise spectrum of the external aerodynamic sound source area calculated by inversion with the abnormal noise spectrum database in order to locate the specific sound source location of the abnormal noise and / or determine its abnormal mode.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the rapid diagnosis method for abnormal vibration and noise of high-speed train fluid-structure interaction as described in any one of claims 1 to 8.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the rapid diagnosis method for abnormal vibration and noise of high-speed train fluid-structure interaction as described in any one of claims 1 to 8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the rapid diagnosis method for abnormal vibration and noise of high-speed train fluid-structure interaction as described in any one of claims 1 to 8.

Images