Electric vehicle charging cable quality detection method and system
Through the combination of infrared temperature sensor array and vision acquisition device, comprehensive abnormality monitoring and quality evaluation of charging cables are achieved, and the problem of insufficient detection of a single data source in the prior art is solved, and the accuracy and safety of detection are improved.
Patent Information
- Application Number
- CN202410845582.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing charging cable quality detection methods usually rely only on a single data source, and cannot comprehensively and accurately evaluate the quality status of the cable, which poses safety risks.
By laying an infrared temperature sensor array and a visual acquisition device, combining multiple temperature acquisition nodes and image acquisition line segments, temperature and image abnormality recognition is performed, overlapping abnormal nodes and detection results are obtained, and the quality detection level is provided in a comprehensive analysis.
It improves the comprehensiveness and accuracy of abnormal detection, ensures the safety, reliability and stability of the charging cable, reduces the potential failure risk, and provides a better charging experience.
Smart Images

Figure CN118914227A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of quality detection, and in particular to a quality detection method and system for electric vehicle charging cables. Background Art
[0002] Electric vehicle charging cables are one of the key components to ensure power transmission and safe charging. Unqualified charging cables may cause safety problems such as abnormal current transmission, overheating, and even fire. With the popularization of electric vehicles and the increase in market demand, the quality inspection of charging cables has become particularly important. However, the charging cable quality inspection methods in the existing technology usually only rely on a single data source for analysis, which cannot comprehensively and accurately evaluate the quality status of the cable, resulting in safety hazards.
[0003] Therefore, developing an efficient and accurate charging cable quality detection method is crucial to ensure the safety and reliability of the charging process. Summary of the invention
[0004] The present application provides an electric vehicle charging cable quality detection method and system, aiming to solve the technical problem that traditional charging cable quality detection methods usually only rely on a single data source for analysis and cannot comprehensively and accurately evaluate the quality status of the cable.
[0005] In view of the above problems, the present application provides a method and system for detecting the quality of electric vehicle charging cables.
[0006] In a first aspect disclosed in the present application, a method for detecting the quality of an electric vehicle charging cable is provided, the method comprising: laying out an infrared temperature sensor array on the surface of a target cable, and establishing a plurality of temperature acquisition nodes according to the sensor laying positions; energizing the target cable, and continuously acquiring temperature data of the plurality of temperature acquisition nodes through the infrared temperature sensor array to obtain a temperature data set within a preset time; identifying temperature anomalies based on the temperature data set to obtain temperature anomaly nodes and temperature anomaly detection results; dividing the target cable into a plurality of image acquisition segments according to the plurality of temperature acquisition nodes; using a visual acquisition device to sequentially acquire images of the plurality of image acquisition segments to obtain a cable image set, the cable image set comprising image anomaly nodes and image anomaly detection results; repeatedly identifying the temperature anomaly nodes and the image anomaly nodes to obtain overlapping anomaly nodes and overlapping anomaly detection results, wherein the overlapping anomaly detection results are detection results corresponding to the same anomaly; and obtaining quality detection levels at different positions of the target cable according to the temperature anomaly detection results, the image anomaly detection results, and the overlapping anomaly detection results.
[0007] Another aspect disclosed in the present application provides an electric vehicle charging cable quality detection system, the system is used for the above method, the system includes: a collection node acquisition module, the collection node acquisition module is used to lay out an infrared temperature sensor array on the surface of the target cable, and establish multiple temperature collection nodes according to the sensor layout position; a temperature data acquisition module, the temperature data acquisition module is used to power on the target cable, and continuously collect temperature data of the multiple temperature collection nodes through the infrared temperature sensor array to obtain a temperature data set within a preset time; a temperature anomaly recognition module, the temperature anomaly recognition module is used to identify temperature anomalies based on the temperature data set, and obtain temperature anomaly nodes and temperature anomaly detection results; a collection line segment acquisition module, the collection line segment acquisition module is used to According to the multiple temperature acquisition nodes, the target cable is divided into multiple image acquisition segments; a cable image acquisition module, the cable image acquisition module is used to use a visual acquisition device to sequentially acquire images of the multiple image acquisition segments to acquire a cable image set, wherein the cable image set includes image abnormality nodes and image abnormality detection results; a repeated recognition module, the repeated recognition module is used to repeatedly recognize the temperature abnormality nodes and the image abnormality nodes to acquire overlapping abnormality nodes and overlapping abnormality detection results, wherein the overlapping abnormality detection results are detection results corresponding to the same abnormality; a detection level acquisition module, the detection level acquisition module is used to acquire the quality detection level of different positions of the target cable according to the temperature abnormality detection result, the image abnormality detection result and the overlapping abnormality detection result.
[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0009] By deploying an infrared temperature sensor array and using a visual acquisition device, combined with multiple temperature acquisition nodes and image acquisition segments, a comprehensive monitoring of the temperature anomaly and image anomaly of the cable is achieved. By comprehensively analyzing the abnormal information of multiple data sources, the comprehensiveness and accuracy of anomaly detection are improved; by repeatedly identifying temperature anomaly nodes and image anomaly nodes, overlapping abnormal nodes and corresponding overlapping abnormal detection results are obtained, which can more accurately determine the situation where the same location has abnormalities in both temperature and image, further improving the accuracy and reliability of abnormality discrimination. In summary, this method can effectively detect the quality problems of electric vehicle charging cables, and provide quality detection levels for cables at different locations based on the comprehensive results of various abnormal detections, which helps to improve the safety, reliability and stability of charging cables, reduce potential failure risks, and provide users with a better charging experience.
[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Figure 1 A flow chart of a method for detecting the quality of an electric vehicle charging cable is provided for an embodiment of the present application;
[0012] Figure 2 A schematic diagram of the structure of an electric vehicle charging cable quality detection system is provided for an embodiment of the present application.
[0013] Explanation of reference numerals: acquisition node acquisition module 10 , temperature data acquisition module 20 , temperature anomaly identification module 30 , acquisition line segment acquisition module 40 , cable image acquisition module 50 , repeat identification module 60 , detection level acquisition module 70 . DETAILED DESCRIPTION
[0014] The embodiment of the present application solves the technical problem that the charging cable quality detection method usually relies on only a single data source for analysis and cannot comprehensively and accurately evaluate the quality status of the cable by providing an electric vehicle charging cable quality detection method.
[0015] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.
[0016] Embodiment 1
[0017] like Figure 1 As shown, the embodiment of the present application provides a method for detecting the quality of an electric vehicle charging cable, the method comprising:
[0018] An infrared temperature sensor array is arranged on the surface of the target cable, and multiple temperature collection nodes are established according to the sensor arrangement positions;
[0019] On the surface of the target cable, by fixing a clamp or pasting, according to the characteristics of the target cable, an infrared temperature sensor array is arranged at a suitable position on the cable surface to measure the temperature of the object surface in real time. The arrangement position includes the coverage range of the entire cable or the possible position of temperature anomaly. According to the arrangement position of the infrared temperature sensor array, the cable is divided into multiple parts, each part corresponds to a temperature collection node, and each temperature collection node has at least one infrared temperature sensor.
[0020] Powering on the target cable, continuously collecting temperature data from the multiple temperature collection nodes through the infrared temperature sensor array, and acquiring a set of temperature data within a preset time;
[0021] Ensure that the target cable is properly connected to the power supply, and turn on the power supply to allow current to flow through the cable. The infrared temperature sensor array is arranged on the surface of the target cable to continuously measure and record the surface temperature of each temperature acquisition node. The preset time is a time period set according to the temperature change of the target cable, such as 30 minutes. The temperature change data of multiple temperature acquisition nodes within the preset time is obtained, and the temperature data is aggregated into a data set, which includes the temperature values of multiple temperature acquisition nodes at different time points.
[0022] Perform temperature anomaly identification based on the temperature data set to obtain temperature anomaly nodes and temperature anomaly detection results;
[0023] Based on the acquired temperature data set, temperature anomaly identification is performed. For example, statistical analysis is performed on the temperature data set, such as calculating indicators such as the average temperature and standard deviation, and judging whether there are abnormal temperature values exceeding a preset range based on the temperature range under normal conditions; or based on experience or expert knowledge, a threshold is set to judge whether there is a temperature anomaly, and temperature data exceeding the preset threshold is considered abnormal.
[0024] After temperature anomaly identification, the temperature anomaly nodes and corresponding temperature anomaly detection results are determined, where the temperature anomaly node refers to the location where abnormal temperature is found on the cable; the temperature anomaly detection results include information such as the coordinates of the abnormal node, the specific temperature value, and the degree of abnormality, thereby providing quality detection levels at different locations of the target cable.
[0025] Dividing the target cable into a plurality of image acquisition line segments according to the plurality of temperature acquisition nodes;
[0026] According to the established multiple temperature acquisition nodes, the position of each temperature acquisition node on the cable is determined, so as to divide the target cable into multiple image acquisition segments, each image acquisition segment corresponding to the area between two adjacent temperature acquisition nodes on the cable.
[0027] Use a visual acquisition device to sequentially acquire images of the plurality of image acquisition line segments to obtain a cable image set, wherein the cable image set includes image abnormality nodes and image abnormality detection results;
[0028] Align the visual acquisition device to the position of the current image acquisition line segment, start the visual acquisition device, and perform image acquisition on the current image acquisition line segment to ensure that high-quality images covering the entire cable area are acquired. By moving the cable, each image acquisition line segment is imaged in sequence. For the acquired images, image processing and analysis algorithms are used to identify and detect abnormal nodes, including detecting damage, foreign matter, breakage, etc. on the cable surface. Each acquired image and its corresponding abnormal node and abnormal detection result are associated to generate a cable image set, which contains multiple images and the abnormal node position and detection results associated with each image.
[0029] Repeatedly identify the temperature abnormality node and the image abnormality node to obtain overlapping abnormality nodes and overlapping abnormality detection results, wherein the overlapping abnormality detection results are detection results corresponding to the same abnormality;
[0030] Compare the temperature abnormal nodes and image abnormal nodes, and find the overlapping abnormal nodes, which represent the positions that are identified as abnormal in both the temperature data set and the image acquisition and analysis. For the overlapping abnormal nodes, the temperature detection results and image detection results corresponding to the abnormal nodes are associated to form the overlapping abnormal detection results, so that the corresponding relationship between different representations of the same abnormality can be determined. This helps to further analyze the abnormal situation of the cable and provide more comprehensive and accurate quality problem detection information.
[0031] The quality detection levels of different positions of the target cable are obtained according to the temperature anomaly detection results, the image anomaly detection results, and the overlap anomaly detection results.
[0032] According to the detection results of temperature anomaly nodes, each abnormal node is compared with the preset abnormal temperature range, and the severity of the abnormality is judged according to the degree and duration of the abnormal temperature exceeding. According to the detection results of image anomaly nodes, the size, shape, color and other characteristics of the abnormal area are analyzed, and the severity of the abnormal situation is evaluated by comparing with the preset abnormal standards. Special attention is paid to the abnormal nodes that overlap with the temperature anomaly detection results and the image anomaly detection results. The overlapping anomaly detection results contain comprehensive information of temperature anomaly and image anomaly, so they can be regarded as more significant abnormal situations.
[0033] Based on the above analysis, the quality inspection levels are divided for different locations of the target cable. For example, according to comprehensive factors such as the number of abnormal nodes, severity, and impact range, different locations are divided into different quality levels, such as excellent, qualified, defective, etc.
[0034] Furthermore, using a visual acquisition device to sequentially acquire images of the plurality of image acquisition line segments to obtain a cable image set includes:
[0035] The visual acquisition device includes a first acquisition device and a second acquisition device, the second acquisition device has a higher resolution than the first acquisition device, and the first acquisition device and the second acquisition device are arranged adjacent to each other in front and back and are in communication connection;
[0036] Performing image acquisition on the first image acquisition line segment by the first acquisition device to obtain a first image acquisition result;
[0037] Performing abnormality recognition on the first image acquisition result to obtain a first abnormality recognition result;
[0038] When the first abnormality recognition result is that there is no abnormality, generating a rotation instruction, and controlling the target cable to rotate a preset angle according to the rotation instruction;
[0039] After the rotation is in place, the first image acquisition line segment is imaged by the first acquisition device, and abnormality recognition is performed to generate a second abnormality recognition result;
[0040] When the second abnormality recognition result is that there is no abnormality, a translation instruction is generated, and the target cable is controlled to translate to the second image acquisition line segment according to the translation instruction;
[0041] The first acquisition device performs image acquisition on the second image acquisition line segment until all target cables are traversed.
[0042] The first acquisition device is a part of the visual acquisition device, which is used to acquire images of a specific area of the target cable. It has a lower resolution and can cover a relatively large area. The second acquisition device is also a visual acquisition device. Compared with the first acquisition device, it has a higher resolution and can capture more detailed details and image information of the specific area of the target cable.
[0043] The first acquisition device and the second acquisition device are arranged adjacent to each other, which means that they are close to each other in space and arranged continuously in position. This arrangement can ensure that the two acquisition devices can cover the same cable area in sequence. The first acquisition device and the second acquisition device can exchange information through communication connection, and then comprehensively utilize the advantages of both to achieve comprehensive and high-resolution image acquisition and analysis of the target cable area.
[0044] According to the divided image acquisition segments, a first image acquisition segment is selected on the target cable as the acquisition area. The segment corresponds to a first acquisition device. The first acquisition device is started to perform image acquisition, and the image of the cable in the area is obtained to obtain a first image acquisition result.
[0045] An image processing and analysis algorithm is used to identify abnormal areas in the first image acquisition result, and a first abnormality identification result is generated. The result can mark abnormal area information to indicate possible problems in the image acquisition result.
[0046] The first abnormality recognition result is checked to determine whether there is an abnormality. If the result shows that there is no abnormality, a rotation instruction is generated, and the rotation instruction includes parameters such as rotation direction (clockwise or counterclockwise), setting rotation speed and angle, and sending the generated rotation instruction to the cable control system. According to the received rotation instruction, the cable control system controls the target cable to rotate according to a preset angle and reach the required rotation angle, such as 180°.
[0047] When the target cable rotates to a preset angle position according to the previous rotation instruction, the first acquisition device is started to acquire an image of the first image acquisition line segment on the target cable that has rotated into position. The same abnormality recognition algorithm as before is used to perform abnormality recognition on the image acquisition result to generate a second abnormality recognition result. These results can mark new abnormal areas and provide more comprehensive and accurate information.
[0048] The second abnormality recognition result is checked to determine whether there is an abnormality. If the result still shows that there is no abnormality, it means that there is no abnormality in the first image acquisition segment. A translation instruction is generated, including parameters such as the cable translation direction (left or right), translation speed, and translation distance. The generated translation instruction is sent to the cable control system. According to the received translation instruction, the cable control system controls the target cable to translate in the specified translation direction and distance and reach the required translation position, that is, the second image acquisition segment.
[0049] The first acquisition device is started to acquire images of the target cable on the second image acquisition line segment, and the above process is repeated, gradually traversing various parts of the target cable until all the target cables are traversed, thereby ensuring that complete image information is obtained.
[0050] Furthermore, performing abnormality recognition on the first image acquisition result to obtain a first abnormality recognition result includes:
[0051] The first image acquisition result includes a target cable area and a background area, and the target cable area is cropped to obtain an image of the target cable area;
[0052] Performing grayscale conversion on the target cable area image to obtain a grayscale image;
[0053] Establishing a preset grayscale interval according to the local features of the target cable area image, and performing binary classification on the grayscale image based on the preset grayscale interval to generate a binary image, wherein the area that does not meet the preset grayscale interval is marked as abnormal;
[0054] A first abnormality recognition result is generated according to the binary image.
[0055] The first image acquisition result is processed using image processing technology, such as image segmentation, to identify the target cable area, and the target cable area is cropped from the original image according to its boundary information and position to obtain an image of the target cable area, which only contains the target cable area.
[0056] For the target cable area image, the color information is removed and only the brightness information is retained to convert it from a color to a grayscale image. For example, the values of the red, green, and blue channels of each pixel in the color image are weighted averaged to obtain the corresponding grayscale value. After the grayscale conversion, a grayscale image containing only brightness information is obtained. The image no longer contains color, and the value of each pixel represents the brightness intensity of the corresponding position.
[0057] For the target cable area image, extract the brightness distribution features, for example, count the grayscale values, take the interval with the most grayscale values as the brightness distribution feature, and establish a preset grayscale interval based on the extracted local features to define the normal range. According to the preset grayscale interval, perform binary classification on the grayscale image, set the pixels that fall within the preset grayscale interval to one value, such as white or 1; and set the pixels that do not meet the interval conditions to another value, such as black or 0. According to the binarization results, the pixels marked as black or 0 are regarded as abnormal.
[0058] A first abnormality recognition result is obtained according to the number of abnormal regions, the size of the regions, and the abnormal positions in the binary image.
[0059] Furthermore, it also includes:
[0060] When the first abnormality identification result is that an abnormality exists, generating an abnormality capture instruction;
[0061] Acquire an abnormal center and an abnormal range according to the first abnormality recognition result, and establish a first abnormal area according to the abnormal center and the abnormal range;
[0062] activating a second acquisition device based on the abnormality capture instruction, focusing and acquiring the first abnormal area through the second acquisition device to obtain a first abnormal image;
[0063] Anomaly detection is performed on the first abnormal image to obtain a first image abnormality detection result.
[0064] The first abnormality identification result is checked to confirm whether there is an abnormality. If the result shows that there is an abnormality, an abnormality capture instruction is generated. The abnormality capture instruction is sent to the second acquisition device through a communication connection to activate the second acquisition device to perform further fine acquisition of the abnormal position.
[0065] According to the first anomaly recognition result, the center position and range of the abnormal area are extracted. The abnormal center indicates the position of the anomaly, and the abnormal range indicates the size or boundary of the abnormal area. The first abnormal area is determined by defining a circular area or a rectangular area with the abnormal center as the center point and the abnormal range as the radius. This can more accurately locate and describe the area where the abnormality exists.
[0066] Based on the received abnormal capture instruction, the second acquisition device is activated, including starting its work, adjusting relevant parameters and settings, etc., so that it is ready to focus on the first abnormal area. The first abnormal area is focused on by the second acquisition device, which has a higher resolution to obtain a more detailed and accurate image of the abnormal area. After completing the focused acquisition of the first abnormal area, the first abnormal image is obtained, which shows the details and features of the first abnormal area, which is helpful for further analysis and processing of the abnormal situation.
[0067] Select a suitable anomaly detection algorithm, such as establishing a twin model, process the first abnormal image, and mark the abnormal area as abnormal. Based on the result of the abnormality detection processing, generate the first image abnormality detection result, including the location, shape, size or other descriptive information of the abnormal area, for further analysis and processing.
[0068] Furthermore, establishing the first abnormal area includes:
[0069] Establishing a coordinate system according to the target cable position, and obtaining the abnormal center coordinates of the first abnormality recognition result in the coordinate system;
[0070] Positioning and expanding the coordinates of the center of the abnormality according to the abnormal range to obtain a positioning and expansion result;
[0071] According to the aspect ratio of the second acquisition device, a rectangular expansion constraint is established, and the positioning expansion result is limited according to the rectangular expansion constraint to obtain the first abnormal area.
[0072] According to the known target cable position information, a rectangular coordinate system is used for positioning and measurement. For the abnormal area of the first abnormal identification result, an image processing method is used, such as contour extraction, center of gravity calculation or feature point positioning of the abnormal area in the image, to determine the abnormal center coordinates of the area in the established coordinate system as the position representation of the abnormal area in the established coordinate system.
[0073] According to the size and shape of the anomaly range, positioning expansion is performed around the coordinates of the anomaly center. For example, according to the maximum radius of the anomaly range, a rectangle, circle or other shape is defined around the coordinates of the anomaly center. According to the positioning expansion method, the positioning expansion result is obtained, which can give a wider area range and cover the position near the original anomaly center, so as to more accurately understand the distribution of the anomaly.
[0074] Determine the aspect ratio of the second acquisition device, that is, the ratio of width to height. Based on the aspect ratio, establish a rectangular expansion constraint to limit the shape of the abnormal area. Use the rectangular expansion constraint to compare and limit the positioning expansion results so that the expansion results are within the range that meets the rectangular expansion constraint. The positioning expansion results that meet the rectangular expansion constraint are used as the first abnormal area, which includes the expansion position around the original abnormal center and is restricted by the rectangular expansion constraint.
[0075] Furthermore, performing abnormality detection on the first abnormal image to obtain an abnormality detection result of the first image includes:
[0076] Obtaining an abnormality detection record of the charging cable, extracting a sample abnormal image, and marking the sample abnormal image as abnormal, and obtaining a sample image abnormality marking result;
[0077] Based on the twin network, an anomaly detection model is constructed, wherein the anomaly detection model includes a first convolutional neural network branch and a second convolutional neural network branch, wherein the input of the first convolutional neural network branch is a sample abnormal image, and the input of the second convolutional neural network branch is the first abnormal image, and the two branches are the same and share parameters;
[0078] Using the first abnormal image, the sample abnormal image and the sample image abnormality identification result as training data, training the abnormality detection model until the convergence requirement is met, and obtaining the abnormality detection model;
[0079] The first abnormal image and the sample abnormal image are similarly calculated according to the abnormality detection model, and a first image abnormality detection result of the first abnormal image is obtained according to the maximum similarity.
[0080] Obtain the anomaly detection records of the same model of charging cables, which contain the images collected when performing anomaly detection on the charging cables. Extract sample images with anomalies from the anomaly detection records, which contain different parts of the charging cables and different anomaly types. Process the sample anomaly images through manual identification or anomaly detection algorithms, such as threshold and outlier detection methods, to identify the abnormal area, abnormality type, abnormality level, etc. as the sample image anomaly identification result.
[0081] Design and define the structure of the twin network, which consists of two convolutional neural network branches. The two branches have the same network structure and parameters, so that the two branches can share information and feature representation during the learning process, improving the robustness and generalization ability of the model.
[0082] The sample abnormal image is taken as input and processed by the first convolutional neural network branch, which is responsible for extracting features from the sample abnormal image and learning abnormal patterns; the first abnormal image is taken as input and processed by the second convolutional neural network branch, which is also responsible for extracting features from the first abnormal image and learning abnormal patterns. Based on the twin network structure and shared parameters, an anomaly detection model is constructed, which has twin network architecture, number of layers, activation function and other components.
[0083] By constructing an anomaly detection model based on the twin network, the shared parameters and feature representations in the sample anomaly image and the first anomaly image can be used for anomaly detection.
[0084] The first abnormal image and the sample abnormal image are used as input, and the sample image abnormal identification result is used as a label for training the model. The training data set is used to iteratively train the anomaly detection model, and the weights and parameters of the model are updated through back propagation and optimization algorithms such as gradient descent. In each training iteration, the model output is compared with the label, the loss function is calculated, and the parameters are adjusted. The indicators in the training process, such as loss function, accuracy, etc., are monitored until the model performs stably on the validation set and meets the convergence requirements, that is, the model can predict anomalies well. After the training process is completed, the trained anomaly detection model is obtained, which can be used to detect anomalies on new samples and identify and mark anomalies.
[0085] The first abnormal image and the sample abnormal images are respectively input into the abnormality detection model, and the similarity between the first abnormal image and each sample abnormal image is calculated, for example, using cosine similarity, Euclidean distance, etc. as a similarity measurement method, and the maximum similarity value is selected from the similarities of all sample abnormal images, which indicates that one of the sample abnormal images is most similar to the first abnormal image. According to the sample abnormal image corresponding to the maximum similarity value, the corresponding sample image abnormality identification result is obtained as the first image abnormality detection result of the first abnormal image.
[0086] Furthermore, repeatedly identifying the temperature abnormal node and the image abnormal node to obtain the coincident abnormal node and the coincident abnormal detection result includes:
[0087] Obtaining the coordinates of the temperature abnormal node and the image abnormal node, and marking the nodes with the same coordinates as overlapping abnormal nodes;
[0088] For the overlap abnormal node, the temperature anomaly detection result and the image anomaly detection result are superimposed to obtain the overlap abnormality detection result.
[0089] The coordinates of the temperature anomaly nodes in the established coordinate system are obtained, and these coordinates represent the temperature anomaly positions detected in a specific area. Similarly, the coordinates of the image anomaly nodes are obtained, and these coordinates represent the positions of the abnormal areas detected on the image. The temperature anomaly node coordinates are compared with the image anomaly node coordinates. If the coordinates are the same, they are marked as coincident anomaly nodes to better understand the correlation between the two data and obtain more accurate anomaly information.
[0090] The temperature anomaly detection results and image anomaly detection results associated with the coincident abnormal nodes are obtained and superimposed, for example, the temperature anomaly values and image anomaly areas are superimposed and displayed on the same image, or they are combined into a comprehensive anomaly score. Using the superimposed coincident anomaly detection results, the characteristics and properties of the coincident abnormal nodes can be more comprehensively understood, and more comprehensive and accurate anomaly detection results can be obtained.
[0091] In summary, the electric vehicle charging cable quality detection method and system provided in the embodiments of the present application have the following technical effects:
[0092] 1. By deploying infrared temperature sensor arrays and using visual acquisition devices, combined with multiple temperature acquisition nodes and image acquisition segments, comprehensive monitoring of cable temperature anomalies and image anomalies is achieved. By comprehensively analyzing the abnormal information of multiple data sources, the comprehensiveness and accuracy of anomaly detection are improved;
[0093] 2. By repeatedly identifying temperature anomaly nodes and image anomaly nodes, we can obtain overlapping anomaly nodes and corresponding overlapping anomaly detection results. This can make it more certain that the same location has abnormalities in both temperature and image, further improving the accuracy and reliability of anomaly discrimination.
[0094] In summary, this method can effectively detect quality problems of electric vehicle charging cables and provide quality detection levels for cables in different locations by integrating various abnormal detection results. This helps to improve the safety, reliability and stability of charging cables, reduce potential failure risks, and provide users with a better charging experience.
[0095] Embodiment 2
[0096] Based on the same inventive concept as the electric vehicle charging cable quality detection method in the aforementioned embodiment, Figure 2 As shown, the present application provides an electric vehicle charging cable quality detection system, the system comprising:
[0097] A collection node acquisition module 10, wherein the collection node acquisition module 10 is used to arrange an infrared temperature sensor array on the surface of the target cable and establish a plurality of temperature collection nodes according to the sensor arrangement positions;
[0098] A temperature data acquisition module 20, wherein the temperature data acquisition module 20 is used to energize the target cable, continuously acquire temperature data of the plurality of temperature acquisition nodes through the infrared temperature sensor array, and obtain a set of temperature data within a preset time;
[0099] A temperature anomaly identification module 30, wherein the temperature anomaly identification module 30 is used to identify temperature anomalies based on the temperature data set, and obtain temperature anomaly nodes and temperature anomaly detection results;
[0100] An acquisition line segment acquisition module 40, wherein the acquisition line segment acquisition module 40 is used to divide the target cable into a plurality of image acquisition line segments according to the plurality of temperature acquisition nodes;
[0101] A cable image acquisition module 50, wherein the cable image acquisition module 50 is used to sequentially acquire images of the plurality of image acquisition line segments using a visual acquisition device to acquire a cable image set, wherein the cable image set includes image abnormality nodes and image abnormality detection results;
[0102] A repeated identification module 60, the repeated identification module 60 is used to repeatedly identify the temperature abnormality node and the image abnormality node, and obtain the coincident abnormality node and the coincident abnormality detection result, wherein the coincident abnormality detection result is the detection result corresponding to the same abnormality;
[0103] The detection level acquisition module 70 is used to acquire the quality detection level of different positions of the target cable according to the temperature anomaly detection result, the image anomaly detection result, and the overlap anomaly detection result.
[0104] Furthermore, the system further includes a target cable traversal module to perform the following operation steps:
[0105] The visual acquisition device includes a first acquisition device and a second acquisition device, the second acquisition device has a higher resolution than the first acquisition device, and the first acquisition device and the second acquisition device are arranged adjacent to each other in front and back and are in communication connection;
[0106] Performing image acquisition on the first image acquisition line segment by the first acquisition device to obtain a first image acquisition result;
[0107] Performing abnormality recognition on the first image acquisition result to obtain a first abnormality recognition result;
[0108] When the first abnormality recognition result is that there is no abnormality, generating a rotation instruction, and controlling the target cable to rotate a preset angle according to the rotation instruction;
[0109] After the rotation is in place, the first image acquisition line segment is imaged by the first acquisition device, and abnormality recognition is performed to generate a second abnormality recognition result;
[0110] When the second abnormality recognition result is that there is no abnormality, a translation instruction is generated, and the target cable is controlled to translate to the second image acquisition line segment according to the translation instruction;
[0111] The first acquisition device performs image acquisition on the second image acquisition line segment until all target cables are traversed.
[0112] Furthermore, the system further includes an abnormality identification result acquisition module to perform the following operation steps:
[0113] The first image acquisition result includes a target cable area and a background area, and the target cable area is cropped to obtain an image of the target cable area;
[0114] Performing grayscale conversion on the target cable area image to obtain a grayscale image;
[0115] Establishing a preset grayscale interval according to the local features of the target cable area image, and performing binary classification on the grayscale image based on the preset grayscale interval to generate a binary image, wherein the area that does not meet the preset grayscale interval is marked as abnormal;
[0116] A first abnormality recognition result is generated according to the binary image.
[0117] Furthermore, the system further includes an image anomaly detection result acquisition module to perform the following operation steps:
[0118] When the first abnormality identification result is that an abnormality exists, generating an abnormality capture instruction;
[0119] Acquire an abnormal center and an abnormal range according to the first abnormality recognition result, and establish a first abnormal area according to the abnormal center and the abnormal range;
[0120] activating a second acquisition device based on the abnormality capture instruction, focusing and acquiring the first abnormal area through the second acquisition device to obtain a first abnormal image;
[0121] Anomaly detection is performed on the first abnormal image to obtain a first image abnormality detection result.
[0122] Furthermore, the system further includes a first abnormal area establishing module to perform the following operation steps:
[0123] Establishing a coordinate system according to the target cable position, and obtaining the abnormal center coordinates of the first abnormality recognition result in the coordinate system;
[0124] Positioning and expanding the coordinates of the center of the abnormality according to the abnormal range to obtain a positioning and expansion result;
[0125] According to the aspect ratio of the second acquisition device, a rectangular expansion constraint is established, and the positioning expansion result is limited according to the rectangular expansion constraint to obtain the first abnormal area.
[0126] Furthermore, the system further includes an image anomaly detection result acquisition module to perform the following operation steps:
[0127] Obtaining an abnormality detection record of the charging cable, extracting a sample abnormal image, and marking the sample abnormal image as abnormal, and obtaining a sample image abnormality marking result;
[0128] Based on the twin network, an anomaly detection model is constructed, wherein the anomaly detection model includes a first convolutional neural network branch and a second convolutional neural network branch, wherein the input of the first convolutional neural network branch is a sample abnormal image, and the input of the second convolutional neural network branch is the first abnormal image, and the two branches are the same and share parameters;
[0129] Using the first abnormal image, the sample abnormal image and the sample image abnormality identification result as training data, training the abnormality detection model until the convergence requirement is met, and obtaining the abnormality detection model;
[0130] The first abnormal image and the sample abnormal image are similarly calculated according to the abnormality detection model, and a first image abnormality detection result of the first abnormal image is obtained according to the maximum similarity.
[0131] Furthermore, the system further includes a coincidence anomaly detection result acquisition module to perform the following operation steps:
[0132] Obtaining the coordinates of the temperature abnormal node and the image abnormal node, and marking the nodes with the same coordinates as overlapping abnormal nodes;
[0133] For the overlap abnormal node, the temperature anomaly detection result and the image anomaly detection result are superimposed to obtain the overlap abnormality detection result.
[0134] Through the above detailed description of the quality detection method for electric vehicle charging cables in this specification, those skilled in the art can clearly understand the quality detection method and system for electric vehicle charging cables in this embodiment. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.
[0135] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for detecting the quality of electric vehicle charging cables, characterized in that: The method comprises: An infrared temperature sensor array is arranged on the surface of the target cable, and multiple temperature collection nodes are established according to the sensor arrangement positions; Powering on the target cable, continuously collecting temperature data from the multiple temperature collection nodes through the infrared temperature sensor array, and acquiring a set of temperature data within a preset time; Perform temperature anomaly identification based on the temperature data set to obtain temperature anomaly nodes and temperature anomaly detection results; Dividing the target cable into a plurality of image acquisition line segments according to the plurality of temperature acquisition nodes; Use a visual acquisition device to sequentially acquire images of the plurality of image acquisition line segments to obtain a cable image set, wherein the cable image set includes image abnormality nodes and image abnormality detection results; Repeatedly identify the temperature abnormality node and the image abnormality node to obtain overlapping abnormality nodes and overlapping abnormality detection results, wherein the overlapping abnormality detection results are detection results corresponding to the same abnormality; The quality detection levels of different positions of the target cable are obtained according to the temperature anomaly detection results, the image anomaly detection results, and the overlap anomaly detection results.
2. The method according to claim 1, characterized in that Using a visual acquisition device to sequentially acquire images of the plurality of image acquisition line segments to obtain a cable image set includes: The visual acquisition device includes a first acquisition device and a second acquisition device, the second acquisition device has a higher resolution than the first acquisition device, and the first acquisition device and the second acquisition device are arranged adjacent to each other in front and back and are in communication connection; Performing image acquisition on the first image acquisition line segment by the first acquisition device to obtain a first image acquisition result; Performing abnormality recognition on the first image acquisition result to obtain a first abnormality recognition result; When the first abnormality recognition result is that there is no abnormality, generating a rotation instruction, and controlling the target cable to rotate a preset angle according to the rotation instruction; After the rotation is in place, the first image acquisition line segment is imaged by the first acquisition device, and abnormality recognition is performed to generate a second abnormality recognition result; When the second abnormality recognition result is that there is no abnormality, a translation instruction is generated, and the target cable is controlled to translate to the second image acquisition line segment according to the translation instruction; The first acquisition device performs image acquisition on the second image acquisition line segment until all target cables are traversed.
3. The method according to claim 2, characterized in that Performing abnormality recognition on the first image acquisition result to obtain a first abnormality recognition result includes: The first image acquisition result includes a target cable area and a background area, and the target cable area is cropped to obtain an image of the target cable area; Performing grayscale conversion on the target cable area image to obtain a grayscale image; Establishing a preset grayscale interval according to the local features of the target cable area image, and performing binary classification on the grayscale image based on the preset grayscale interval to generate a binary image, wherein the area that does not meet the preset grayscale interval is marked as abnormal; A first abnormality recognition result is generated according to the binary image.
4. The method according to claim 2, characterized in that Also includes: When the first abnormality identification result is that an abnormality exists, generating an abnormality capture instruction; Acquire an abnormal center and an abnormal range according to the first abnormality recognition result, and establish a first abnormal area according to the abnormal center and the abnormal range; activating a second acquisition device based on the abnormality capture instruction, focusing and acquiring the first abnormal area through the second acquisition device to obtain a first abnormal image; Anomaly detection is performed on the first abnormal image to obtain a first image abnormality detection result.
5. The method of claim 4, establishing the first abnormal area, comprising: Establishing a coordinate system according to the target cable position, and obtaining the abnormal center coordinates of the first abnormality recognition result in the coordinate system; Positioning and expanding the coordinates of the center of the abnormality according to the abnormal range to obtain a positioning and expansion result; According to the aspect ratio of the second acquisition device, a rectangular expansion constraint is established, and the positioning expansion result is limited according to the rectangular expansion constraint to obtain the first abnormal area.
6. The method according to claim 4, performing anomaly detection on the first abnormal image to obtain an abnormality detection result of the first image, comprising: Obtaining an abnormality detection record of the charging cable, extracting a sample abnormal image, and marking the sample abnormal image as abnormal, and obtaining a sample image abnormality marking result; Based on the twin network, an anomaly detection model is constructed, wherein the anomaly detection model includes a first convolutional neural network branch and a second convolutional neural network branch, wherein the input of the first convolutional neural network branch is a sample abnormal image, and the input of the second convolutional neural network branch is the first abnormal image, and the two branches are the same and share parameters; Using the first abnormal image, the sample abnormal image and the sample image abnormality identification result as training data, training the abnormality detection model until the convergence requirement is met, and obtaining the abnormality detection model; The first abnormal image and the sample abnormal image are similarly calculated according to the abnormality detection model, and a first image abnormality detection result of the first abnormal image is obtained according to the maximum similarity.
7. The method according to claim 1, characterized in that Repeatedly identifying the temperature abnormal node and the image abnormal node to obtain the coincident abnormal node and the coincident abnormal detection result, including: Obtaining the coordinates of the temperature abnormal node and the image abnormal node, and marking the nodes with the same coordinates as overlapping abnormal nodes; For the overlap abnormal node, the temperature anomaly detection result and the image anomaly detection result are superimposed to obtain the overlap abnormality detection result.
8. The electric vehicle charging cable quality detection system is characterized by: The method for detecting the quality of an electric vehicle charging cable according to any one of claims 1 to 7 comprises: A collection node acquisition module, the collection node acquisition module is used to lay out an infrared temperature sensor array on the surface of the target cable and establish multiple temperature collection nodes according to the sensor layout positions; A temperature data acquisition module, wherein the temperature data acquisition module is used to energize the target cable, continuously acquire temperature data of the plurality of temperature acquisition nodes through the infrared temperature sensor array, and obtain a set of temperature data within a preset time; A temperature anomaly identification module, the temperature anomaly identification module is used to identify temperature anomalies based on the temperature data set, and obtain temperature anomaly nodes and temperature anomaly detection results; An acquisition line segment acquisition module, the acquisition line segment acquisition module is used to divide the target cable into a plurality of image acquisition line segments according to the plurality of temperature acquisition nodes; A cable image acquisition module, the cable image acquisition module is used to use a visual acquisition device to sequentially acquire images of the multiple image acquisition line segments to acquire a cable image set, wherein the cable image set includes image abnormality nodes and image abnormality detection results; A repeated identification module, the repeated identification module is used to repeatedly identify the temperature abnormality node and the image abnormality node, and obtain the coincident abnormality node and the coincident abnormality detection result, wherein the coincident abnormality detection result is the detection result corresponding to the same abnormality; The detection level acquisition module is used to obtain the quality detection level of different positions of the target cable according to the temperature anomaly detection result, the image anomaly detection result, and the overlap anomaly detection result.
Citation Information
Patent Citations
Cable protective jacket surface defect detector and detection method thereof
CN105911067A
Intelligent visual perception system
CN111770266A
Omnibearing medicine bottle appearance defect detection method
CN113324993A
Surface defect online detection system for cable processing
CN114755235A
Digital visual cable temperature sensing early warning method and system
CN116399462A
