A rail train gasket performance detection method and system

Abstract

The present invention relates to the technical field of gasket performance detection, and discloses a method and system for detecting the performance of rail train gaskets, the method comprising: arranging a data acquisition device around the train track gasket; obtaining data required for gasket performance detection; transmitting the data required for gasket performance detection to a data processing module for preprocessing to obtain preprocessed data; inputting the preprocessed data into a performance detection module to obtain a physical performance score of the gasket, a durability performance score of the gasket, and a surface damage score of the gasket; transmitting the physical performance score of the gasket, the durability performance score of the gasket, and the surface damage score of the gasket to a data fusion module to obtain a comprehensive score and a health index, and transmitting the comprehensive score and the health index to an early warning module; comparing the comprehensive score and the health index with corresponding thresholds in the early warning module, and if the corresponding thresholds are exceeded, an alarm is issued. The present invention can solve the problem of low efficiency of rail train gasket performance detection in the current technology.

Description

Technical Field

[0001] The present invention relates to the technical field of gasket performance detection, and in particular to a method and system for detecting the performance of a rail train gasket. Background Art

[0002] During the high-speed operation of rail trains, their stability and safety are the key to ensuring the quality of operation. As an important part of the track system, track gaskets have the functions of shock absorption, buffering, sound insulation and bearing the weight of the train. During the long-term operation of the train, track gaskets may experience problems such as aging, wear, and cracks due to repeated impact loads and complex environmental influences, which directly affect the operation safety and riding comfort of the train. Therefore, how to accurately and quickly detect the performance of track gaskets has important research value and practical significance.

[0003] Existing rail gasket inspection methods mainly rely on manual visual inspection, mechanical properties testing, and surface morphology observation. Although these methods can evaluate the performance of gaskets to a certain extent, they also have the following problems:

[0004] Traditional detection methods mostly use a single indicator for detection, such as the hardness and elastic modulus of the gasket, but fail to comprehensively consider the mechanical properties, durability and surface damage of the gasket, which easily leads to one-sided detection results and cannot fully reflect the overall performance status of the gasket. For complex track operation environments, gasket performance is affected by many factors, including force distribution, strain, vibration, temperature and humidity. Existing methods lack efficient data collection and processing capabilities, and it is difficult to evaluate gasket performance from multiple dimensions, especially in terms of real-time and data fusion analysis. There is still much room for improvement. Existing detection methods rely on offline detection and laboratory analysis, with a long detection cycle and low efficiency, and cannot meet the rapid detection needs in train operations. At the same time, manual detection relies heavily on the experience of the detection personnel, and there are subjective errors, which affect the accuracy of the detection results. Existing technologies lack a comprehensive evaluation model and early warning mechanism based on the detection results, making it difficult to timely evaluate and predict the status of the gasket, resulting in the failure to timely discover and deal with potential safety hazards.

[0005] Therefore, there is an urgent need for a gasket performance detection method and system that integrates multi-sensor data acquisition, preprocessing, intelligent analysis and health assessment, which can efficiently and accurately evaluate the physical properties, durability and surface damage of the gasket, and improve the safety and reliability of rail train operation. Summary of the invention

[0006] In view of this, the present invention proposes a rail train gasket performance detection method and system, aiming to solve the problem of low efficiency of rail train gasket performance detection in current technology.

[0007] The present invention provides a method for detecting the performance of a rail vehicle gasket, comprising:

[0008] Arrange a sensor array and an image acquisition device around the train track pad;

[0009] Obtain gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image;

[0010] The gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image data are transmitted to the data processing module for preprocessing to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data, preprocessed gasket environmental temperature and humidity data and preprocessed gasket surface image data;

[0011] Input the pre-processed gasket force distribution data, pre-processed gasket strain data, pre-processed gasket vibration data, pre-processed gasket environmental temperature and humidity data, and pre-processed gasket surface image data into the performance detection module to obtain the gasket physical performance score, gasket durability performance score, and gasket surface damage score;

[0012] The physical performance score, durability score and surface damage score of the gasket are transmitted to the data fusion module, and the comprehensive score and health index are obtained by using the weighted average method and linear regression model, and the comprehensive score and health index are transmitted to the early warning module;

[0013] The comprehensive score and health index are compared with the corresponding thresholds in the early warning module. If the corresponding thresholds are exceeded, an alarm is issued.

[0014] Furthermore, the sensor array includes a piezoelectric sensor, a pressure sensor, an acceleration sensor, and a temperature and humidity sensor.

[0015] Further, the specific content of transmitting the gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image data to the data processing module for preprocessing is: using the wavelet transform method to denoise the gasket force distribution data, gasket strain data, gasket vibration data and gasket environmental temperature and humidity data, and then normalizing them to obtain normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data, and using a machine learning algorithm to extract features from the normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data and preprocessed gasket environmental temperature and humidity data;

[0016] The gasket surface image data is grayed, denoised, enhanced, and then normalized to obtain pre-processed gasket surface image data.

[0017] Furthermore, the gasket force distribution data, gasket strain data, gasket vibration data and gasket environmental temperature and humidity data are denoised by wavelet transform method, and then normalized to obtain normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data, which are expressed as:

[0018] ;

[0019] in, is the denoised gasket force distribution data, is the wavelet basis function, is the wavelet coefficient of the force distribution data after threshold processing, a is the scale, and b is the position parameter;

[0020] ;

[0021] in, is the denoised gasket strain data, is the wavelet coefficient of strain data after threshold processing;

[0022] ;

[0023] in, is the denoised gasket vibration data, is the wavelet coefficient of vibration data after threshold processing;

[0024] ;

[0025] in, is the denoised gasket environmental temperature and humidity data, is the wavelet coefficient of the ambient temperature and humidity data after threshold processing;

[0026] ;

[0027] in, is the normalized gasket force distribution data, is the minimum value of the gasket force distribution data after denoising, is the maximum value of the gasket force distribution data after denoising;

[0028] ;

[0029] in, is the normalized gasket strain data, is the minimum value of the gasket strain data after denoising, is the maximum value of the gasket strain data after denoising;

[0030] ;

[0031] in, is the normalized gasket vibration data, is the minimum value of the gasket vibration data after denoising, is the maximum value of the gasket vibration data after denoising;

[0032] ;

[0033] in, is the normalized gasket environmental temperature and humidity data, is the minimum value of the pad environmental temperature and humidity data after denoising, is the maximum value of the gasket environmental temperature and humidity data after denoising.

[0034] Further, the normalized gasket force distribution data, the normalized gasket strain data, the normalized gasket vibration data and the normalized gasket ambient temperature and humidity data are subjected to feature extraction using a machine learning algorithm to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data and preprocessed gasket ambient temperature and humidity data, which are expressed as:

[0035] ;

[0036] in, is the gasket force distribution data after preprocessing, is the feature extraction function;

[0037] ;

[0038] in, is the gasket strain data after preprocessing;

[0039] ;

[0040] in, is the pre-processed gasket vibration data;

[0041] ;

[0042] in, It is the environmental temperature and humidity data of the gasket after preprocessing.

[0043] Furthermore, the performance detection module includes a physical performance detection unit, a durability performance detection unit, and a surface damage detection unit.

[0044] Furthermore, the physical property score of the gasket, the durability score of the gasket and the surface damage score of the gasket are obtained by inputting the pre-processed gasket force distribution data, the pre-processed gasket strain data and the pre-processed gasket vibration data into the physical property detection unit, obtaining the force distribution score according to the force distribution data, obtaining the deformation score according to the strain data, and obtaining the vibration response score according to the vibration data, and obtaining the physical property score by using the weighted average method for the force distribution score, the deformation score and the vibration response score, which is expressed as:

[0045] ;

[0046] in, Score the physical properties, Score the force distribution, Score the deformation. Score the vibration response, is the first weight, is the second weight, is the third weight;

[0047] The pre-processed gasket force distribution data, pre-processed gasket strain data and pre-processed gasket environmental temperature and humidity data are input into the durability performance testing unit, and the fatigue degree score is obtained according to the force distribution data, the aging rate score is obtained according to the strain data, and the corrosion degree score is obtained according to the environmental temperature and humidity data. The fatigue degree score, aging rate score and corrosion degree score are weighted averaged to obtain the durability performance score, which is expressed as:

[0048] ;

[0049] in, Rating for durability, 'Rate your fatigue level, 'Score the aging speed, 'Score the degree of corrosion, ' is the fourth weight, ' is the fifth weight, ' is the sixth weight;

[0050] The pre-processed gasket surface image data and the pre-processed gasket force distribution data are input into the surface damage detection unit, and the crack score and wear score are obtained according to the surface image data, and the damage score is obtained according to the force distribution data. The surface damage score is obtained by using the weighted average method for the crack score, wear score and damage score, which is expressed as:

[0051] ;

[0052] in, Score surface damage, ''Score the cracks, '' is the wear score, '' is the injury score, '' is the seventh weight, '' is the eighth weight, '' is the ninth weight.

[0053] Furthermore, the physical performance score of the gasket, the durability performance score of the gasket and the surface damage score of the gasket are transmitted to the data fusion module, and the weighted average method and the linear regression model are used to obtain the comprehensive score and the health index expression as follows:

[0054] ;

[0055] in, For the comprehensive rating, ''' is the tenth weight, ''' is the eleventh weight, ''' is the twelfth weight;

[0056] ;

[0057] in, is the health index, a is the first regression coefficient, b is the second regression coefficient, and c is the third regression coefficient.

[0058] Furthermore, the comprehensive score and health index are compared with the corresponding thresholds in the early warning module. If the corresponding thresholds are exceeded, an alarm is issued. The specific contents are: the comprehensive score is compared with the comprehensive score threshold in the early warning module, and the comprehensive score threshold is divided into a poor threshold, a general threshold, a good threshold and an excellent threshold; the health index is compared with the health index threshold in the early warning module, and the health index threshold is divided into a healthy threshold and an unhealthy threshold;

[0059] If the health index is greater than the health threshold and the comprehensive score is greater than the poor threshold, no alarm will be issued;

[0060] If the health index is greater than the unhealthy threshold, less than the healthy threshold, and the overall score is less than the good threshold, a warning is issued;

[0061] If the health index is less than the unhealthy threshold and the comprehensive score is less than the general threshold, an alarm is issued.

[0062] On the other hand, the present invention also proposes a rail vehicle gasket performance detection system, the system comprising:

[0063] A data acquisition module is configured to collect gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data, and gasket surface images;

[0064] A data processing module is configured to pre-process gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data, and gasket surface image data;

[0065] The performance detection module is configured to obtain a physical performance score of the gasket, a durability performance score of the gasket, and a surface damage score of the gasket according to the preprocessed gasket force distribution data, the preprocessed gasket strain data, the preprocessed gasket vibration data, the preprocessed gasket environmental temperature and humidity data, and the preprocessed gasket surface image data;

[0066] A data fusion module is configured to obtain a comprehensive score and a health index according to a physical performance score of the gasket, a durability performance score of the gasket, and a surface damage score of the gasket;

[0067] The early warning module is configured to generate an alarm based on the comparison result of the comprehensive score and health index with the corresponding threshold value in the early warning module.

[0068] Compared with the prior art, the beneficial effects of the present invention are as follows: a rail train gasket performance detection method and system of the present invention deploys a sensor array including piezoelectric sensors, pressure sensors, acceleration sensors and temperature and humidity sensors, and combines an image acquisition device to comprehensively acquire the force distribution, strain, vibration, ambient temperature and humidity and surface image data of the gasket, thereby forming a complete data set covering various aspects of the gasket performance; uses wavelet transform to perform denoising to improve data quality and signal-to-noise ratio; eliminates dimensional differences in data of different dimensions through normalization processing; combines machine learning algorithms to perform feature extraction, deeply mines the potential information of the data, and lays the foundation for subsequent performance testing; the performance detection module evaluates the gasket performance from multiple dimensions, and calculates the physical performance score, durability performance score and surface damage score of the gasket respectively. This multi-dimensional scoring method makes up for the shortcomings of the single evaluation index of the traditional detection method and reflects the status of the gasket more comprehensively. The data fusion module is used to calculate the comprehensive score and health index through weighted average and linear regression model, and multi-level warning thresholds (poor, general, good, excellent, etc.) are set to promptly alarm the potential failure of the gasket, effectively improving the safety and reliability of the rail train system. Combined with the data analysis of external factors such as ambient temperature, humidity, and vibration, it can accurately evaluate the use status of the gasket under different working conditions and adapt to the complex rail train operating environment. The system realizes the automation of the entire process from data collection to performance analysis, without the need for human intervention, which greatly reduces labor costs, while improving detection efficiency and the objectivity of results. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0070] Figure 1 This is a flow chart of a method for detecting performance of a rail vehicle gasket according to an embodiment of the present invention;

[0071] Figure 2 The present invention is a structural block diagram of a rail train gasket performance detection system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0072] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0073] See also Figure 1 As shown, an embodiment of the present invention provides a method for detecting performance of a rail vehicle gasket, comprising:

[0074] Arrange a sensor array and an image acquisition device around the train track pad;

[0075] Obtain gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image;

[0076] The gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image data are transmitted to the data processing module for preprocessing to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data, preprocessed gasket environmental temperature and humidity data and preprocessed gasket surface image data;

[0077] Input the pre-processed gasket force distribution data, pre-processed gasket strain data, pre-processed gasket vibration data, pre-processed gasket environmental temperature and humidity data, and pre-processed gasket surface image data into the performance detection module to obtain the gasket physical performance score, gasket durability performance score, and gasket surface damage score;

[0078] The physical performance score, durability score and surface damage score of the gasket are transmitted to the data fusion module, and the comprehensive score and health index are obtained by using the weighted average method and linear regression model, and the comprehensive score and health index are transmitted to the early warning module;

[0079] The comprehensive score and health index are compared with the corresponding thresholds in the early warning module. If the corresponding thresholds are exceeded, an alarm is issued.

[0080] Furthermore, the sensor array includes a piezoelectric sensor, a pressure sensor, an acceleration sensor, and a temperature and humidity sensor.

[0081] Further, the specific content of transmitting the gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image data to the data processing module for preprocessing is: using the wavelet transform method to denoise the gasket force distribution data, gasket strain data, gasket vibration data and gasket environmental temperature and humidity data, and then normalizing them to obtain normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data, and using a machine learning algorithm to extract features from the normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data and preprocessed gasket environmental temperature and humidity data;

[0082] The gasket surface image data is grayed, denoised, enhanced, and then normalized to obtain pre-processed gasket surface image data.

[0083] It should be noted that wavelet transform denoising is suitable for processing multidimensional signals. It can effectively filter out noise in the original data while retaining important feature information in the data to avoid noise interference in subsequent analysis and judgment. The removal of high-frequency noise in vibration data and strain data improves the accuracy of feature extraction. Abnormal fluctuations in environmental temperature and humidity data are smoothed to enhance data stability.

[0084] Improves the reliability and accuracy of sensor data, laying a solid foundation for subsequent normalization and feature extraction. Reduces the impact of external interference on gasket monitoring data, allowing the model to focus more on key features that reflect the true state of the gasket. Data normalization can scale data of different physical quantities (such as force, vibration, temperature and humidity) to a unified range to avoid imbalanced feature weights caused by different dimensions. Eliminates differences in the acquisition range and accuracy of different sensors, so that data has consistent expression capabilities in machine learning models.

[0085] Improve the convergence speed of the model and avoid gradient explosion or vanishing problems caused by inconsistent data range during training. Optimize the sensitivity of the feature extraction algorithm to multi-source data to ensure that key indicators of the health status of the gasket are effectively captured.

[0086] By using machine learning algorithms (such as PCA, LSTM or CNN) to extract features from normalized data, the core features of the gasket health status can be identified from multi-dimensional sensor data, and key features can be extracted from massive monitoring data, which significantly reduces the data dimension and reduces the subsequent analysis and storage pressure. It strengthens the monitoring system's ability to perceive potential risks of gaskets and provides high-quality input for real-time warning and health assessment.

[0087] Grayscale processing converts color images into single-channel grayscale images, which simplifies the data dimension and facilitates the subsequent algorithm to extract surface texture and defect features. Denoising and enhancement processing removes image noise through median filtering, Gaussian smoothing and other methods, while enhancing the edge and texture features of the gasket surface and improving the sensitivity of defect detection. Normalization processing maps pixel values ​​to a uniform range, improves the contrast consistency of image data, and provides standardized input for subsequent feature extraction.

[0088] The accuracy and robustness of gasket surface defect detection are improved, which helps to timely detect cracks, wear, deformation and other problems. The complexity of image processing is simplified, so that subsequent AI models can efficiently analyze image data and improve processing efficiency.

[0089] Furthermore, the gasket force distribution data, gasket strain data, gasket vibration data and gasket environmental temperature and humidity data are denoised by wavelet transform method, and then normalized to obtain normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data, which are expressed as:

[0090] ;

[0091] in, is the denoised gasket force distribution data, is the wavelet basis function, is the wavelet coefficient of the force distribution data after threshold processing, a is the scale, and b is the position parameter;

[0092] ;

[0093] in, is the denoised gasket strain data, is the wavelet coefficient of strain data after threshold processing;

[0094] ;

[0095] in, is the denoised gasket vibration data, is the wavelet coefficient of vibration data after threshold processing;

[0096] ;

[0097] in, is the denoised gasket environmental temperature and humidity data, is the wavelet coefficient of the ambient temperature and humidity data after threshold processing;

[0098] ;

[0099] in, is the normalized gasket force distribution data, is the minimum value of the gasket force distribution data after denoising, is the maximum value of the gasket force distribution data after denoising;

[0100] ;

[0101] in, is the normalized gasket strain data, is the minimum value of the gasket strain data after denoising, is the maximum value of the gasket strain data after denoising;

[0102] ;

[0103] in, is the normalized gasket vibration data, is the minimum value of the gasket vibration data after denoising, is the maximum value of the gasket vibration data after denoising;

[0104] ;

[0105] in, is the normalized gasket environmental temperature and humidity data, is the minimum value of the pad environmental temperature and humidity data after denoising, is the maximum value of the gasket environmental temperature and humidity data after denoising.

[0106] Further, the normalized gasket force distribution data, the normalized gasket strain data, the normalized gasket vibration data and the normalized gasket ambient temperature and humidity data are subjected to feature extraction using a machine learning algorithm to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data and preprocessed gasket ambient temperature and humidity data, which are expressed as:

[0107] ;

[0108] in, is the gasket force distribution data after preprocessing, is the feature extraction function;

[0109] ;

[0110] in, is the gasket strain data after preprocessing;

[0111] ;

[0112] in, is the pre-processed gasket vibration data;

[0113] ;

[0114] in, It is the environmental temperature and humidity data of the gasket after preprocessing.

[0115] Furthermore, the performance detection module includes a physical performance detection unit, a durability performance detection unit, and a surface damage detection unit.

[0116] Furthermore, the physical property score of the gasket, the durability score of the gasket and the surface damage score of the gasket are obtained by inputting the pre-processed gasket force distribution data, the pre-processed gasket strain data and the pre-processed gasket vibration data into the physical property detection unit, obtaining the force distribution score according to the force distribution data, obtaining the deformation score according to the strain data, and obtaining the vibration response score according to the vibration data, and obtaining the physical property score by using the weighted average method for the force distribution score, the deformation score and the vibration response score, which is expressed as:

[0117] ;

[0118] in, Score the physical properties, Score the force distribution, Score the deformation. Score the vibration response, is the first weight, is the second weight, is the third weight;

[0119] The pre-processed gasket force distribution data, pre-processed gasket strain data and pre-processed gasket environmental temperature and humidity data are input into the durability performance testing unit, and the fatigue degree score is obtained according to the force distribution data, the aging rate score is obtained according to the strain data, and the corrosion degree score is obtained according to the environmental temperature and humidity data. The fatigue degree score, aging rate score and corrosion degree score are weighted averaged to obtain the durability performance score, which is expressed as:

[0120] ;

[0121] in, Rating for durability, 'Rate your fatigue level, 'Score the aging speed, 'Score the degree of corrosion, ' is the fourth weight, ' is the fifth weight, ' is the sixth weight;

[0122] The pre-processed gasket surface image data and the pre-processed gasket force distribution data are input into the surface damage detection unit, and the crack score and wear score are obtained according to the surface image data, and the damage score is obtained according to the force distribution data. The surface damage score is obtained by using the weighted average method for the crack score, wear score and damage score, which is expressed as:

[0123] ;

[0124] in, Score surface damage, ''Score the cracks, '' is the wear score, '' is the injury score, '' is the seventh weight, '' is the eighth weight, '' is the ninth weight.

[0125] Furthermore, the physical performance score of the gasket, the durability performance score of the gasket and the surface damage score of the gasket are transmitted to the data fusion module, and the weighted average method and the linear regression model are used to obtain the comprehensive score and the health index expression as follows:

[0126] ;

[0127] in, For the comprehensive rating, ''' is the tenth weight, ''' is the eleventh weight, ''' is the twelfth weight;

[0128] ;

[0129] in, is the health index, a is the first regression coefficient, b is the second regression coefficient, and c is the third regression coefficient.

[0130] Specifically, the pre-processed gasket force distribution data, the pre-processed gasket strain data, and the pre-processed gasket vibration data are input into the physical performance detection unit:

[0131] Stress distribution analysis: By analyzing the stress distribution of the gasket under different loading conditions, we can check whether the stress is uniform and whether there are any stress concentration areas, so as to obtain the stress distribution score. Uniformly distributed stress will increase the overall stability of the gasket and avoid local excessive wear and fatigue.

[0132] Strain data analysis: Based on the strain response of the gasket, its elasticity and deformation capacity are evaluated, and its deformation degree under stress is tested to see whether it is within a safe range, and the deformation score is obtained. The smaller or more uniform the deformation score, the better the gasket's ability to resist deformation.

[0133] Vibration response: Vibration data can reveal the dynamic stability of the gasket. Check whether there is a large vibration or the frequency exceeds the safe range during use, and get the vibration response score. The larger the vibration response score, the worse the stability of the gasket in the working environment, which affects its physical performance score.

[0134] Physical performance score results: Combining the above indicators, a physical performance score is generated through a weighted formula or model, usually distributed between 0 and 100. The higher the score, the better the load-bearing capacity and vibration resistance of the gasket.

[0135] Input the pre-processed gasket force distribution data, pre-processed gasket strain data and pre-processed gasket environmental temperature and humidity data into the durability performance testing unit:

[0136] Fatigue strength assessment: By analyzing the strain changes of the gasket under long-term stress conditions, simulating its fatigue behavior in long-term use, and obtaining a fatigue degree score.

[0137] Environmental adaptability analysis: Ambient temperature and humidity have a great impact on the life of the gasket. The greater the temperature and humidity changes, the lower the corrosion score. By incorporating temperature and humidity conditions into the scoring system, the durability performance of the gasket in various climates can be more accurately simulated.

[0138] Aging resistance: Combined with long-term strain data, the structural and performance change trends of the gasket under aging conditions are determined to obtain an aging rate score.

[0139] Durability score results: By combining these data, a durability score is obtained. This score can indicate the expected life of the gasket in various usage environments, and the higher the value, the stronger the durability.

[0140] Input the preprocessed gasket surface image data and preprocessed gasket force distribution data into the surface damage detection unit:

[0141] Image analysis: The damage features (such as cracks, scratches, corrosion) on the gasket surface are detected through image processing algorithms (such as edge detection, texture analysis, etc.) to obtain the wear score. The larger the damage area or depth, the lower the wear score.

[0142] Damage type analysis: The damage type is distinguished based on the damage shape and size in the image to obtain a crack score. For example, a deeper crack affects the structural integrity and will have a lower score.

[0143] Force area correlation: Combine the force distribution data with the image damage data to check whether the damage area is concentrated in the area with greater force and obtain the damage score.

[0144] Surface damage score result: A surface damage score is obtained based on the damage detection results. The lower the score, the higher the degree of surface damage and the worse the surface quality.

[0145] Furthermore, the comprehensive score and health index are compared with the corresponding thresholds in the early warning module. If the corresponding thresholds are exceeded, an alarm is issued. The specific contents are: the comprehensive score is compared with the comprehensive score threshold in the early warning module, and the comprehensive score threshold is divided into a poor threshold, a general threshold, a good threshold and an excellent threshold; the health index is compared with the health index threshold in the early warning module, and the health index threshold is divided into a healthy threshold and an unhealthy threshold;

[0146] If the health index is greater than the health threshold and the comprehensive score is greater than the poor threshold, no alarm will be issued;

[0147] If the health index is greater than the unhealthy threshold, less than the healthy threshold, and the overall score is less than the good threshold, a warning is issued;

[0148] If the health index is less than the unhealthy threshold and the comprehensive score is less than the general threshold, an alarm is issued.

[0149] It should be noted that the threshold of the comprehensive score is divided into four levels: "poor", "average", "good" and "excellent", and the health index is divided into two levels: "healthy" and "unhealthy". By dividing the threshold, a clear gasket status evaluation standard is established, which is helpful for scientific analysis of the current performance and future status trend of the gasket.

[0150] Based on the combined status of the health index and comprehensive score, the system is divided into the following three levels of response:

[0151] Normal state: When the health index is higher than the health threshold and the comprehensive score is higher than the poor threshold, the gasket performance is good and no alarm is required, reducing unnecessary interference.

[0152] Warning status: When the health index approaches the unhealthy threshold but has not completely exceeded it, and the overall score is average or good, the system issues a warning prompt to guide maintenance personnel to pay attention to potential problems.

[0153] Alarm status: When the health index is lower than the unhealthy threshold and the overall score is poor, the system triggers an alarm, indicating that the gasket may have serious performance problems and immediate maintenance measures are required.

[0154] By comparing the dual thresholds of health index and comprehensive score, accurate judgment of rail gasket performance can be achieved, reducing misjudgment caused by a single parameter. Real-time monitoring and response mechanism ensures that problems can be discovered at an early stage.

[0155] The hierarchical alarm mechanism helps maintenance personnel prioritize high-risk issues, allocate resources reasonably, and reduce the cost of blind repairs and comprehensive inspections.

[0156] Through dynamic monitoring and alarm of comprehensive scores and health indexes, sudden accidents caused by performance degradation of rail gaskets can be effectively avoided, improving the operational safety and reliability of the rail system.

[0157] The automated assessment and alarm mechanism combined with data analysis and threshold comparison enables the system to have autonomous judgment capabilities, reduce manual intervention, and achieve intelligent monitoring of gasket status.

[0158] See also Figure 2 As shown, an embodiment of the present invention further provides a rail vehicle gasket performance detection system, the system comprising:

[0159] A data acquisition module is configured to collect gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data, and gasket surface images;

[0160] A data processing module is configured to pre-process gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data, and gasket surface image data;

[0161] The performance detection module is configured to obtain a physical performance score of the gasket, a durability performance score of the gasket, and a surface damage score of the gasket according to the preprocessed gasket force distribution data, the preprocessed gasket strain data, the preprocessed gasket vibration data, the preprocessed gasket environmental temperature and humidity data, and the preprocessed gasket surface image data;

[0162] A data fusion module is configured to obtain a comprehensive score and a health index according to a physical performance score of the gasket, a durability performance score of the gasket, and a surface damage score of the gasket;

[0163] The early warning module is configured to generate an alarm based on the comparison result of the comprehensive score and health index with the corresponding threshold value in the early warning module.

[0164] The data acquisition module includes a sensor unit and an image acquisition unit;

[0165] The performance detection module includes a physical performance detection unit, a durability performance detection unit, and a surface damage detection unit.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for detecting the performance of a rail vehicle gasket, characterized in that: include: Arrange a sensor array and an image acquisition device around the train track pad; Obtain gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data, and gasket surface image data; The gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image data are transmitted to the data processing module for preprocessing to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data, preprocessed gasket environmental temperature and humidity data and preprocessed gasket surface image data; The pre-processed gasket force distribution data, pre-processed gasket strain data, pre-processed gasket vibration data, pre-processed gasket environmental temperature and humidity data and pre-processed gasket surface image data are input into the performance detection module to obtain the gasket physical performance score, the gasket durability performance score and the gasket surface damage score. The performance detection module includes a physical performance detection unit, a durability performance detection unit and a surface damage detection unit. The physical performance score, the durability performance score and the surface damage score of the gasket are obtained by: inputting the pre-processed gasket force distribution data, the pre-processed gasket strain data and the pre-processed gasket vibration data into the physical performance detection unit, obtaining the force distribution score according to the force distribution data, obtaining the deformation score according to the strain data, and obtaining the vibration response score according to the vibration data. The physical performance score is obtained by using the weighted average method for the force distribution score, the deformation score and the vibration response score, which is expressed as: ； in, Score the physical properties, Score the force distribution, Score the deformation. Score the vibration response, is the first weight, is the second weight, is the third weight; The pre-processed gasket force distribution data, pre-processed gasket strain data and pre-processed gasket environmental temperature and humidity data are input into the durability performance testing unit, and the fatigue degree score is obtained according to the force distribution data, the aging rate score is obtained according to the strain data, and the corrosion degree score is obtained according to the environmental temperature and humidity data. The fatigue degree score, aging rate score and corrosion degree score are weighted averaged to obtain the durability performance score, which is expressed as: ； in, Rating for durability, 'Rate your fatigue level, 'Score the aging speed, 'Score the degree of corrosion, ' is the fourth weight, ' is the fifth weight, ' is the sixth weight; The pre-processed gasket surface image data and the pre-processed gasket force distribution data are input into the surface damage detection unit, and the crack score and wear score are obtained according to the surface image data, and the damage score is obtained according to the force distribution data. The surface damage score is obtained by using the weighted average method for the crack score, wear score and damage score, which is expressed as: ； in, Score surface damage, ''Score the cracks, '' is the wear score, '' is the injury score, '' is the seventh weight, '' is the eighth weight, '' is the ninth weight; The physical performance score, durability score and surface damage score of the gasket are transmitted to the data fusion module, and the weighted average method and linear regression model are used to obtain the comprehensive score and health index, which are expressed as: ； in, For the comprehensive rating, ''' is the tenth weight, ''' is the eleventh weight, ''' is the twelfth weight; ； in, is the health index, a is the first regression coefficient, b is the second regression coefficient, and c is the third regression coefficient; The comprehensive score and health index are transmitted to the early warning module, and compared with the corresponding thresholds in the early warning module. If the corresponding thresholds are exceeded, an alarm is issued.

2. A rail vehicle gasket performance detection method according to claim 1, characterized in that: The specific content of transmitting the gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data and gasket surface image data to the data processing module for preprocessing is as follows: the gasket force distribution data, gasket strain data, gasket vibration data and gasket environmental temperature and humidity data are denoised by wavelet transform method, and then normalized to obtain normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data; the normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data are feature extracted by machine learning algorithm to obtain preprocessed gasket force distribution data, preprocessed gasket strain data, preprocessed gasket vibration data and preprocessed gasket environmental temperature and humidity data; The gasket surface image data is grayed, denoised, enhanced, and then normalized to obtain pre-processed gasket surface image data.

3. A rail vehicle gasket performance detection method according to claim 2, characterized in that: The gasket force distribution data, gasket strain data, gasket vibration data and gasket environmental temperature and humidity data are denoised by wavelet transform method, and then normalized to obtain normalized gasket force distribution data, normalized gasket strain data, normalized gasket vibration data and normalized gasket environmental temperature and humidity data, which are expressed as: ； in, is the denoised gasket force distribution data, is the wavelet basis function, is the wavelet coefficient of the force distribution data after threshold processing, a is the scale, and b is the position parameter; ； in, is the denoised gasket strain data, is the wavelet coefficient of strain data after threshold processing; ； in, is the denoised gasket vibration data, is the wavelet coefficient of vibration data after threshold processing; ； in, is the denoised gasket environmental temperature and humidity data, is the wavelet coefficient of the ambient temperature and humidity data after threshold processing; ； in, is the normalized gasket force distribution data, is the minimum value of the gasket force distribution data after denoising, is the maximum value of the gasket force distribution data after denoising; ； in, is the normalized gasket strain data, is the minimum value of the gasket strain data after denoising, is the maximum value of the gasket strain data after denoising; ； in, is the normalized gasket vibration data, is the minimum value of the gasket vibration data after denoising, is the maximum value of the gasket vibration data after denoising; ； in, is the normalized gasket environmental temperature and humidity data, is the minimum value of the pad environmental temperature and humidity data after denoising, is the maximum value of the gasket environmental temperature and humidity data after denoising.

4. A rail vehicle gasket performance detection method according to claim 3, characterized in that: The normalized gasket force distribution data, the normalized gasket strain data, the normalized gasket vibration data and the normalized gasket ambient temperature and humidity data are subjected to feature extraction using a machine learning algorithm to obtain pre-processed gasket force distribution data, pre-processed gasket strain data, pre-processed gasket vibration data and pre-processed gasket ambient temperature and humidity data, which are expressed as: ； in, is the gasket force distribution data after preprocessing, is the feature extraction function; ； in, is the gasket strain data after preprocessing; ； in, is the pre-processed gasket vibration data; ； in, It is the environmental temperature and humidity data of the gasket after preprocessing.

5. A rail vehicle gasket performance detection method according to claim 4, characterized in that: The comprehensive score and health index are compared with the corresponding thresholds in the early warning module. If the corresponding thresholds are exceeded, an alarm is issued. The specific contents are: the comprehensive score is compared with the comprehensive score threshold in the early warning module. The comprehensive score threshold is divided into a poor threshold, a general threshold, a good threshold and an excellent threshold. The health index is compared with the health index threshold in the early warning module. The health index threshold is divided into a healthy threshold and an unhealthy threshold; If the health index is greater than the health threshold and the comprehensive score is greater than the poor threshold, no alarm will be issued; If the health index is greater than the unhealthy threshold, less than the healthy threshold, and the overall score is less than the good threshold, a warning is issued; If the health index is less than the unhealthy threshold and the comprehensive score is less than the general threshold, an alarm is issued.

6. A rail train gasket performance detection system, implemented in a rail train gasket performance detection method according to any one of claims 1 to 5, characterized in that: The system comprises: A data acquisition module is configured to collect gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data, and gasket surface images; A data processing module is configured to pre-process gasket force distribution data, gasket strain data, gasket vibration data, gasket environmental temperature and humidity data, and gasket surface image data; The performance detection module is configured to obtain a physical performance score of the gasket, a durability performance score of the gasket, and a surface damage score of the gasket according to the preprocessed gasket force distribution data, the preprocessed gasket strain data, the preprocessed gasket vibration data, the preprocessed gasket environmental temperature and humidity data, and the preprocessed gasket surface image data; A data fusion module is configured to obtain a comprehensive score and a health index according to a physical performance score of the gasket, a durability performance score of the gasket, and a surface damage score of the gasket; The early warning module is configured to generate an alarm based on the comparison result of the comprehensive score and health index with the corresponding threshold value in the early warning module.

Images