Method and apparatus for detecting abnormality of cable terminal of rail vehicle

Abstract

This application relates to the technical field of rail transit, and particularly to a method and device for real-time detection of abnormalities in cable terminals of rail vehicles. This method includes collecting environmental temperature data and analysis data of the cable terminals to be detected, and performing analysis processing using a preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminals to be detected. Here, the analysis data includes image data, temperature data, and audio data, and the preset detection model includes a detection model based on analysis processing of the environmental temperature data, image data, and temperature data, or a detection model based on analysis processing of the environmental temperature data, image data, temperature data, and audio data. The purpose of this application is to solve the problem that in the conventional detection method, when detecting cable terminals, reporting omissions and false alarms are likely to occur, resulting in a low accuracy rate of fault detection of cable terminals of rail vehicles.

Description

Cross-reference

[0001] This application claims the priority of a Chinese patent application filed on December 8, 2022, with application number 202211575348.4 and invention title "Method and Device for Detecting Abnormalities of Cable Terminals of Rail Vehicles", the entire content of which is incorporated herein by reference.

Technical Field

[0002] This application relates to the technical field of rail transit, and in particular, to a method and device for real-time detection of abnormalities of cable terminals of rail vehicles.

Background Art

[0003] A cable terminal refers to a device attached to the end of a cable to ensure electrical connection with other parts of the system and maintain insulation up to the connection point. Here, a semi-rigid cable terminal is an important component of the high-voltage system of a rail vehicle. When a failure occurs in the semi-rigid cable terminal, it directly affects the performance of the vehicle's high-voltage system and further affects the safe operation of the vehicle. Therefore, it is necessary to monitor and detect the state of the semi-rigid cable terminal in real time to ensure the safe operation of the vehicle.

[0004] Conventionally, there is no direct monitoring means for the semi-rigid cable terminals of rail vehicles. Referring to the monitoring means for other components of the high-voltage system, temperature detection is the most widely applied detection means because its technology is relatively mature and the cost is relatively low. The conventional temperature detection method collects the temperature value of the device and compares it with a simple temperature threshold to determine whether an abnormality has occurred in the device. However, such a method cannot fully consider factors such as environmental temperature, device integrity, and temperature changes, so reporting omissions and false alarms often occur. Therefore, it is necessary to establish a real-time detection method for cable terminals with high accuracy, accurate positioning, and high efficiency.

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the conventional detection method, when detecting a cable terminal, there are likely to be reporting omissions and false alarms, so the accuracy rate of fault detection of the cable terminal of a rail vehicle is low. To solve this problem, the present application provides a method and device for real-time detection of abnormalities of the cable terminal of a rail vehicle.

Means for Solving the Problem

[0006] The present application provides a method for detecting abnormalities of a cable terminal of a rail vehicle, collecting environmental temperature data and analysis data of the cable terminal to be detected, performing analysis processing using a preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected, and wherein the analysis data includes image data, temperature data, and audio data, the preset detection model includes a detection model based on analysis processing of the environmental temperature data, image data, and temperature data, or a detection model based on analysis processing of the environmental temperature data, image data, temperature data, and audio data.

[0007] The present application further provides a method for detecting abnormalities of a cable terminal of a rail vehicle. In the step of performing analysis processing using a preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected, the step of using a detection model based on analysis processing of the environmental temperature data, image data, and temperature data includes extracting temperature data of each frame in the temperature data, calibrating the position coordinate point range of the cable terminal to be detected in the image data, extracting a temperature value corresponding to the position coordinate point range based on the temperature data of each frame, performing processing and conversion on the temperature value based on the environmental temperature data, Using a preset temperature warning policy, determine the temperature value of each frame after processing and conversion, and obtain a detection result of the cable terminal to be detected.

[0008] The present application further provides a method for detecting an abnormality of a cable terminal of a rail vehicle. The preset temperature warning policy includes one or more of an absolute temperature average value warning policy, an absolute temperature ratio value warning policy, a relative temperature average value warning policy, a relative temperature ratio value warning policy, a temperature difference average value warning policy, a temperature difference ratio value warning policy, a temperature rise rate average value warning policy, and a temperature rise rate ratio value warning policy.

[0009] The present application further provides a method for detecting an abnormality of a cable terminal of a rail vehicle. Using the above-mentioned preset temperature warning policy to determine the temperature value of each frame after processing and conversion, and obtaining a detection result of the cable terminal to be detected is Step 1 of setting counter F2 and initializing the counter so that F2 = 0, where counter F2 represents the number of consecutive integrated frames of abnormal temperature values of the cable terminal in Step 1, Step 2 of using a preset temperature warning policy to determine the temperature value of the current frame of the cable terminal. If it is abnormal, add 1 to counter F2. If it is normal, reset counter F2 to 0, After determining counter F2 in Step 2, if the number of consecutive integrated frames of counter F2 is greater than the frame number set value, conclude that the cable terminal is abnormal, and reset the counter to 0. If the number of consecutive integrated frames of counter F2 is less than the frame number set value, conclude that the cable terminal is normal in Step 3, Step 4 of repeating Steps 2 and 3 to determine the temperature values of subsequent frames of the cable terminal, and then obtaining a detection result of the cable terminal to be detected.

[0010] The present application further provides a method for detecting abnormalities in cable terminals of rail vehicles. In the step of performing analysis processing using a preset detection model based on the environmental temperature data and analysis data described above to obtain a detection result of the cable terminal to be detected, the analysis processing is performed using a detection model based on the analysis processing of environmental temperature data, image data, temperature data, and audio data, and the step of obtaining a detection result of the cable terminal to be detected is as follows: extracting the temperature feature of the image data; extracting the audio feature at the time corresponding to the image data in the audio data; performing normalization processing on the temperature feature and the audio feature to construct a fused feature of temperature and audio; including inputting the temperature feature, the audio feature, and the fused feature of temperature and audio into a preset cluster analysis model for detection to obtain a detection result of the cable terminal to be detected; Here, the preset cluster analysis model is obtained by training with normal temperature features, audio features, and fused features of temperature and audio, and abnormal temperature features, audio features, and fused features of temperature and audio as samples, and labels corresponding to using normal temperature features, audio features, and fused features of temperature and audio, and abnormal temperature features, audio features, and fused features of temperature and audio as samples.

[0011] The present application further provides a method for detecting abnormalities in cable terminals of rail vehicles. After obtaining a detection result of the cable terminal to be detected by performing analysis processing using a preset detection model based on the environmental temperature data and analysis data described above, the method further includes a step of positioning the detection result of the abnormality. In the step of positioning the detection result of the abnormality, when the detection result of the cable terminal to be detected is abnormal, based on the analysis data, the position of the failure of the abnormal cable terminal is obtained by using a method for positioning the abnormality of the cable terminal.

[0012] The present application further provides a method for detecting abnormalities in cable terminals of rail vehicles. When the detection result of the cable terminal to be detected, as described above, is abnormal, based on the analysis data of the cable terminal to be detected, using the method for positioning abnormalities in cable terminals, obtaining the location of the failure of the abnormal cable terminal is locking the location of the abnormal cable terminal based on the voice characteristics; and matching the abnormal cable terminal using the multi-point positioning method based on the image data of the cable terminal to be detected for abnormality detection and the cable terminal recognition model; and dividing the image area of the abnormal cable terminal after matching into a plurality of partitions, performing temperature determination on the abnormal cable terminal after division based on the temperature data, and obtaining the failure location of the abnormal cable terminal, and here, the cable terminal recognition model collects the design data of the cable terminal, and preliminarily creates a cable terminal recognition template based on the size of the design drawing in the design data. After training the preliminarily created cable terminal recognition template by matching at least 1000 actual contour images of cable terminals, the cable terminal recognition model is obtained.

[0013] The present application further provides an apparatus for detecting abnormalities in cable terminals of rail vehicles. a collection module configured to collect analysis data of the cable terminal to be detected; and a detection module configured to perform analysis processing using a preset detection model based on the analysis data and obtain a detection result of the cable terminal to be detected, and here, the analysis data includes environmental temperature data, image data of the cable terminal to be detected, temperature data, and voice data. The preset detection model includes a detection model based on the environmental temperature data, image data, and temperature data, or a detection model based on the analysis processing of the environmental temperature data, image data, temperature data, and voice data.

[0014] The present application further provides an abnormal detection device for a cable terminal of a rail vehicle. In the step where the detection module performs analysis processing using a preset detection model based on the environmental temperature data and analysis data to obtain a detection result of the cable terminal to be detected, the step of using a detection model based on analysis processing of the environmental temperature data, image data, and temperature data includes: extracting the temperature data of each frame in the temperature data; calibrating the position coordinate point range of the cable terminal to be detected in the image data; extracting the temperature value corresponding to the position coordinate point range based on the temperature data of each frame; performing processing and conversion on the temperature value based on the environmental temperature data; using a preset temperature warning policy to determine the temperature value of each frame after processing and conversion, and obtaining a detection result of the cable terminal to be detected.

[0015] The present application further provides an abnormal detection device for a cable terminal of a rail vehicle. The detection module includes a preset temperature warning policy, and the preset temperature warning policy includes one or more of an absolute temperature average value warning policy, an absolute temperature ratio value warning policy, a relative temperature average value warning policy, a relative temperature ratio value warning policy, a temperature difference average value warning policy, a temperature difference ratio value warning policy, a temperature rise rate average value warning policy, and a temperature rise rate ratio value warning policy.

Advantages of the Invention

[0016] The real-time detection method and device for abnormalities of cable terminals of rail vehicles according to the present application use a preset detection model to analyze and detect the collected environmental temperature data and analysis data including image data, temperature data, and audio data through the abnormality detection method of cable terminals of rail vehicles. By doing so, it fully considers situations such as environmental temperature and temperature changes, avoids the detection results being affected by the environmental temperature and the occurrence of reporting omissions and false alarms, thereby improving the accuracy rate of detection for cable terminals.

Brief Description of the Drawings

[0017] Hereinafter, to more clearly explain the technical means in the present application or the prior art, the drawings necessary for the description of the embodiments or the prior art will be briefly described. Of course, the drawings described below are only part of the embodiments of the present application, and those skilled in the art can obtain other drawings based on these drawings without creative labor.

[0018]

Figure 1

Figure 2

Figure 3

Modes for Carrying Out the Invention

[0019] Hereinafter, to more clearly clarify the objectives, technical means, and advantages of the present application, while referring to the drawings in the present application, the technical means in the present application will be clearly and completely described. Of course, the described embodiments are only part of the embodiments of the present application, not all of them. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the protection scope of the present application.

[0020] Example 1 The following describes the abnormality detection method for cable terminals of rail vehicles according to the present application with reference to FIG. 1. The method is as follows: S1 that collects environmental temperature data and analysis data of cable terminals to be detected, Based on the environmental temperature data and analysis data, performing analysis processing using a preset detection model to obtain a detection result of the cable terminal to be detected, including S2, Here, the analysis data includes image data, temperature data, and audio data.

[0021] Specifically, use an infrared camera to collect data for the entire cable terminal area, obtain temperature data of at least 60 frames of the cable terminal and image data of the collection area, and obtain environmental temperature data. Collect audio data with an audio sensor attached near the cable terminal. Thereby, ensure the completeness of temperature collection for the cable terminal.

[0022] The preset detection model includes a detection model based on analysis processing of the environmental temperature data, image data, and temperature data, or a detection model based on analysis processing of the environmental temperature data, image data, temperature data, and audio data.

[0023] This application uses a preset detection model to analyze and detect the collected environmental temperature data and analysis data including image data, temperature data, and audio data, fully considering situations such as environmental temperature and temperature changes, avoiding the occurrence of reporting omissions and false alarms in the detection results, and thereby improving the accuracy of detection for cable terminals.

[0024] In this embodiment, in step S2, using the preset temperature warning policy to determine the temperature value of each frame after processing and conversion, and obtaining the detection result of the cable terminal to be detected includes step 1, step 2, step 3, step 4, and step 5.

[0025] In step 1, extract the temperature data of each frame in the temperature data.

[0026] In step 2, calibrate the position coordinate point range of the cable terminal to be detected in the image data. For example, adopt image data of 100*100 pixel points, select any cable terminal block where the upper left coordinate point in the image data is (2, 5) and the lower right coordinate point is (3, 7), and it can be obtained that the coordinate points of the cable terminal are ((2, 5), (2, 6), (2, 7), (3, 5), (3, 6), (3, 7)).

[0027] In step 3, based on the temperature data of each frame, extract the temperature value corresponding to the position coordinate point range. The temperature value T is (34, 36, 35, 35, 36, 34).

[0028] In step 4, based on the environmental temperature data, perform processing and conversion on the temperature value. Here, the environmental temperature in the environmental temperature data is Tamb (for example, Tamb = 25°C).

[0029] In step 5, use the preset temperature alarm policy to determine the temperature value of each frame after processing and conversion, and obtain the detection result of the cable terminal to be detected.

[0030] Specifically, the method of performing processing and conversion on the temperature value is as follows.

[0031] The average temperature value Tave of the cable terminal area is the average value of the temperature values T corresponding to all coordinate points in the cable terminal area of the current frame. For example, (the temperature values corresponding to all coordinate points are 34, 36, 35, 35, 36, 34), and the average temperature value Tave of the cable terminal area = 35 can be obtained.

[0032] Regarding the relative temperature value Trel of the cable terminal area, Trel = T - Tamb is obtained by subtracting the current environmental temperature value Tamb from the temperature value T corresponding to all coordinate points in the cable terminal area of the current frame. For example, Tamb = 25 and Trel = (9, 11, 10, 10, 11, 9).

[0033] The average value Trel-ave of the relative value Trel of the cable terminal area temperature is the average value of the relative value Trel of the temperature corresponding to all coordinate points in the cable terminal area of the current frame. For example, Trel = (9, 11, 10, 10, 11, 9), and Trel-ave = 10.

[0034] Regarding the temperature difference Tdif of the cable terminal area, Tdif = T - T1 is obtained by subtracting the average temperature value T1 of other cable terminals from the temperature values corresponding to all coordinate points in the cable terminal area of the current frame. For example, T1 = (33, 35, 34, 34, 35, 33), and Tdif = (1, 1, 1, 1, 1, 1).

[0035] The average value Tdif-ave of the temperature difference Tdif of the cable terminal area is the average value of the temperature difference Tdif corresponding to all coordinate points in the cable terminal area of the current frame. For example, Tdif = (1, 1, 1, 1, 1, 1), and Tdif-ave = 1.

[0036] Regarding the temperature rise rate Trat of the cable terminal area, Trat = T - T0 is obtained by subtracting the temperature value T0 of the previous frame of the same cable terminal from the temperature values corresponding to all coordinate points in the cable terminal area of the current frame. For example, T0 = (30, 30, 30, 30, 30, 30), and Trat = (4, 6, 5, 5, 6, 4).

[0037] The average value Trat-ave of the temperature rise rate Trat of the cable terminal area is the average value of the temperature rise rate Trat corresponding to all coordinate points in the cable terminal area of the current frame. For example, Trat = (4, 6, 5, 5, 6, 4), and Trat-ave = 5.

[0038] Specifically, the preset temperature alarm policy includes one or more of an absolute temperature average value alarm policy, an absolute temperature ratio value alarm policy, a relative temperature average value alarm policy, a relative temperature ratio value alarm policy, a temperature difference average value alarm policy, a temperature difference ratio value alarm policy, a temperature rise rate average value alarm policy, and a temperature rise rate ratio value alarm policy.

[0039] 1) Absolute temperature average value alarm policy: The average value of the temperature Tave in the current frame is statistically calculated and compared with the set value Tave-max (for example, 100 °C). If Tave > Tave-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0040] 2) Absolute temperature ratio value alarm policy: All temperature values T in the current frame are compared with the set value Tmax. The ratio (P1 = (number of points where T > Tmax) / total number of points) of the number of temperature points exceeding the set value to the total number of points is statistically calculated, and P1 is compared with the set value P1-max (for example, 20%). If P1 > P1-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0041] 3) Relative temperature average value alarm policy: The relative temperature value Trel = T - Tamb in the current frame is calculated, and the average value Trel-ave of the relative temperature values in the current frame is calculated and compared with the set value Trel-ave-max (for example, 80 °C). If Trel-ave > Trel-ave-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0042] 4) Relative temperature ratio value alarm policy: The relative temperature value Trel = T - Tamb in the current frame is calculated. All relative temperature values Trel in the current frame are compared with the set value Trel-max. The ratio (P2 = (number of points where Trel > Trel-max) / total number of points) of the number of temperature points exceeding the set value to the total number of points is statistically calculated, and P2 is compared with the set value P2-max (for example, 20%). If P2 > P2-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0043] 5) Temperature difference average value alarm policy: Calculate the temperature difference (Tdif = T - T1) between the temperature T of the current frame and the temperature T1 of other cable terminals in the same train (take the average value if there are multiple). Calculate the average value Tdif-ave of the temperature differences of the current frame, compare it with the set value Tdif-ave-max (for example, 20°C). If Tdif-ave > Tdif-ave-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0044] 6) Temperature difference ratio value alarm policy: Calculate the temperature difference (Tdif = T - T1) between the temperature T of the current frame and the temperature T1 of other cable terminals in the same train (take the average value if there are multiple). Compare all the temperature differences Tdif of the current frame with the set value Tdif-max, count the ratio (P3 = (number of points where Tdif > Tdif-max) / total number of points) that the number of points exceeding the temperature difference set value occupies in the total number of points, and compare P3 with the set value P3-max (for example, 20%). If P3 > P3-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0045] 7) Average value of temperature rise rate alarm policy: Calculate the difference value (Trat = T - T0) between the temperature T of the current frame and the temperature T0 of the previous frame of the same cable terminal. Trat is the temperature rise rate of this cable terminal in the current frame. Calculate the average value Trat-ave of the temperature rise rates of the current frame, compare it with the set value Trat-ave-max (for example, 10°C). If Trat-ave > Trat-ave-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0046] 8) Heating rate percentage alarm policy: Calculate the difference value (Trat = T - T0) between the temperature T of the current frame and the temperature T0 of the frame before the same cable terminal. Trat is the heating rate of the current frame of this cable terminal. Compare all the heating rates Trat of the current frame with the set value Trat-max, and count the ratio (P4 = (number of points where Trat > Trat-max) / total number of points) that the number of points exceeding the heating rate set value occupies in the total number of points. Compare P4 with the set value P4-max (for example, 20%). If P4 > P4-max, temperature data indicating that the current frame is abnormal is obtained; otherwise, temperature data indicating that the current frame is normal is obtained.

[0047] The policy for determining the state of the current frame of the above-mentioned cable terminal may be used alone, that is, the state of the current frame is determined using one of them, or multiple policies may be combined and used. The combination method is not limited to the voting method (for example, if there is an alarm according to more than half of the policies, it is determined that the current frame is abnormal), the weighting method (for example, the weight of each policy is set to 12.5%), etc. Due to the alarm policy, situations such as environmental temperature, differences between different cable terminals, and abnormal temperature rise of cable terminals are fully considered, and the accuracy rate of the proposed cable terminal abnormality detection method is high.

[0048] In step 5, determine the number of frames of temperature data with abnormalities after the determination, and obtain the detection result of the cable terminal to be detected.

[0049] Here, determining the number of frames of temperature data with abnormalities after the determination and obtaining the detection result of the cable terminal to be detected includes step 1, step 2, step 3, and step 4.

[0050] In step 1, set the counter F2 and initialize the counter so that F2 = 0.

[0051] Here, the counter F2 represents the continuous integrated number of frames of abnormal temperature values of the cable terminal.

[0052] In step 2, using the preset temperature warning policy, determine the temperature value of the current frame of the cable terminal. If it is abnormal, add 1 to counter F2 and then use it as the counter for the next frame. If it is normal, reset counter F2 to 0.

[0053] In step 3, after determining the counter in step 2, if the continuous integrated frame number of counter F2 is greater than the frame number setting value F2-max (for example, F2-max is set to 10, 20, or 30), conclude that the cable terminal is abnormal and reset the counter to 0. If the continuous integrated frame number of counter F2 is less than the frame number setting value, conclude that the cable terminal is normal.

[0054] In step 4, after repeatedly performing steps 2 and 3 to determine the temperature values of subsequent frames of the cable terminal, obtain the detection result of the cable terminal to be detected.

[0055] The setting values in the above process may, on the one hand, be directly set based on expert experience. For example, they may be set as Tave-max = 100°C and Trel-ave-max = 80°C. On the other hand, they may also be obtained by performing statistics and analysis on the temperature data of a large number of cable terminals. If all the obtained temperature data is normal data, the 3σ limit value can be set as the setting value such as Tave-max or Trel-ave-max using the 3σ principle of the normal distribution. If the obtained temperature data contains abnormal data, based on the distributions of the normal data and abnormal data, the setting value (for example, normal temperature average value * 0.3 + abnormal temperature average value * 0.7) can be set between the normal data (for example, normal temperature average value) and abnormal data (for example, abnormal temperature average value).

[0056] In this embodiment, based on the above-mentioned environmental temperature data and analysis data, after performing analysis processing using the preset detection model and obtaining the detection result of the cable terminal to be detected, it further includes the step of positioning the abnormal detection result. In the step of positioning the abnormal detection result, when the detection result of the cable terminal to be detected is abnormal, based on the analysis data, using the abnormal positioning method of the cable terminal, obtain the position of the failure of the abnormal cable terminal.

[0057] Specifically, when the detection result of the cable terminal to be detected is abnormal as described above, obtaining the position of the failure of the abnormal cable terminal by using the abnormal positioning method of the cable terminal based on the analysis data of the cable terminal to be detected means locking the position of the abnormal cable terminal according to the voice feature; and matching the abnormal cable terminal by using a multi-point positioning method (for example, a 3-point positioning method) based on the image data of the cable terminal where the abnormality is to be detected and the cable terminal recognition model; and dividing the image area of the abnormal cable terminal after matching into a plurality of partitions (for example, 10 partitions), making a temperature determination for the abnormal cable terminal after division based on the temperature data, and obtaining the position of the abnormal partition of the cable terminal. Thereby, positioning for the failure of the abnormal cable terminal is realized according to the position of the abnormal partition of the cable terminal.

[0058] Here, the cable terminal recognition model collects the design data of the cable terminal, and preliminarily creates a cable terminal recognition template based on the size of the design drawing in the design data. For example, the design drawing shows coordinate points ((1, 2), (1, 3),... (2, 2)) corresponding to the outer contour curve of the cable terminal.

[0059] After matching the actual contour images of at least 1,000 cable terminals to train the preliminarily created cable terminal recognition template, the cable terminal recognition model is obtained. For example, an image of 100*100 pixels is adopted, coordinate points corresponding to the outer contour curve are drawn in the image, fine-tuning is performed based on the image, it is matched with the cable terminal in the image, and at least 1,000 images are matched and trained to obtain a cable terminal recognition model. Thereby, the actual contour of the cable terminal in the image is shown.

[0060] Example 2 Based on Example 1, in this example, the preset detection model in Example 1 is changed to a detection model based on the analysis processing of the environmental temperature data, image data, temperature data, and audio data.

[0061] Specifically, in the step of performing analysis processing using the preset detection model based on the above-mentioned environmental temperature data and analysis data to obtain the detection result of the cable terminal to be detected, the step of performing analysis processing using the detection model based on the analysis processing of the environmental temperature data, image data, temperature data, and audio data to obtain the detection result of the cable terminal to be detected includes the following.

[0062] Extract the temperature characteristics of the image data. For example, the temperature characteristics include the temperature average value Tave = 35, the temperature relative value Trel (Trel = T - Tamb = (9, 11, 10, 10, 11, 9)), the temperature difference Tdif (Tdif = T - T1 = (1, 1, 1, 1, 1, 1)), the temperature rise rate Trat (Trat = T - T0 = (4, 6, 5, 5, 6, 4)), etc.

[0063] Extract the voice features corresponding to the image data in the voice data at the corresponding time. For example, as the time domain features in the voice features, there are the average value of the voice amplitude within 1 s, the voice waveform index within 1 s, the voice pulse index within 1 s, the voice kurtosis index within 1 s, the voice margin index within 1 s, and the voice peak-to-peak value within 1 s. As the frequency domain features, there are the voice centroid frequency within 1 s, the root mean square frequency of the voice within 1 s, the short-time power spectral density of the voice within 1 s, the voice spectral entropy within 1 s, the voice formant within 1 s, etc.

[0064] Perform normalization processing on the temperature features and voice features. Thereby, the temperature features and voice features are respectively comprehensively transformed into the range from 0 to 1.

[0065] Construct the fusion features of temperature and voice. For example, perform weighted addition of the average value of the voice temperature and the average value of the voice amplitude within 1 s, perform weighted addition of the relative value of the voice temperature and the voice peak-to-peak value within 1 s, obtain the maximum value from the voice temperature difference and the voice waveform index within 1 s, obtain the average value from the voice temperature rise rate and the voice spectral entropy within 1 s, etc.

[0066] Input the temperature features, voice features, and the fusion features of temperature and voice into a preset cluster analysis model for detection, and obtain the detection result of the cable terminal to be detected. Here, the preset cluster analysis model may adopt the K-MEANS clustering algorithm, the mean shift clustering algorithm, the DBSCAN clustering algorithm, or the hierarchical clustering algorithm, etc.

[0067] Here, the preset cluster analysis model takes the normal temperature features, voice features, and the fusion features of temperature and voice, as well as the abnormal temperature features, voice features, and the fusion features of temperature and voice as samples, and is obtained by training with the labels corresponding to taking the normal temperature features, voice features, and the fusion features of temperature and voice, and the abnormal temperature features, voice features, and the fusion features of temperature and voice as samples.

[0068] The following describes an abnormal detection device for a cable terminal of a rail vehicle according to the present application. The abnormal detection device for a cable terminal of a rail vehicle described below and the above-described abnormal detection method for a cable terminal of a rail vehicle can be referred to each other.

[0069] The present application further provides an abnormal detection device for a cable terminal of a rail vehicle, including a collection module 210 and a detection module 220.

[0070] The collection module 210 is configured to collect environmental temperature data and analysis data of the cable terminal to be detected.

[0071] In this embodiment, the collection module includes an infrared camera and a voice sensor. By using the infrared camera to collect the temperature of the entire cable terminal area, the surface temperature of the cable terminal is obtained and the environmental temperature is collected. Furthermore, in order to ensure the completeness of detection, a plurality of infrared cameras may be installed at different angles and positions so as to obtain more complete temperature data of the semi-rigid terminal. The number of cameras is three or more, distributed on a circle with a distance from the cable terminal area greater than 20 cm and less than 100 cm, and the infrared cameras are uniformly distributed along the circumferential direction. The voice sensor is attached near the cable terminal on the roof of the rail vehicle to collect the voice emitted by the cable terminal during the running process. The number of voice sensors is three or more, distributed on a circle with a distance from the cable terminal area greater than 20 cm and less than 100 cm, and the voice sensors are uniformly distributed along the circumferential direction. After obtaining the analysis data, it is transmitted to the detection module via a wired or wireless transmission method.

[0072] The detection module 220 performs analysis processing using a preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected.

[0073] Here, the analysis data includes image data, temperature data, and voice data.

[0074] The preset detection model 220 includes a detection model based on analysis processing of the environmental temperature data, image data, and temperature data, or a detection model based on analysis processing of the environmental temperature data, image data, temperature data, and audio data. In this application, a collection module collects environmental temperature data and analysis data including image data, temperature data, and audio data of a cable terminal to be detected, and a detection module performs analysis processing using the preset detection model based on the analysis data to detect a detection result of the cable terminal to be detected. By fully considering the environmental temperature, differences between different devices, and situations such as temperature rise due to device failures, the accuracy of detecting the cable terminal is improved.

[0075] Here, in the step where the detection module performs analysis processing using the preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected, the step of using the detection model based on analysis processing of the environmental temperature data, image data, and temperature data includes: extracting the temperature data of each frame in the temperature data; calibrating the position coordinate point range of the cable terminal to be detected in the image data; extracting the temperature value corresponding to the position coordinate point range based on the temperature data of each frame; performing processing and conversion on the temperature value based on the environmental temperature data; using a preset temperature alarm policy to determine the temperature value of each frame after processing and conversion, and obtaining a detection result of the cable terminal to be detected.

[0076] Meanwhile, the detection module includes a preset temperature warning policy, and the preset temperature warning policy includes one or more of an absolute temperature average value warning policy, an absolute temperature ratio value warning policy, a relative temperature average value warning policy, a relative temperature ratio value warning policy, a temperature difference average value warning policy, a temperature difference ratio value warning policy, a temperature rise rate average value warning policy, and a temperature rise rate ratio value warning policy.

[0077] Figure 3 is a schematic diagram of the structure of the entity of the electronic device. As shown in Figure 3, the electronic device may include a processor 310, a communications interface 320, a memory 330, and a communication bus 340. Here, the processor 310, the communications interface 320, and the memory 330 complete their communications with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute the method for detecting abnormalities in the cable terminal of the rail vehicle. The method includes: S1: Collecting environmental temperature data and analysis data of the cable terminal to be detected; S2: Performing analysis processing using a preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected. Here, the analysis data includes image data, temperature data, and audio data.

[0078] Specifically, use an infrared camera to collect data for the entire cable terminal area, obtain temperature data of the cable terminal of at least 60 frames and image data of the collection area, and obtain environmental temperature data. Collect audio data with an audio sensor attached near the cable terminal. Thereby, ensure the completeness of temperature collection for the cable terminal.

[0079] In addition, the logical instructions in the memory 330 described above can be realized in the form of software function units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical means of the present application may be embodied in the form of a software product in terms of its essence, or the part contributing to the prior art, or all or part of the technical means. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The storage medium includes various media capable of storing program codes, such as a USB disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0080] On the other hand, the present application further provides a computer program product, the computer program product includes a computer program, the computer program may be stored in a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute the method for detecting abnormalities of the cable terminal of the rail vehicle according to each of the above methods, and the method includes S1 of collecting environmental temperature data and analysis data of the cable terminal to be detected; S2 of performing analysis processing using a preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected, and includes Here, the analysis data includes image data, temperature data, and audio data.

[0081] Specifically, use an infrared camera to collect data for the entire cable terminal area, obtain temperature data of the cable terminal for at least 60 frames and image data of the collection area, and obtain environmental temperature data. Collect audio data with an audio sensor attached near the cable terminal. Thereby, the integrity of temperature collection for the cable terminal is ensured.

[0082] On the other hand, the present application further provides a non-transitory computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is realized to execute the method for detecting abnormalities in the cable terminal of the rail vehicle according to each of the above methods. The method includes: S1: Collecting environmental temperature data and analysis data of the cable terminal to be detected; S2: Performing analysis processing using a preset detection model based on the environmental temperature data and the analysis data to obtain a detection result of the cable terminal to be detected. Here, the analysis data includes image data, temperature data, and audio data.

[0083] Specifically, use an infrared camera to collect data for the entire cable terminal area, obtain temperature data of the cable terminal for at least 60 frames and image data of the collection area, and obtain environmental temperature data. Collect audio data with an audio sensor attached near the cable terminal. Thereby, the integrity of temperature collection for the cable terminal is ensured.

[0084] The embodiments of the above-described device are merely exemplary. The units described as the above-mentioned separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed in a plurality of network units. According to actual needs, some or all of the modules can be selected to achieve the purpose of the technical means of this embodiment. Those skilled in the art can understand and implement it without creative labor.

[0085] From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment may be implemented in a form with the general-purpose hardware platform required for software, and of course, it may also be implemented in hardware. Based on such understanding, the above technical means may be embodied in the form of a software product in terms of its essence or the part that contributes to the prior art. The computer software product may be stored in a computer-readable storage medium such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, server, network device, etc.) to execute the method described in each embodiment or a certain part of the embodiment.

[0086] However, the above embodiments are merely for explaining the technical means of the present application and do not limit it. Although the present application has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that they can still modify the technical means described in each of the above embodiments or replace some technical features with equivalent ones, and these modifications or replacements do not deviate from the spirit and scope of the technical means of each embodiment of the present application in terms of the essence of the corresponding technical means.

Claims

1. A method for detecting an abnormality of a cable terminal of a rail vehicle, comprising: collecting ambient temperature data and analysis data of a cable terminal to be detected; performing analysis processing using a preset detection model based on the ambient temperature data and the analysis data to obtain a detection result of the cable terminal to be detected; wherein the analysis data includes image data, temperature data, and audio data; the preset detection model includes a detection model based on analysis processing of the ambient temperature data, image data, and temperature data, or a detection model based on analysis processing of the ambient temperature data, image data, temperature data, and audio data. A method for detecting an abnormality of a cable terminal of a rail vehicle, characterized in that it comprises the above.

2. In the step of performing analysis processing using a preset detection model based on the ambient temperature data and the analysis data to obtain a detection result of the cable terminal to be detected, the step of using a detection model based on analysis processing of the ambient temperature data, image data, and temperature data comprises: extracting temperature data of each frame in the temperature data; calibrating a position coordinate point range of the cable terminal to be detected in the image data; extracting a temperature value corresponding to the position coordinate point range based on the temperature data of each frame; performing processing and conversion on the temperature value based on the ambient temperature data; judging the temperature value of each frame after processing and conversion using a preset temperature alarm policy to obtain a detection result of the cable terminal to be detected. The method for detecting an abnormality of a cable terminal of a rail vehicle according to Claim 1.

3. The preset temperature alarm policy includes one or more of an absolute temperature average value alarm policy, an absolute temperature ratio value alarm policy, a relative temperature average value alarm policy, a relative temperature ratio value alarm policy, a temperature difference average value alarm policy, a temperature difference ratio value alarm policy, a temperature rise rate average value alarm policy, and a temperature rise rate ratio value alarm policy. The method for detecting an abnormality of a cable terminal of a rail vehicle according to Claim 2.

4. The above-mentioned step of judging the temperature value of each frame after processing and conversion using a preset temperature alarm policy to obtain a detection result of the cable terminal to be detected is: Step 1 of setting counter F2 and initializing the counter so that F2 = 0, where the counter F2 represents the number of consecutive integration frames of the abnormal temperature value of the cable terminal Step 1 of using a preset temperature alarm policy to determine the temperature value of the current frame of the cable terminal. If it is abnormal, add 1 to counter F2. If it is normal, reset counter F2 to 0 After determining counter F2 in Step 2, if the number of consecutive integration frames of counter F2 is greater than the frame number set value, conclude that the cable terminal is abnormal and reset the counter to 0. If the number of consecutive integration frames of counter F2 is less than the frame number set value, conclude that the cable terminal is normal Step 3 of repeating Steps 2 and 3 to determine the temperature values of subsequent frames of the cable terminal, and then obtaining the detection result of the cable terminal to be detected The abnormal detection method of the cable terminal of the rail vehicle according to claim 2

5. In the step of performing analysis processing using a preset detection model based on the environmental temperature data and analysis data described above to obtain the detection result of the cable terminal to be detected, the step of performing analysis processing using a detection model based on the analysis processing of environmental temperature data, image data, temperature data, and audio data to obtain the detection result of the cable terminal to be detected includes Extracting the temperature feature of the image data Extracting the audio feature of the time corresponding to the image data in the audio data Performing normalization processing on the temperature feature and the audio feature to construct a fusion feature of temperature and audio Inputting the temperature feature, audio feature, and fusion feature of temperature and audio into a preset cluster analysis model for detection to obtain the detection result of the cable terminal to be detected, where the preset cluster analysis model is trained with normal temperature features, audio features, and fusion features of temperature and audio, and abnormal temperature features, audio features, and fusion features of temperature and audio as samples, and labels corresponding to using normal temperature features, audio features, and fusion features of temperature and audio, and abnormal temperature features, audio features, and fusion features of temperature and audio as samples It is characterized in that it is obtained ​ Method for detecting abnormality of cable terminal of rail vehicle according to claim 1.

6. Further comprising the step of performing analysis processing using a preset detection model based on the environmental temperature data and the analysis data, obtaining a detection result of the cable terminal to be detected, and then positioning the detection result of the abnormality. In the step of positioning the detection result of the abnormality, when the detection result of the cable terminal to be detected is abnormal, based on the analysis data, using the cable terminal abnormality positioning method to obtain the position of the failure of the abnormal cable terminal. Method for detecting abnormality of cable terminal of rail vehicle according to any one of claims 1 to 5.

7. When the detection result of the cable terminal to be detected is abnormal, obtaining the position of the failure of the abnormal cable terminal based on the analysis data of the cable terminal to be detected and using the cable terminal abnormality positioning method. Locking the position of the abnormal cable terminal by the voice feature. Based on the image data of the cable terminal to be detected for abnormality and the cable terminal recognition model, using the multi-point positioning method to match the abnormal cable terminal. Dividing the image area of the abnormal cable terminal after matching into a plurality of partitions, performing temperature determination on the abnormal cable terminal after division based on the temperature data, and obtaining the failure position of the abnormal cable terminal. Here, the cable terminal recognition model collects the design data of the cable terminal, and preliminarily creates a cable terminal recognition template based on the size of the design drawing in the design data. Characterized in that after training the preliminarily created cable terminal recognition template by matching at least 1000 actual contour images of the cable terminal, the cable terminal recognition model is obtained. Method for detecting abnormality of cable terminal of rail vehicle according to claim 6.

8. An abnormal detection device for a cable terminal of a rail vehicle, comprising: A collection module configured to collect environmental temperature data and analysis data of a cable terminal to be detected. A detection module configured to perform analysis processing using a preset detection model based on the environmental temperature data and the analysis data, and obtain a detection result of the cable terminal to be detected. Here, the analysis data includes image data, temperature data, and audio data. The preset detection model includes a detection model based on analysis processing of the environmental temperature data, image data, and temperature data, or a detection model based on analysis processing of the environmental temperature data, image data, temperature data, and audio data. An abnormal detection device for a cable terminal of a rail vehicle is characterized in that it includes the above.

9. In the step where the detection module performs analysis processing using a preset detection model based on the environmental temperature data and analysis data to obtain a detection result of the cable terminal to be detected, the step of using a detection model based on analysis processing of the environmental temperature data, image data, and temperature data includes: extracting the temperature data of each frame in the temperature data; calibrating the position coordinate point range of the cable terminal to be detected in the image data; extracting the temperature value corresponding to the position coordinate point range based on the temperature data of each frame; performing processing and conversion on the temperature value based on the environmental temperature data; using a preset temperature warning policy to determine the temperature value of each frame after processing and conversion, and obtaining a detection result of the cable terminal to be detected. The abnormal detection device for a cable terminal of a rail vehicle according to claim 8.

10. The detection module includes a preset temperature warning policy, and the preset temperature warning policy includes one or more of an absolute temperature average value warning policy, an absolute temperature ratio value warning policy, a relative temperature average value warning policy, a relative temperature ratio value warning policy, a temperature difference average value warning policy, a temperature difference ratio value warning policy, a temperature rise rate average value warning policy, and a temperature rise rate ratio value warning policy. The abnormal detection device for a cable terminal of a rail vehicle according to claim 8 or 9.

Images