Method and device for detecting abnormality of a cable terminal of a rail vehicle

Abstract

The present application relates to the field of rail transit technology, and more particularly to a real-time detection method and device for abnormal cable terminal of a rail vehicle. The method comprises: collecting environmental temperature data and analysis data of the cable terminal to be detected; and using a preset detection model to analyze and process the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected; wherein the analysis data comprises image data, temperature data and sound data; and the preset detection model comprises a detection model based on analysis and processing of the environmental temperature data, image data and temperature data, or a detection model based on analysis and processing of the environmental temperature data, image data, temperature data and sound data. The present application aims to solve the problem of low accuracy of fault detection of the cable terminal of the rail vehicle due to the easy occurrence of missed reports and false reports when using the existing detection method to detect the cable terminal.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular to a method and device for real-time detection of abnormalities in rail vehicle cable terminals. Background Technology

[0002] Cable terminations are devices installed at the ends of cables to ensure electrical connection to other parts of the system and maintain insulation to the connection point. Semi-rigid cable terminations are a crucial component of the high-voltage system of rail vehicles. Failure of a semi-rigid cable termination directly affects the performance of the vehicle's high-voltage system, thus impacting vehicle operational safety. Therefore, real-time monitoring and detection of the status of semi-rigid cable terminations are necessary to ensure safe vehicle operation.

[0003] Currently, there are no direct monitoring methods for semi-rigid cable terminals in rail vehicles. Referring to monitoring methods for other components in high-voltage systems, temperature detection is the most widely used method due to its mature technology and relatively low cost. Existing temperature detection methods collect the equipment's temperature values ​​and compare them with simple temperature thresholds to determine if the equipment is malfunctioning. However, this method frequently results in missed or false alarms because it fails to adequately consider ambient temperature, equipment consistency, and temperature variations. Therefore, there is an urgent need to establish a real-time cable terminal detection method that is highly accurate, precisely positioned, and efficient. Summary of the Invention

[0004] This invention provides a method and apparatus for real-time detection of abnormalities in cable terminals of rail vehicles, which solves the problem that existing detection methods are prone to missed detections and false alarms when detecting cable terminals, resulting in low accuracy of fault detection of cable terminals in rail vehicles.

[0005] This invention provides a method for detecting anomalies in railway vehicle cable terminals, comprising:

[0006] Collect ambient temperature data and analyze data from the cable terminals under test;

[0007] Based on the ambient temperature data and analysis data, the preset detection model is used for analysis and processing to obtain the detection results of the cable terminal to be tested.

[0008] The analysis data includes image data, temperature data, and sound data;

[0009] The preset detection model includes a detection model based on the analysis and processing of the ambient temperature data, image data, and temperature data, or a detection model based on the analysis and processing of the ambient temperature data, image data, temperature data, and sound data.

[0010] This invention also provides an anomaly detection method for cable terminals of rail vehicles. The step of analyzing and processing the ambient temperature data and analysis data using a preset detection model to obtain the detection result of the cable terminal to be detected includes the following steps: The method employs a detection model based on the analysis and processing of the ambient temperature data, image data, and temperature data.

[0011] Extract the temperature data from each frame of the temperature data;

[0012] Calibrate the range of position coordinate points of the cable terminal to be detected in the image data;

[0013] Based on the temperature data of each frame, extract the temperature value corresponding to the range of the location coordinates;

[0014] Based on the ambient temperature data, the temperature value is processed and transformed;

[0015] Using a preset temperature alarm strategy, the temperature value of each frame after processing and transformation is determined to obtain the detection result of the cable terminal to be tested.

[0016] The present invention also provides an anomaly detection method for a cable terminal of a rail vehicle, wherein the preset temperature alarm strategy includes one or more of the following: an absolute temperature average value alarm strategy, an absolute temperature ratio alarm strategy, a relative temperature average value alarm strategy, a relative temperature ratio alarm strategy, a temperature difference average value alarm strategy, a temperature difference ratio alarm strategy, a temperature rise rate average value alarm strategy, and a temperature rise rate ratio alarm strategy.

[0017] This invention also provides an anomaly detection method for cable terminals of rail vehicles, which utilizes a preset temperature alarm strategy to determine the temperature value of each frame after processing and transformation, and obtains the detection result of the cable terminal to be detected, including:

[0018] Step 1: Set up counter F2 and initialize the counter: F2 = 0;

[0019] Where: Counter F2 represents the number of consecutive cumulative frames of abnormal temperature values ​​at the cable terminal;

[0020] Step 2: Using the preset temperature alarm strategy, determine the temperature value of the current frame of the cable terminal. If abnormal, increment counter F2 by 1; if normal, reset counter F2 to 0.

[0021] Step 3: After judging the counter F2 in Step 2, if the number of consecutive accumulated frames of counter F2 is greater than the set value of frames, it is concluded that the cable terminal is abnormal and the counter is reset to 0. If the number of consecutive accumulated frames of counter F2 is less than the set value of frames, it is concluded that the cable terminal is normal.

[0022] Step 4: Repeat steps 2 and 3 to determine the temperature value of subsequent frames of the cable terminal, and then obtain the detection result of the cable terminal to be tested.

[0023] This invention also provides an anomaly detection method for cable terminals of rail vehicles. The step of analyzing and processing ambient temperature data and analytical data using a preset detection model to obtain the detection result of the cable terminal to be detected includes the following steps:

[0024] Extract the temperature features from the image data;

[0025] Extract the sound features from the sound data that correspond to the time in the image data;

[0026] The temperature and sound features are normalized to construct a fused temperature and sound feature.

[0027] The temperature features, sound features, and temperature and sound fusion features are input into a preset clustering analysis model for detection to obtain the detection results of the cable terminal to be detected.

[0028] The preset clustering analysis model is trained using normal temperature features, sound features, and temperature and sound fusion features as samples, and abnormal temperature features, sound features, and temperature and sound fusion features as samples, along with corresponding labels.

[0029] This invention also provides a method for detecting anomalies in cable terminals of rail vehicles. After analyzing ambient temperature data and analytical data using a preset detection model to obtain the detection results for the cable terminal to be detected, the method further includes a step of locating the anomaly detected in the results.

[0030] If the test result of the cable terminal to be tested is abnormal, the fault location of the abnormal cable terminal is obtained by using the cable terminal anomaly location method based on the analysis data.

[0031] This invention also provides an anomaly detection method for cable terminals of rail vehicles. If the detection result of the cable terminal to be tested is abnormal, the fault location of the abnormal cable terminal is determined using an anomaly location method based on the analysis data of the cable terminal to be tested. The method includes:

[0032] The location of the abnormal cable terminal can be determined by the sound characteristics;

[0033] Based on the image data of the abnormal cable terminals to be detected and the cable terminal recognition model, the abnormal cable terminals are matched using the multi-point positioning method.

[0034] The image area of ​​the abnormal cable terminal after matching is divided into multiple partitions. The temperature of the abnormal cable terminal after partitioning is judged based on the temperature data, and the fault location of the abnormal cable terminal is obtained.

[0035] The cable terminal identification model collects the design data of the cable terminal and, based on the dimensions of the design drawings in the design data, initially produces a cable terminal identification template.

[0036] After training the initially fabricated cable terminal recognition template by matching at least one thousand real contour images of cable terminals, the cable terminal recognition model is obtained.

[0037] The present invention also provides an anomaly detection device for cable terminals of rail vehicles, comprising:

[0038] The data acquisition module is used to collect analytical data from the cable terminals under test.

[0039] The detection module analyzes and processes the data using a preset detection model to obtain the detection results of the cable terminal to be tested.

[0040] The analysis data includes ambient temperature data, image data, temperature data, and sound data of the cable terminal to be tested;

[0041] The preset detection model includes a detection model based on the ambient temperature data, image data, and temperature data, or a detection model based on the analysis and processing of ambient temperature data, image data, temperature data, and sound data.

[0042] This invention also provides an anomaly detection device for cable terminals of rail vehicles. The step of the detection module analyzing and processing the ambient temperature data and analysis data using a preset detection model to obtain the detection result of the cable terminal to be detected includes the following steps:

[0043] Extract the temperature data from each frame of the temperature data;

[0044] Calibrate the range of position coordinate points of the cable terminal to be detected in the image data;

[0045] Based on the temperature data of each frame, extract the temperature value corresponding to the range of the location coordinates;

[0046] Based on the ambient temperature data, the temperature value is processed and transformed;

[0047] Using a preset temperature alarm strategy, the temperature value of each frame after processing and transformation is determined to obtain the detection result of the cable terminal to be tested.

[0048] The present invention also provides an anomaly detection device for a cable terminal of a rail vehicle. The detection module includes a preset temperature alarm strategy, which includes one or more of the following: an absolute temperature average alarm strategy, an absolute temperature ratio alarm strategy, a relative temperature average alarm strategy, a relative temperature ratio alarm strategy, a temperature difference average alarm strategy, a temperature difference ratio alarm strategy, a temperature rise rate average alarm strategy, and a temperature rise rate ratio alarm strategy.

[0049] This invention provides a real-time detection method and device for abnormalities in railway vehicle cable terminals. By using a preset detection model, the method analyzes and detects collected ambient temperature data and analytical data including image data, temperature data, and sound data. This fully considers ambient temperature and temperature changes, avoiding the influence of ambient temperature on the detection results and preventing missed or false alarms, thereby improving the accuracy of cable terminal detection. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating an anomaly detection method for a rail vehicle cable terminal provided by the present invention.

[0052] Figure 2 This is a schematic diagram of the structure of an anomaly detection device for a rail vehicle cable terminal provided by the present invention;

[0053] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0055] Example 1

[0056] The following is combined Figure 1 A method for detecting anomalies in a rail vehicle cable terminal according to the present invention includes:

[0057] S1. Collect ambient temperature data and analysis data of the cable terminal to be tested.

[0058] S2. Based on the ambient temperature data and analysis data, the preset detection model is used for analysis and processing to obtain the detection results of the cable terminal to be tested.

[0059] The analysis data includes image data, temperature data, and sound data.

[0060] Specifically, an infrared camera is used to collect data across the entire cable terminal area, obtaining at least 60 frames of temperature data for the cable terminal and image data of the acquisition area, as well as ambient temperature data. A sound sensor installed near the cable terminal collects sound data. This ensures the completeness of the temperature data acquisition for the cable terminal.

[0061] The preset detection model includes a detection model based on the analysis and processing of the ambient temperature data, image data, and temperature data, or a detection model based on the analysis and processing of the ambient temperature data, image data, temperature data, and sound data.

[0062] This invention utilizes a pre-set detection model to analyze and detect collected ambient temperature data and analytical data including image data, temperature data, and sound data. It fully considers factors such as ambient temperature and temperature changes, avoiding missed or false alarms in the detection results, thereby improving the accuracy of cable terminal detection.

[0063] In this embodiment, step S2, which involves using a preset temperature alarm strategy to determine the temperature value of each frame after processing and transformation to obtain the detection result of the cable terminal to be detected, includes:

[0064] Step 1: Extract the temperature data from each frame of the temperature data.

[0065] Step 2: Define the range of coordinate points of the cable terminal to be detected in the image data. For example, using image data of 100*100 pixels, select the upper left corner coordinate point (2,5) and the lower right corner coordinate point (3,7) of a certain cable terminal frame in the image data. The coordinate points of the cable terminal can be obtained as ((2,5), (2,6), (2,7), (3,5), (3,6), (3,7)).

[0066] Step 3: Based on the temperature data of each frame, extract the temperature value corresponding to the range of the location coordinates. The temperature value T is (34, 36, 35, 35, 36, 34).

[0067] Step 4: Process and transform the temperature value based on the ambient temperature data. The ambient temperature data includes the ambient temperature Tamb (e.g., Tamb = 25℃).

[0068] Step 5: Using a preset temperature alarm strategy, determine the temperature value of each frame after processing and transformation to obtain the detection result of the cable terminal to be tested.

[0069] Specifically, the methods for processing and transforming temperature values ​​are as follows:

[0070] The average temperature Tave in the cable termination area is the average of the temperature values ​​T corresponding to all coordinate points in the cable termination area of ​​the current frame. For example (the temperature values ​​corresponding to all coordinate points are: 34, 36, 35, 35, 36, 34), the average temperature Tave in the cable termination area is 35.

[0071] The relative temperature value Trel of the cable termination area, Trel = T - Tamb: is obtained by subtracting the current ambient temperature value Tamb from the temperature values ​​T corresponding to all coordinate points in the cable termination area of ​​the current frame. For example, Tamb = 25, Trel = (9, 11, 10, 10, 11, 9).

[0072] The average value of the relative temperature Trel in the cable termination area, Trel-ave: This is the average value of the relative temperature Trel corresponding to all coordinate points in the cable termination area in the current frame. For example, Trel = (9, 11, 10, 10, 11, 9), Trel-ave = 10.

[0073] Temperature difference Tdif in cable terminal area: Tdif = T - T1: is the temperature value of all coordinate points in the current frame of the cable terminal area minus the average temperature T1 of other cable terminals. For example, T1 = (33,35,34,34,35,33), Tdif = (1,1,1,1,1,1).

[0074] The average value of the temperature difference Tdif in the cable terminal area, Tdif-ave: is the average value of the temperature difference Tdif corresponding to all coordinate points in the cable terminal area in the current frame. For example, Tdif = (1,1,1,1,1,1), Tdif-ave = 1.

[0075] Temperature rise rate of cable terminal area Trat: Trat = T - T0: is the temperature value of all coordinate points in the current frame of the cable terminal area minus the temperature value T0 of the same cable terminal in the previous frame. For example, T0 = (30,30,30,30,30,30), Trat = (4,6,5,5,6,4).

[0076] The average value of the temperature rise rate Trat in the cable terminal area is Trat-ave: This is the average value of the temperature rise rate Trat corresponding to all coordinate points in the cable terminal area in the current frame. For example, Trat = (4,6,5,5,6,4), Trat-ave = 5.

[0077] Specifically, the preset temperature alarm strategy includes one or more of the following: absolute temperature average value alarm strategy, absolute temperature ratio value alarm strategy, relative temperature average value alarm strategy, relative temperature ratio value alarm strategy, temperature difference average value alarm strategy, temperature difference ratio value alarm strategy, temperature rise rate average value alarm strategy, and temperature rise rate ratio value alarm strategy.

[0078] 1) Absolute temperature average alarm strategy: Calculate the average temperature Tave of the current frame and compare it with the set value Tave-max (e.g., 100℃). If Tave > Tave-max, the current frame is considered to have abnormal temperature data; otherwise, the current frame is considered to have normal temperature data.

[0079] 2) Absolute temperature ratio alarm strategy: Compare all temperature values ​​T in the current frame with the set value Tmax, and count the proportion of temperature points exceeding the set value to the total number of points P1 = (number of points T>Tmax) / total number of points. Then compare P1 with the set value P1-max (e.g., 20%). If P1>P1-max, the current frame is considered to have abnormal temperature data; otherwise, the current frame is considered to have normal temperature data.

[0080] 3) Relative temperature average alarm strategy: Calculate the relative temperature value Trel of the current frame: Trel = T - Tamb, calculate the average value Trel - ave of the relative temperature value of the current frame, and compare it with the set value Trel - ave - max (e.g., 80℃). If Trel - ave > Trel - ave - max, then the current frame is considered to have abnormal temperature data; otherwise, the current frame is considered to have normal temperature data.

[0081] 4) Relative Temperature Ratio Alarm Strategy: Calculate the relative temperature value Trel = T - Tamb for the current frame. Compare all relative temperature values ​​Trel for the current frame with the set value Trel - max. Calculate the percentage of temperature points exceeding the set value out of the total number of points, P2 = (number of points Trel > Trel - max) / total number of points. Compare P2 with the set value P2 - max (e.g., 20%). If P2 > P2 - max, the current frame is considered to have abnormal temperature data; otherwise, the current frame is considered to have normal temperature data.

[0082] 5) Temperature difference average alarm strategy: Calculate the temperature difference Tdif = T - T1 between the current frame temperature T and the temperatures T1 of other cable terminals on the same train (if there are multiple, take the average value). Calculate the average temperature difference Tdif - ave of the current frame and compare it with the set value Tdif - ave - max (e.g., 20℃). If Tdif - ave > Tdif - ave - max, then the current frame is considered to have abnormal temperature data; otherwise, the current frame is considered to have normal temperature data.

[0083] 6) Temperature Difference Ratio Alarm Strategy: Calculate the temperature difference Tdif = T - T1 between the current frame temperature T and the temperatures T1 of other cable terminals on the same train (or the average if there are multiple temperatures). Compare all temperature differences Tdif in the current frame with the set value Tdif-max. Calculate the percentage of points exceeding the set temperature difference value, P3 = (number of points where Tdif > Tdif-max) / total number of points. Compare P3 with the set value P3-max (e.g., 20%). If P3 > P3-max, the current frame is considered to have abnormal temperature data; otherwise, the current frame is considered to have normal temperature data.

[0084] 7) Average temperature rise rate alarm strategy: Calculate the difference Trat = T - T0 between the current frame temperature T and the previous frame temperature T0 of the same cable terminal. Trat is the temperature rise rate of this cable terminal in the current frame. Calculate the average temperature rise rate Trat-ave of the current frame and compare it with the set value Trat-ave-max (e.g., 10℃). If Trat-ave > Trat-ave-max, the current frame is considered to have abnormal temperature data; otherwise, the current frame is considered to have normal temperature data.

[0085] 8) Temperature rise rate ratio alarm strategy: Calculate the difference Trat = T - T0 between the current frame temperature T and the previous frame temperature T0 of the same cable terminal. Trat is the temperature rise rate of this cable terminal in the current frame. Compare all temperature rise rates Trat in the current frame with the set value Trat-max. Count the percentage of points exceeding the set value of temperature rise rate P4 = (number of points Trat > Trat-max) / total number of points. Compare P4 with the set value P4-max (e.g., 20%). If P4 > P4-max, the current frame is considered abnormal temperature data; otherwise, the current frame is considered normal temperature data.

[0086] The above-mentioned strategies for determining the current frame status of a cable terminal can be used individually, i.e., only one strategy can be used to determine the current frame status; or multiple strategies can be combined, and the combination method is not limited to the voting method (e.g., if more than half of the strategies alarm, the current frame is considered abnormal) or the weighting method (e.g., each strategy is assigned a weight of 12.5%). Through the alarm strategy, the differences in ambient temperature, different cable terminals, and the temperature rise during cable terminal faults are fully considered, resulting in a high accuracy rate for the proposed cable terminal anomaly detection method. Step 5: The number of frames with abnormal temperature data after the initial assessment is determined to obtain the detection result for the cable terminal to be tested.

[0087] The step of determining the number of frames of abnormal temperature data after the initial assessment to obtain the detection result of the cable terminal to be tested includes:

[0088] Step 1: Set up counter F2 and initialize the counter: F2 = 0.

[0089] Wherein: Counter F2 represents the cumulative number of consecutive frames of abnormal temperature values ​​at the cable terminal.

[0090] Step 2: Using the preset temperature alarm strategy, judge the temperature value of the current frame of the cable terminal. If it is abnormal, the counter F2 is incremented by 1 and used as the counter for the next frame; if it is normal, the counter F2 is reset to 0.

[0091] Step 3: After judging the counter in Step 2, if the number of consecutive accumulated frames of counter F2 is greater than the set frame number F2-max (for example, F2-max is set to 10, 20 or 30), it is concluded that the cable terminal is abnormal and the counter is reset to 0. If the number of consecutive accumulated frames of counter F2 is less than the set frame number, it is concluded that the cable terminal is normal.

[0092] Step 4: Repeat steps 2 and 3 to determine the temperature value of subsequent frames of the cable terminal, and obtain the detection result of the cable terminal to be tested.

[0093] The setpoints in the above process can be set directly through expert experience, for example, setting Tave-max = 100℃ and Trel-ave-max = 80℃. Alternatively, they can be obtained through statistical analysis of a large amount of temperature data from cable terminals. If the obtained temperature data is all normal, the 3σ principle of normal distribution can be used to set the 3σ limit as the setpoint Tave-max or Trel-ave-max, etc. If the obtained temperature data includes abnormal data, the setpoint can be set between the normal data (e.g., the average normal temperature) and the abnormal data (e.g., the average abnormal temperature) based on the distribution of normal and abnormal data (e.g., the average normal temperature * 0.3 + the average abnormal temperature * 0.7).

[0094] In this embodiment, after analyzing and processing the ambient temperature data and analysis data using a preset detection model to obtain the detection results of the cable terminal to be tested, the method further includes the step of locating abnormal detection results:

[0095] If the test result of the cable terminal to be tested is abnormal, the fault location of the abnormal cable terminal is obtained by using the cable terminal anomaly location method based on the analysis data.

[0096] Specifically, if the detection result of the cable terminal to be tested is abnormal, based on the analysis data of the cable terminal to be tested, the fault location of the abnormal cable terminal is obtained using the cable terminal anomaly location method, including:

[0097] The location of the abnormal cable terminal can be determined by the sound characteristics.

[0098] Based on the image data of the abnormal cable terminals to be detected and the cable terminal identification model, the abnormal cable terminals are matched using a multi-point positioning method (such as the three-point positioning method).

[0099] The image area of ​​the abnormal cable terminal after matching is divided into multiple partitions (e.g., 10 partitions). Temperature data is used to determine the temperature of the abnormal cable terminal within each partition, thus identifying the location of the abnormal partition. This allows for fault location of the abnormal cable terminal.

[0100] The cable terminal identification model collects the design data of the cable terminal and, based on the dimensions of the design drawings in the design data, initially creates a cable terminal identification template. For example, the design drawings provide the coordinates of the outer contour curve of the cable terminal ((1,2), (1,3)...(2,2)).

[0101] The cable terminal recognition model is obtained by matching at least 1000 real contour images of cable terminals and training the initially created cable terminal recognition template. For example, using a 100*100 pixel image, the coordinate points corresponding to the outer contour curve are drawn in the image and fine-tuned according to the image. By matching at least 1000 images and training, the cable terminal recognition model can be obtained, representing the real contour of the cable terminal in the image.

[0102] Example 2

[0103] Based on Example 1, this example modifies the detection model preset in Example 1 to a detection model based on the analysis and processing of the ambient temperature data, image data, temperature data, and sound data.

[0104] Specifically, the step of analyzing and processing ambient temperature data and analysis data using a preset detection model to obtain the detection result of the cable terminal to be tested, including the following steps:

[0105] Extract the temperature features from the image data. For example, temperature features include: average temperature Tave = 35, relative temperature Trel: Trel = T - Tamb = (9, 11, 10, 10, 11, 9), temperature difference Tdif: Tdif = T - T1 = (1, 1, 1, 1, 1, 1), and temperature rise rate Trat: Trat = T - T0 = (4, 6, 5, 5, 6, 4), etc.

[0106] Extract the sound features from the sound data corresponding to the time in the image data. Examples of sound features include: time-domain features such as: average sound amplitude over 1 second, sound waveform index over 1 second, sound impulse index over 1 second, sound kurtosis index over 1 second, sound margin index over 1 second, and sound peak-to-peak value over 1 second. Frequency-domain features include: sound centroid frequency over 1 second, sound root-mean-square frequency over 1 second, short-time power spectral density over 1 second, sound spectral entropy over 1 second, and sound formants over 1 second.

[0107] The temperature and sound features are normalized, converting them to a uniform range of 0 to 1.

[0108] Construct integrated features of temperature and sound. For example: weighted sum of the average sound temperature and average sound amplitude over 1 second; weighted sum of the relative sound temperature and peak-to-peak value over 1 second; maximum value of sound temperature difference and sound waveform index over 1 second; average value of sound temperature rise rate and sound spectral entropy over 1 second, etc.

[0109] The temperature characteristics, sound characteristics, and temperature and sound fusion characteristics are input into a preset clustering analysis model for detection to obtain the detection results of the cable terminal to be detected. The preset clustering analysis model can employ algorithms such as K-MEANS clustering, mean-shift clustering, DBSCAN clustering, or hierarchical clustering.

[0110] The preset clustering analysis model is trained using normal temperature features, sound features, and temperature and sound fusion features as samples, and abnormal temperature features, sound features, and temperature and sound fusion features as samples, along with corresponding labels.

[0111] The abnormal detection device for railway vehicle cable terminals provided by the present invention will be described below. The abnormal detection device for railway vehicle cable terminals described below can be referred to in correspondence with the abnormal detection method for railway vehicle cable terminals described above.

[0112] The present invention also provides an anomaly detection device for cable terminals of rail vehicles, comprising:

[0113] The data acquisition module 210 is used to acquire ambient temperature data and analysis data of the cable terminal to be tested.

[0114] In this embodiment, the data acquisition module includes an infrared camera and a sound sensor. The infrared camera collects temperature data across the entire cable terminal area to obtain the surface temperature of the cable terminal, while simultaneously collecting ambient temperature data. Furthermore, to ensure the completeness of the detection, multiple infrared cameras can be placed at different angles and positions to obtain more comprehensive semi-rigid terminal temperature data. Three or more cameras are used, distributed on a circle at a distance greater than 20cm and less than 100cm from the cable terminal area, and the infrared cameras are evenly distributed along the circumference. The sound sensor is installed near the cable terminal on the roof of the rail vehicle to collect the sound emitted by the cable terminal during operation. Three or more sound sensors are used, distributed on a circle at a distance greater than 20cm and less than 100cm from the cable terminal area, and the sound sensors are evenly distributed along the circumference. After data acquisition, the data is transmitted to the detection module via wired or wireless transmission.

[0115] The detection module 220 analyzes and processes the ambient temperature data and analysis data using a preset detection model to obtain the detection results of the cable terminal to be tested.

[0116] The analysis data includes image data, temperature data, and sound data.

[0117] The preset detection model 220 includes either a detection model based on analyzing and processing the ambient temperature data, image data, and temperature data, or a detection model based on analyzing and processing the ambient temperature data, image data, temperature data, and sound data. This invention uses an acquisition module to collect ambient temperature data and analysis data (including image data, temperature data, and sound data) of the cable terminal to be tested. The detection module then uses the preset detection model to analyze and process the analysis data to detect the cable terminal. By fully considering factors such as ambient temperature, differences between different devices, and temperature rise due to device failure, the accuracy of cable terminal detection is improved.

[0118] The step in which the detection module analyzes and processes the ambient temperature data and analysis data using a preset detection model to obtain the detection result of the cable terminal to be tested includes the following steps: The step of using a detection model based on the analysis and processing of the ambient temperature data, image data, and temperature data includes:

[0119] Extract the temperature data from each frame of the temperature data.

[0120] The range of coordinate points of the cable terminal to be detected in the image data is calibrated.

[0121] Based on the temperature data of each frame, extract the temperature value corresponding to the range of the location coordinates.

[0122] Based on the ambient temperature data, the temperature value is processed and transformed.

[0123] Using a preset temperature alarm strategy, the temperature value of each frame after processing and transformation is determined to obtain the detection result of the cable terminal to be tested.

[0124] Meanwhile, the detection module includes a preset temperature alarm strategy, which includes one or more of the following: absolute temperature average value alarm strategy, absolute temperature ratio value alarm strategy, relative temperature average value alarm strategy, relative temperature ratio value alarm strategy, temperature difference average value alarm strategy, temperature difference ratio value alarm strategy, temperature rise rate average value alarm strategy, and temperature rise rate ratio value alarm strategy.

[0125] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute an anomaly detection method for the cable terminal of a rail vehicle, the method including:

[0126] S1. Collect ambient temperature data and analysis data of the cable terminal to be tested.

[0127] S2. Based on the ambient temperature data and analysis data, the preset detection model is used for analysis and processing to obtain the detection results of the cable terminal to be tested.

[0128] The analysis data includes image data, temperature data, and sound data.

[0129] Specifically, an infrared camera is used to collect data across the entire cable terminal area, obtaining at least 60 frames of temperature data for the cable terminal and image data of the acquisition area, as well as ambient temperature data. A sound sensor installed near the cable terminal collects sound data. This ensures the completeness of the temperature data acquisition for the cable terminal.

[0130] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0131] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the abnormal detection method for rail vehicle cable terminals provided by the above methods, the method comprising:

[0132] S1. Collect ambient temperature data and analysis data of the cable terminal to be tested.

[0133] S2. Based on the ambient temperature data and analysis data, the preset detection model is used for analysis and processing to obtain the detection results of the cable terminal to be tested.

[0134] The analysis data includes image data, temperature data, and sound data.

[0135] Specifically, an infrared camera is used to collect data across the entire cable terminal area, obtaining at least 60 frames of temperature data for the cable terminal and image data of the acquisition area, as well as ambient temperature data. A sound sensor installed near the cable terminal collects sound data. This ensures the completeness of the temperature data acquisition for the cable terminal.

[0136] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an anomaly detection method for rail vehicle cable terminals provided by the methods described above, the method comprising:

[0137] S1. Collect ambient temperature data and analysis data of the cable terminal to be tested.

[0138] S2. Based on the ambient temperature data and analysis data, the preset detection model is used for analysis and processing to obtain the detection results of the cable terminal to be tested.

[0139] The analysis data includes image data, temperature data, and sound data.

[0140] Specifically, an infrared camera is used to collect data across the entire cable terminal area, obtaining at least 60 frames of temperature data for the cable terminal and image data of the acquisition area, as well as ambient temperature data. A sound sensor installed near the cable terminal collects sound data. This ensures the completeness of the temperature data acquisition for the cable terminal.

[0141] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0142] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting anomalies in a rail vehicle cable terminal, characterized in that, include: Collect ambient temperature data and analyze data from the cable terminals under test; The step of analyzing and processing the ambient temperature data and analysis data using a preset detection model to obtain the detection result of the cable terminal to be tested, including the following steps: Extract the temperature features from the image data; Extract sound features from the sound data corresponding to the time of the image data; the sound features include time-domain features and frequency-domain features, wherein the time-domain features include one or more of the following: average sound amplitude over 1 second, sound waveform index over 1 second, sound impulse index over 1 second, and sound peak-to-peak value over 1 second; the frequency-domain features include one or more of the following: root mean square frequency of sound over 1 second, short-time power spectral density of sound over 1 second, and sound spectral entropy over 1 second. The temperature and sound features are normalized to construct a fused temperature and sound feature; wherein the fused temperature and sound feature includes at least one of the following: a weighted sum of the average temperature and the average sound amplitude over 1 second, a weighted sum of the relative temperature and the peak-to-peak value of the sound over 1 second, the maximum value of the temperature difference and the sound waveform index over 1 second, and the average value of the temperature rise rate and the sound spectral entropy over 1 second. The temperature features, sound features, and temperature and sound fusion features are input into a preset clustering analysis model for detection to obtain the detection results of the cable terminal to be detected. The analysis data includes image data, temperature data, and sound data; The preset detection model includes a detection model based on the analysis and processing of the ambient temperature data, image data, and temperature data, or a detection model based on the analysis and processing of the ambient temperature data, image data, temperature data, and sound data.

2. The abnormal detection method for rail vehicle cable terminals according to claim 1, characterized in that, The step of obtaining the detection result of the cable terminal to be tested by analyzing and processing the ambient temperature data and analysis data using a preset detection model includes the following steps: Extract the temperature data from each frame of the temperature data; Calibrate the range of position coordinate points of the cable terminal to be detected in the image data; Based on the temperature data of each frame, extract the temperature value corresponding to the range of the location coordinates; Based on the ambient temperature data, the temperature value is processed and transformed; Using a preset temperature alarm strategy, the temperature value of each frame after processing and transformation is determined to obtain the detection result of the cable terminal to be tested.

3. The abnormality detection method for railway vehicle cable terminals according to claim 2, characterized in that, The preset temperature alarm strategies include one or more of the following: absolute temperature average value alarm strategy, absolute temperature ratio alarm strategy, relative temperature average value alarm strategy, relative temperature ratio alarm strategy, temperature difference average value alarm strategy, temperature difference ratio alarm strategy, temperature rise rate average value alarm strategy, and temperature rise rate ratio alarm strategy.

4. The abnormal detection method for railway vehicle cable terminals according to claim 2, characterized in that, The process of using a preset temperature alarm strategy to determine the temperature value of each processed and transformed frame to obtain the detection result of the cable terminal under test includes: Step 1: Set up counter F2 and initialize the counter: F2=0; Where: Counter F2 represents the number of consecutive cumulative frames of abnormal temperature values ​​at the cable terminal; Step 2: Using the preset temperature alarm strategy, determine the temperature value of the current frame of the cable terminal. If abnormal, increment counter F2 by 1; if normal, reset counter F2 to 0. Step 3: After judging the counter F2 in Step 2, if the number of consecutive accumulated frames of counter F2 is greater than the set value of frames, it is concluded that the cable terminal is abnormal and the counter is reset to 0. If the number of consecutive accumulated frames of counter F2 is less than the set value of frames, it is concluded that the cable terminal is normal. Step 4: Repeat steps 2 and 3 to determine the temperature value of subsequent frames of the cable terminal, and then obtain the detection result of the cable terminal to be tested.

5. The abnormal detection method for railway vehicle cable terminals according to claim 1, characterized in that, The preset clustering analysis model is trained using normal temperature features, sound features, and temperature and sound fusion features as samples, and abnormal temperature features, sound features, and temperature and sound fusion features as samples, along with corresponding labels.

6. The abnormality detection method for railway vehicle cable terminals according to any one of claims 1 to 5, characterized in that, Based on ambient temperature data and analysis data, the system uses a pre-set detection model to analyze and process the data to obtain the detection results for the cable terminal to be tested. The process also includes locating any abnormal detection results. If the test result of the cable terminal to be tested is abnormal, the fault location of the abnormal cable terminal is obtained by using the cable terminal anomaly location method based on the analysis data.

7. The abnormality detection method for railway vehicle cable terminals according to claim 6, characterized in that, If the detection result of the cable terminal to be tested is abnormal, based on the analysis data of the cable terminal to be tested, the fault location of the abnormal cable terminal is obtained using the cable terminal anomaly location method, including: The location of the abnormal cable terminal can be determined by the sound characteristics; Based on the image data of the abnormal cable terminals to be detected and the cable terminal recognition model, the abnormal cable terminals are matched using the multi-point positioning method. The image area of ​​the abnormal cable terminal after matching is divided into multiple partitions. The temperature of the abnormal cable terminal after partitioning is judged based on the temperature data, and the fault location of the abnormal cable terminal is obtained. The cable terminal identification model collects the design data of the cable terminal and, based on the dimensions of the design drawings in the design data, initially produces a cable terminal identification template. After training the initially created cable terminal recognition template by matching at least one thousand real contour images of cable terminals, the cable terminal recognition model is obtained.

8. An anomaly detection device for a rail vehicle cable terminal, characterized in that, include: The data acquisition module is used to collect ambient temperature data and analysis data of the cable terminal under test; The step of the detection module analyzing and processing the ambient temperature data and analysis data using a preset detection model to obtain the detection result of the cable terminal to be tested, includes the following steps: Extract the temperature features from the image data; Extract sound features from the sound data corresponding to the time of the image data; the sound features include time-domain features and frequency-domain features, wherein the time-domain features include one or more of the following: average sound amplitude over 1 second, sound waveform index over 1 second, sound impulse index over 1 second, and sound peak-to-peak value over 1 second; the frequency-domain features include one or more of the following: root mean square frequency of sound over 1 second, short-time power spectral density of sound over 1 second, and sound spectral entropy over 1 second. The temperature and sound features are normalized to construct a fused temperature and sound feature; wherein the fused temperature and sound feature includes at least one of the following: a weighted sum of the average temperature and the average sound amplitude over 1 second, a weighted sum of the relative temperature and the peak-to-peak value of the sound over 1 second, the maximum value of the temperature difference and the sound waveform index over 1 second, and the average value of the temperature rise rate and the sound spectral entropy over 1 second. The temperature features, sound features, and temperature and sound fusion features are input into a preset clustering analysis model for detection to obtain the detection results of the cable terminal to be detected. The analysis data includes image data, temperature data, and sound data; The preset detection model includes a detection model based on the analysis and processing of the ambient temperature data, image data, and temperature data, or a detection model based on the analysis and processing of the ambient temperature data, image data, temperature data, and sound data.

9. The abnormality detection device for railway vehicle cable terminals according to claim 8, characterized in that, The step in which the detection module analyzes and processes the ambient temperature data and analysis data using a preset detection model to obtain the detection result of the cable terminal to be tested includes the following steps: The step of using a detection model based on the analysis and processing of the ambient temperature data, image data, and temperature data includes: Extract the temperature data from each frame of the temperature data; Calibrate the range of position coordinate points of the cable terminal to be detected in the image data; Based on the temperature data of each frame, extract the temperature value corresponding to the range of the location coordinates; Based on the ambient temperature data, the temperature value is processed and transformed; Using a preset temperature alarm strategy, the temperature value of each frame after processing and transformation is determined to obtain the detection result of the cable terminal to be tested.

10. The abnormality detection device for railway vehicle cable terminals according to claim 8 or 9, characterized in that, The detection module includes a preset temperature alarm strategy, which includes one or more of the following: absolute temperature average value alarm strategy, absolute temperature ratio alarm strategy, relative temperature average value alarm strategy, relative temperature ratio alarm strategy, temperature difference average value alarm strategy, temperature difference ratio alarm strategy, temperature rise rate average value alarm strategy, and temperature rise rate ratio alarm strategy.

Images