Cable defect detection system and method based on image recognition

Abstract

The invention discloses a cable defect detection system and method based on image recognition, and the method comprises the steps: obtaining a to-be-detected cable image, determining to-be-evaluated defect parameters of a to-be-detected cable based on the to-be-detected cable image, and enabling the to-be-evaluated defect parameters to be used for evaluating the performance and health state of the to-be-detected cable; determining a cable use parameter associated with the to-be-evaluated defect parameter based on historical data of the to-be-detected cable, collecting a parameter information value of the cable use parameter based on a preset frequency, and recording a change condition of the parameter information value; and determining a defect detection result of the to-be-detected cable based on the change condition of the parameter information value. According to the scheme, the cable use parameters can be intelligently utilized to predict and evaluate the cable defects, and the accuracy and efficiency of cable defect detection are improved.

Description

Technical Field

[0001] The present application belongs to the technical field of cable detection, and specifically relates to a cable defect detection system and method based on image recognition. Background Art

[0002] Cable is one of the main media for power transmission. The safety and reliability of cables directly affect the quality and stability of power supply. Cable defects increase the risk of cable failures, such as short circuits and overloads, which may cause fires or electric shock accidents, posing a serious threat to personal and property safety. Through timely cable defect detection, potential safety hazards can be discovered and eliminated in a timely manner to protect people's lives and property.

[0003] In the existing cable defect detection process, it is usually determined whether the cable has defects based on image recognition or human visual observation. The defect comparison basis depends on various pre-set standards. However, for subtle defects or parameters without comparison basis, existing detection methods often find it difficult to give accurate evaluation results. An evaluation mechanism is needed to improve the accuracy and efficiency of cable defect detection. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a cable defect detection system and method based on image recognition, with the aim of solving the technical problem that existing detection methods often find it difficult to give accurate evaluation results for defects that are not obvious or for which there are no parameters for comparison. The system and method can intelligently use cable usage parameters to predict and evaluate cable defects, thereby improving the accuracy and efficiency of cable defect detection.

[0005] In a first aspect, an embodiment of the present application provides a cable defect detection method based on image recognition, the method comprising:

[0006] Acquire an image of the cable to be tested, and determine defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested, wherein the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested;

[0007] Determine the cable usage parameters associated with the defect parameters to be evaluated based on the historical data of the cable to be tested, collect parameter information values ​​of the cable usage parameters based on a preset frequency, and record changes in the parameter information values;

[0008] The defect detection result of the cable to be tested is determined based on the change of the parameter information value.

[0009] Further, the determining the defect parameter to be evaluated of the cable to be tested based on the image of the cable to be tested includes:

[0010] Identify the image of the cable to be tested based on an image processing algorithm to obtain characteristic information of the cable to be tested;

[0011] Comparing the characteristic information with preset defect characteristic information to determine the defect type of the cable to be tested;

[0012] Determine the defect parameter to be evaluated corresponding to the defect type based on the preset association relationship.

[0013] Furthermore, the determining of the cable usage parameters associated with the defect parameters to be evaluated based on the historical data of the cable to be tested includes:

[0014] Collecting historical defect data of the cable to be tested within a preset historical time period, wherein the historical defect data includes defect type, defect parameter, occurrence time, occurrence location, and historical usage parameters;

[0015] Filter out historical usage parameters that meet preset change trends within a preset time period before and after the occurrence of historical defects;

[0016] When the number of times the historical usage parameter is screened reaches a preset threshold, an association relationship is established between the historical usage parameter and the defect type, defect parameter, occurrence time, and occurrence location;

[0017] The cable usage parameter associated with the defect parameter to be evaluated is determined based on the association relationship.

[0018] Further, the determining, based on the association relationship, the cable usage parameter associated with the defect parameter to be evaluated includes:

[0019] When the defect parameter to be evaluated is insulation resistance, the associated cable usage parameter is the actual voltage of the cable to be tested;

[0020] When the defect parameter to be evaluated is partial discharge, the associated cable usage parameter is the load current of the cable to be tested;

[0021] When the defect parameter to be evaluated is thermal stability, the associated cable usage parameters are the length, cross section and surface temperature of the cable to be tested;

[0022] When the defect parameter to be evaluated is compressive strength, the associated cable usage parameter is the pressure value of the cable to be tested;

[0023] When the defect parameter to be evaluated is the bending degree, the associated cable usage parameter is the bending radius of the cable to be tested.

[0024] Furthermore, after the acquisition task is started based on the preset frequency and the usage parameter value of the cable to be tested is read from the sensor, the method further includes:

[0025] Calculating the difference between the usage parameter values ​​of the cable to be tested collected at adjacent time points;

[0026] When the difference is greater than a preset threshold, the usage parameter value of the cable to be tested collected at the corresponding time point is eliminated.

[0027] Further, the determining the defect detection result of the cable to be tested based on the change of the parameter information value includes:

[0028] Recording the usage parameter values ​​of the cable to be tested in chronological order to form time series data;

[0029] Draw a change curve of the usage parameter of the cable to be tested based on the time series data;

[0030] Obtaining the fluctuation amplitude of the usage parameter of the cable to be tested based on the change curve;

[0031] When the similarity between the fluctuation amplitude and the fluctuation amplitude of the preset defect meets a preset threshold, it is determined that the cable to be tested has a defect.

[0032] Further, when the similarity between the fluctuation amplitude and the fluctuation amplitude of the preset defect meets the preset threshold, after determining that the tested cable has a defect, the method further includes:

[0033] The cable usage parameters and environmental data of known defect causes are obtained in advance, and the defect causes are used as labels, and the cable usage parameters and environmental data are used as input variables for supervised training to learn the correlation between the defect causes and the environmental data and the cable usage parameters, thereby constructing a cause analysis model; wherein the environmental data includes temperature data, precipitation data and landform data;

[0034] Obtain environmental data around the cable under test;

[0035] A cause analysis model is pre-constructed using the acquired environmental data and the cable usage parameter input values, and the cause analysis model predicts the cause of the defect based on the pre-learned correlation between the defect cause and the environmental data and the usage parameters.

[0036] In a second aspect, an embodiment of the present application provides a cable defect detection system based on image recognition, the system comprising:

[0037] An image acquisition module, used for acquiring an image of the cable to be tested;

[0038] A first parameter determination module, used to determine the defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested, wherein the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested;

[0039] A second parameter determination module, configured to determine the cable usage parameter associated with the defect parameter to be evaluated based on the historical data of the cable to be tested;

[0040] A parameter value recording module, used to collect parameter information values ​​of the cable usage parameters based on a preset frequency, and record changes in the parameter information values;

[0041] The result determination module is used to determine the defect detection result of the cable to be tested based on the change of the parameter information value.

[0042] In a third aspect, an embodiment of the present application provides a cable defect detection device in combination with cable usage parameters, the device comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the cable defect detection method based on image recognition described in the embodiment of the present application.

[0043] In a fourth aspect, an embodiment of the present application provides a storage medium storing computer executable instructions, which, when executed by a computer processor, are used to execute the cable defect detection method based on image recognition described in an embodiment of the present invention.

[0044] In an embodiment of the present application, an image of a cable to be tested is obtained, and defect parameters to be evaluated of the cable to be tested are determined based on the image of the cable to be tested, and the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested; cable usage parameters associated with the defect parameters to be evaluated are determined based on historical data of the cable to be tested, parameter information values ​​of the cable usage parameters are collected based on a preset frequency, and changes in the parameter information values ​​are recorded; defect detection results of the cable to be tested are determined based on changes in the parameter information values. This solution performs cable defect detection through cable usage parameters, which solves the technical problem that existing detection methods often have difficulty in giving accurate evaluation results for non-obvious defects or parameters for which there is no comparison basis, and can intelligently use cable usage parameters to predict and evaluate cable defects, thereby improving the accuracy and efficiency of cable defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of a flow chart of a cable defect detection method based on image recognition provided by an embodiment of the present invention;

[0046] Figure 2A flow chart of a method for determining defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested provided by an embodiment of the present invention;

[0047] Figure 3 A flow chart of a method for determining the cable usage parameters associated with the defect parameters to be evaluated based on the historical data of the cable to be tested provided by an embodiment of the present invention;

[0048] Figure 4 A schematic diagram of a flow chart of another cable defect detection method based on image recognition provided by an embodiment of the present invention;

[0049] Figure 5 A module structure block diagram of a cable defect detection system based on image recognition provided by an embodiment of the present invention;

[0050] Figure 6 A schematic structural diagram of a cable defect detection device based on image recognition provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only the part related to the present application but not all the contents are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0053] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0054] The cable defect detection method based on image recognition provided in the embodiment of the present application can be applied to the detection scenario of inconspicuous cable defects, such as monitoring the existence of defects in the cable during use, but the existing image recognition technology alone cannot clearly identify the cause of the cable defect. The cable defect can be determined by the defect detection method provided in the embodiment of the present application. In the cable defect detection method based on image recognition provided in the embodiment of the present application, the execution subject of each step can be a computer device, which refers to any electronic device with data calculation, processing and storage capabilities, such as mobile phones, PCs (Personal Computers), tablet computers and other terminal devices, and can also be servers and other devices, which are not limited in the embodiment of the present application.

[0055] Figure 1 A flow chart of a cable defect detection method based on image recognition provided by an embodiment of the present invention is as follows: Figure 1 As shown, specifically including:

[0056] S101. Acquire an image of a cable to be tested, and determine defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested, where the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested.

[0057] Among them, the image of the cable to be tested refers to the image data of the cable obtained by the image acquisition device in order to detect, analyze or record a certain characteristic or state of the cable. The image of the cable to be tested can be an infrared image of the cable to be tested or a visible light image of the cable to be tested. The infrared image of the cable to be tested is an image of the temperature distribution on the surface of the cable obtained by an infrared thermal imager. This image is displayed in grayscale or pseudo-color form, where different colors or grayscale levels represent different temperature values. The infrared image can intuitively show the temperature changes on the cable, thereby helping the inspector to find potential defects or faults. The visible light image of the cable to be tested is an image of the appearance of the cable obtained by an ordinary camera or a high-definition camera. The visible light image is used to observe the appearance of the cable, such as wear, cracks, deformation, etc. on the cable surface. The physical state of the cable can be intuitively displayed, which is convenient for quickly identifying obvious defects.

[0058] Among them, the defect parameters of the cable refer to the parameters used to describe and quantify the various characteristics or conditions in the cable that do not meet the standards, specifications or expected performance requirements. The defect parameters of the cable can be electrical characteristic parameters, thermal characteristic parameters and mechanical characteristic parameters. Among them, the electrical characteristic parameters include insulation resistance and partial discharge, the thermal characteristic parameters include thermal stability, and the mechanical characteristic parameters include compressive strength and bending degree. The defect parameters related to the cable defects determined based on the image of the cable to be tested are determined as the defect parameters to be evaluated. The performance and health status of the cable to be tested can directly reflect the transmission efficiency, safety and service life of the cable to be tested.

[0059] In one embodiment, a camera installed on an inspection vehicle or unmanned aerial vehicle captures an image of the entire cable or a specific area to obtain an image of the cable to be tested. The image of the cable to be tested is input into a pre-trained image defect detection model, and the image defect detection model outputs the defect type of the cable to be tested and the defect parameters to be evaluated.

[0060] Figure 2 A flow chart of a method for determining the defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested is provided in an embodiment of the present invention, such as Figure 2 As shown, the specific steps include:

[0061] S1011. Identify the image of the cable to be tested based on an image processing algorithm to obtain characteristic information of the cable to be tested.

[0062] Among them, image processing algorithms are a series of methods for computer processing images, including image analysis algorithms, image transformation algorithms, and image recognition and matching algorithms. The characteristic information of the cable to be tested refers to the data or parameters that can describe the physical, electrical, chemical or other characteristics of the cable to be tested, obtained from the cable to be tested through a series of measurements, analyses or image processing.

[0063] In one embodiment, feature extraction is performed on the cable to be tested based on an image analysis algorithm, thereby obtaining shape features, texture features, and color features of the cable to be tested.

[0064] Optionally, before extracting features from the image of the cable to be tested, the image may be preprocessed, for example, denoising the image of the cable to be tested based on a filtering algorithm, or contrast enhancing the image based on an image enhancement algorithm.

[0065] S1012: Compare the characteristic information with preset defect characteristic information to determine the defect type of the cable to be tested.

[0066] Among them, the preset defect characteristic information is set based on previous cable defect data, which covers various defect types that may exist in the cable and their corresponding characteristic information. This information usually exists in the form of a database, table or software system, which is convenient for quick retrieval and comparison. The defect type of the cable refers to various types of problems that may occur in the structure or material of the cable, which will affect the performance, safety and life of the cable. The defect type of the cable can include physical defects, electrical defects, thermal defects and mechanical defects.

[0067] In one embodiment, the characteristic information of the cable to be tested is sorted with the preset defect characteristic information to ensure the consistency of the two in terms of format, unit, etc. The various characteristic information of the cable to be tested is compared with the preset defect characteristic information one by one. According to the comparison result, the type of defect that may exist in the cable to be tested is identified. For example, if the similarity between the length of the cable break and the shape of the cross section and the length of the cable break and the shape of the cross section in the preset defect reaches 80%, the defect type of the cable to be tested is determined to be an electrical defect. If the bending degree of the cable to be tested is greater than or equal to the bending degree of the preset defect, the defect type of the cable to be tested is determined to be a mechanical property defect.

[0068] S1013: Determine the defect parameter to be evaluated corresponding to the defect type based on the preset association relationship.

[0069] The preset association relationship refers to the relationship between the predefined defect type and the defect parameter to be evaluated.

[0070] In one embodiment, the known cable defect types and their associated defect parameters are obtained to build an association analysis model, and the association relationship between the defect type and the defect parameter obtained based on the association analysis model is counted to form an association relationship list. The association relationship list is queried based on the defect type to determine the defect parameter associated with the defect type. The current defect parameter is determined as the defect parameter to be evaluated.

[0071] S102: determining the cable usage parameters associated with the defect parameters to be evaluated based on the historical data of the cable to be tested, collecting parameter information values ​​of the cable usage parameters based on a preset frequency, and recording changes in the parameter information values.

[0072] Among them, the historical data of the cable under test refers to the data records accumulated in the operation, maintenance, detection, fault handling, etc. of the cable under test in the past period of time. Specifically, the historical data of the cable under test of the cable under test includes historical defect data and historical cable usage data. The preset frequency usually refers to a sampling rate or sampling frequency that is set or preset in advance during the data acquisition process, which is reasonably set according to actual needs and data change characteristics. This frequency should be high enough to capture important changes in parameters, but not too high to waste resources. Cable usage parameters are indicators that describe the performance and applicable conditions of the cable in actual use. The parameter information value of the cable usage parameter refers to the specific actual value of the cable usage parameter during the operation of the cable.

[0073] In one embodiment, all relevant data of the cable to be tested within a historical time period are collected, including defect data and historical records of cable usage parameters. The correlation between historical defect parameters and historical cable usage parameters is analyzed using a data mining algorithm to identify cable usage parameters that have a significant correlation with the defect parameters to be evaluated. Based on the historical defect data and historical usage parameters with significant correlation, an association analysis model is established to determine the correlation between the defect parameters and the cable usage parameters. The defect parameters to be evaluated are input into the association analysis model, and the association analysis model outputs the cable usage parameters related thereto.

[0074] In one embodiment, the data acquisition system automatically collects data at a preset frequency and stores the collected data in a database or data recording system. The system records parameter information values ​​in real time and updates the data at a preset frequency. The data processing center regularly analyzes the recorded data to evaluate parameter change trends and abnormal conditions. For example, trend graphs, comparison graphs, etc. are used to observe parameter changes.

[0075] Figure 3 A flow chart of a method for determining the cable usage parameters associated with the defect parameters to be evaluated based on the historical data of the cable to be tested is provided in an embodiment of the present invention, such as Figure 3 As shown, the specific steps include:

[0076] S1021. Collect historical defect data of the cable to be tested within a preset historical time period, wherein the historical defect data includes defect type, defect parameters, occurrence time, occurrence location, and historical usage parameters.

[0077] The preset historical time period is a period of time in the past selected to study the correlation between the defect parameters to be evaluated and the cable usage parameters in the historical cable data. For example, the historical defect data of the past one, five or ten years can be selected as the analysis object, and this period of time is the preset historical time period. Historical defect data refers to the data information about the existing or past problems, faults and abnormal conditions of the cable recorded and accumulated through detection, maintenance, inspection and other means during the long-term use of the cable. Historical usage parameters refer to the historical parameter data used for historical defect and fault analysis.

[0078] In one embodiment, based on a set time period, the corresponding historical defect data is screened from a relevant data system or database that records historical defect data. These data may include defect type (such as physical defects, electrical defects, thermal defects, and mechanical defects, etc.), defect parameters (such as thermal stability, compressive strength, and bending degree, etc.), time of defect occurrence, location of occurrence (such as the specific location of the cable section), and historical usage parameters (such as current, voltage, temperature, humidity, load changes, etc.).

[0079] S1022. Filter out historical usage parameters that meet a preset change trend within a preset time period before and after the occurrence of the historical defect.

[0080] The preset time period before and after the occurrence of a historical defect refers to a window of time before and after the occurrence of the defect, such as one week before the defect occurs to one week after the defect occurs. The preset change trend refers to the change rule that a historical usage parameter needs to meet during this period, such as continuous increase, decrease or fluctuation.

[0081] In one embodiment, trend analysis is performed on the data retrieved in the above steps to determine whether the parameter values ​​conform to a preset change trend within a preset time period. Based on the result of the trend analysis, historical usage parameters that conform to the preset change trend are screened out.

[0082] S1023. When the number of times the historical usage parameter is screened reaches a preset threshold, an association relationship is established between the historical usage parameter and the defect type, defect parameter, occurrence time, and occurrence location.

[0083] The number of times that the historical usage parameter is screened reaches a preset threshold value means that the number of times that the historical parameter data corresponding to a certain type of historical usage parameter is screened to meet a preset change trend reaches a preset number threshold value.

[0084] In one embodiment, when the historical parameter data corresponding to the type of historical usage parameter is screened out and the number of times it meets the preset change trend reaches a preset number threshold, it is considered that there is a significant correlation between the parameter and the defect type and defect parameter, and an association relationship list is established based on the parameter and the corresponding defect type, defect parameter, occurrence time, occurrence location and other defect information.

[0085] S1024. Determine the cable usage parameters associated with the defect parameters to be evaluated based on the association relationship.

[0086] In one embodiment, after establishing the association relationship lists between the historical usage parameters and the historical defect information, the cable usage parameters associated with the defect parameters to be evaluated may be determined based on the association relationship lists.

[0087] Optionally, determining the cable usage parameter associated with the defect parameter to be evaluated based on the association relationship includes:

[0088] When the defect parameter to be evaluated is insulation resistance, the associated cable usage parameter is the actual voltage of the cable to be tested;

[0089] When the defect parameter to be evaluated is partial discharge, the associated cable usage parameter is the load current of the cable to be tested;

[0090] When the defect parameter to be evaluated is thermal stability, the associated cable usage parameters are the length, cross section and surface temperature of the cable to be tested;

[0091] When the defect parameter to be evaluated is compressive strength, the associated cable usage parameter is the pressure value of the cable to be tested;

[0092] When the defect parameter to be evaluated is the bending degree, the associated cable usage parameter is the bending radius of the cable to be tested.

[0093] Among them, insulation resistance is an important indicator to measure the insulation performance of cables. It reflects the ability of cables to prevent current from passing through insulating materials. The larger the value of insulation resistance, the better the insulation performance of the cable, and vice versa. The reduction in insulation resistance is directly related to the insulation aging caused by the long-term operation of the cable at the rated voltage. Therefore, when insulation resistance is used as a defect parameter to be evaluated, the associated cable usage parameter is the actual voltage of the cable to be tested, and it is necessary to pay attention to whether the cable continues to operate at its rated voltage. The rated voltage is fixed, and it can be indirectly verified by monitoring the actual operating voltage of the cable to verify whether it is close to or exceeds its rated voltage.

[0094] Among them, partial discharge refers to a discharge phenomenon in which the insulation system in high-voltage equipment discharges in some areas under the action of high electric field strength without forming a penetrating discharge channel. It usually occurs inside or on the surface of the insulating medium, appears in the form of pulses, and lasts for a very short time (nanoseconds to microseconds). The energy of partial discharge is small, but its long-term existence will cause erosion and damage to the insulating material, resulting in a decrease in insulation performance and even insulation breakdown. Load current refers to the current passing through electrical equipment. In some cases, changes in load current may indirectly affect the intensity and frequency of partial discharge. For example, when the load current increases, the heat generated by the electrical equipment increases, which may cause the temperature of the insulating material to rise, thereby exacerbating the occurrence of partial discharge. Therefore, when the defect parameter to be evaluated is partial discharge, the load current of the cable to be tested is determined as the associated cable usage parameter.

[0095] Among them, thermal stability refers to the long-term stability of the physical and electrical properties of the cable under certain temperature and time conditions. This feature is crucial to the service life and safety of the cable. During long-term use, the cable will be affected by various factors such as resistance heating and external ambient temperature. If the thermal stability of the cable is insufficient, it will easily lead to cable aging, leakage and other faults, and in severe cases may even cause serious consequences such as fire. The thermal stability of the cable is closely related to its heat dissipation capacity, which depends on the length, cross-section and surface temperature of the cable. Cables that are too long, too small in cross-section or too high in surface temperature may cause the cable to overheat, thereby affecting its thermal stability. Therefore, when the defect parameter to be evaluated of the cable is thermal stability, the associated cable usage parameters are the length, cross-section and surface temperature of the cable to be tested.

[0096] In one embodiment, the length and cross-sectional dimensions (such as diameter or area) of the cable can be obtained from the cable specification, and the surface temperature can be measured by an infrared thermal imager or a temperature sensor. Specifically, the surface temperature of the cable can be measured regularly (such as every day or every week) using an infrared thermal imager, and the length and cross-sectional data of the cable can be recorded.

[0097] The compressive resistance of a cable refers to the ability of the cable to maintain the stability of its structure and performance without deformation, rupture or damage when subjected to external pressure. The compressive resistance of a cable is related to its structure and material properties, but the increase in external pressure may exceed the cable's ability to withstand and cause damage. Therefore, when the defect parameter to be evaluated is compressive resistance, the associated cable usage parameter is the pressure value of the cable to be tested, and the pressure value of the cable to be monitored exceeds its compressive resistance range.

[0098] Among them, the bending degree of the cable refers to the bending state that the cable needs to present during the use of the cable. The bending radius of the cable is the minimum radius of the bending part of the cable. The bending radius of the cable is an important parameter to measure the bending degree of the cable. Therefore, when the defect parameter to be evaluated is the bending degree, the bending radius of the cable to be tested is used as the associated cable usage parameter. Generally speaking, the bending radius of the cable should not be less than a certain multiple of the cable diameter (such as 6 times, 8 times, etc.) to ensure that the cable will not be subjected to excessive mechanical stress when bent. If the bending radius is too small, it will aggravate the deformation and damage of the cable and increase the risk of cable failure.

[0099] Optionally, after collecting the parameter information value of the cable usage parameter based on the preset frequency, the method further includes:

[0100] Calculating the difference between the usage parameter values ​​of the cable to be tested collected at adjacent time points;

[0101] When the difference is greater than a preset threshold, the usage parameter value of the cable to be tested collected at the corresponding time point is eliminated.

[0102] Among them, the preset threshold is set according to the normal working conditions of the cable and the reasonable range of parameter changes. This threshold should be large enough to allow normal parameter fluctuations, but small enough to identify abnormal parameter changes that may be caused by defects. It should be noted that the selection of the preset threshold is a key step, which directly affects the accuracy and effectiveness of outlier removal. Therefore, in practical applications, the threshold should be reasonably set according to the specific conditions and historical data of the cable, and adjusted and optimized multiple times as needed.

[0103] In one embodiment, for the data collected at each time point, the difference between it and the data collected at the previous time point is calculated. This difference reflects the change in the parameter value over time. The difference at each time point is compared with a preset threshold. If the difference is greater than the threshold, the parameter value collected at the current moment is considered abnormal. For all time points where the difference is greater than the preset threshold, the cable usage parameter values ​​collected at the corresponding time point are eliminated. This ensures that subsequent data analysis and processing are based on accurate and reliable data sets. After eliminating outliers, other data cleaning and preprocessing steps can also be performed, such as filling missing values, smoothing data, etc.

[0104] S103: Determine a defect detection result of the cable to be tested based on a change in the parameter information value.

[0105] The cable defect detection result refers to the problems or non-compliance status of the cable found after various tests on the cable.

[0106] In one embodiment, trend analysis is performed on each parameter information value to observe its changes over time. Note whether the parameter value has abnormal conditions such as sudden drop, increased fluctuation, or continuous deviation from the normal range. The current parameter information value is compared with the historical data to determine whether the change in the parameter value exceeds the normal range or the historical fluctuation range. Based on the change in the parameter information value, a preliminary determination is made as to whether the cable has defects and the type and severity of the defects. If a parameter value exceeds the set threshold or shows obvious abnormal changes, it may indicate that the cable has corresponding defects.

[0107] As can be seen from the above, an image of the cable to be tested is obtained, and defect parameters to be evaluated of the cable to be tested are determined based on the image of the cable to be tested, and the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested; the cable usage parameters associated with the defect parameters to be evaluated are determined based on the historical data of the cable to be tested, and the parameter information values ​​of the cable usage parameters are collected based on a preset frequency, and the changes in the parameter information values ​​are recorded; the defect detection results of the cable to be tested are determined based on the changes in the parameter information values. This solution performs cable defect detection through cable usage parameters, which solves the technical problem that existing detection methods often find it difficult to give accurate evaluation results for non-obvious defects or parameters for which there is no comparison basis. It can intelligently use cable usage parameters to predict and evaluate cable defects, thereby improving the accuracy and efficiency of cable defect detection.

[0108] Figure 4 A schematic flow chart of another cable defect detection method based on image recognition provided by an embodiment of the present invention provides another optional specific process of determining the defect detection result of the cable to be tested based on the change of the parameter information value, such as Figure 4 As shown, the method specifically includes:

[0109] S201. Acquire an image of a cable to be tested, and determine defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested, where the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested.

[0110] S202: Determine the cable usage parameters associated with the defect parameters to be evaluated based on historical data of the cable to be tested, collect parameter information values ​​of the cable usage parameters based on a preset frequency, and record changes in the parameter information values.

[0111] S203: Record the usage parameter values ​​of the cable to be tested in chronological order to form time series data.

[0112] Among them, recording in time series means arranging and recording the collected cable parameter values ​​in chronological order. Time series is a data representation method that emphasizes the temporal order relationship between data, that is, data changes over time. Recording in time series usually requires following a certain time interval, such as every second, every minute, or every hour.

[0113] In one embodiment, the collected data is sorted in chronological order to ensure that each data point has an accurate timestamp and that the timestamps are evenly distributed at set intervals. The data is sorted into a format suitable for analysis, such as CSV, Excel, or a database table. The sorted data is arranged in chronological order to form time series data. This can be a univariate time series (each time point has only one parameter value) or a multivariate time series (each time point has multiple related parameter values). The use parameter values ​​of the cable to be tested are recorded in chronological order to form time series data, which can ensure the continuity and timing of the data, and facilitate subsequent trend analysis and anomaly detection.

[0114] S204: Draw a variation curve of the usage parameters of the tested cable based on the time series data.

[0115] Among them, the change curve of cable usage parameters is a graphical representation of the changes of cable usage parameters over time under specific conditions.

[0116] In one embodiment, a mapping relationship is established between the timestamp column and the parameter value column, and a line chart is selected as the chart type. Ensure that the timestamp is correctly interpreted as the X-axis (horizontal axis) and the parameter value is interpreted as the Y-axis (vertical axis). Import the processed time series data into a drawing tool and draw a parameter change curve in chronological order.

[0117] Optionally, after generating the change curve, you can also preview the chart to check whether the data is displayed as expected and adjust the time range, zoom ratio, or add data point labels as needed.

[0118] S205: Obtain the fluctuation amplitude of the usage parameter of the cable to be tested based on the change curve.

[0119] The fluctuation range of cable parameters refers to the range of fluctuations of cable parameters relative to a certain reference value or average value during the operation of the cable. This range is usually used to describe the stability and dynamic characteristics of cable parameters as they change over time.

[0120] In one embodiment, according to the change curve, the highest and lowest points of the parameter value in the specified time period are found. The extreme value usually represents the extreme situation of the parameter value and is the key to calculating the fluctuation range. Specifically, the built-in function or tool can be used to automatically find the highest and lowest values, or it can be checked manually. The fluctuation range is determined based on the difference between the highest and lowest values, which is the range of change of the parameter value in the specified time period, that is, the maximum amplitude of its fluctuation.

[0121] It should be noted that when calculating the fluctuation amplitude, it is necessary to ensure that the selected time period is representative and can reflect the actual operation of the cable. If there are multiple obvious fluctuation cycles or trend changes in the change curve, it is necessary to calculate the fluctuation amplitude in different time periods or different intervals separately in order to more accurately evaluate the performance of the cable.

[0122] S206: When the similarity between the fluctuation amplitude and the fluctuation amplitude of a preset defect meets a preset threshold, it is determined that the tested cable has a defect.

[0123] The fluctuation amplitude of the preset defect refers to the fluctuation amplitude of the cable usage parameters extracted based on the historical cable defect data during operation. The preset threshold refers to the similarity threshold that the fluctuation amplitude and the fluctuation amplitude of the preset defect need to meet.

[0124] In one embodiment, the quantitative index values ​​of the fluctuation amplitude and the fluctuation amplitude of the preset defect are calculated respectively, which may be the difference between the maximum value and the minimum value, the standard deviation, etc. The ratio between the two quantitative indexes is calculated, and then converted into a similarity value between 0 and 1 through normalization. When the calculated similarity meets the preset similarity threshold, it can be considered that the cable to be tested has a fluctuation pattern similar to the preset defect, thereby preliminarily determining that the cable has a defect.

[0125] On the basis of the above embodiment, when the similarity between the fluctuation amplitude and the fluctuation amplitude of the preset defect meets the preset threshold, it is determined that the tested cable has a defect, and further includes:

[0126] The cable usage parameters and environmental data of known defect causes are obtained in advance, and the defect causes are used as labels, and the cable usage parameters and environmental data are used as input variables for supervised training to learn the correlation between the defect causes and the environmental data and the cable usage parameters, thereby constructing a cause analysis model; wherein the environmental data includes temperature data, precipitation data and landform data;

[0127] Obtain environmental data around the cable under test;

[0128] A cause analysis model is pre-constructed using the acquired environmental data and the cable usage parameter input values, and the cause analysis model predicts the cause of the defect based on the pre-learned correlation between the defect cause and the environmental data and the usage parameters.

[0129] Among them, the known defect cause refers to the specific cause of the defect or failure that has been identified and confirmed in the historical cable defect failure. Cable environmental data refers to various environmental parameter data related to the cable operation and installation environment. The label is the output or result corresponding to each sample in the data set. Labels are used in the supervised learning process. The model predicts or classifies by learning the relationship between the input data (features) and the corresponding labels. Supervised training is a basic method in machine learning, which involves training the model using a set of data sets with known inputs (features) and corresponding outputs (labels or target variables). During the supervised training process, the model learns how to extract useful information or patterns from the input data so that it can predict the output of new data points.

[0130] In one embodiment, first, a large number of cable usage parameters and corresponding environmental data with known defect causes are collected. These data should cover multiple defect types and situations under different environmental conditions. For each collected data record, its defect cause needs to be labeled. These labels are the target variables for subsequent model training. After removing or correcting errors, missing or outliers in the data, features that affect the cause of the defect are selected from the cable usage parameters and environmental data as input variables. These features may include cable usage parameters such as current, voltage, and resistance, as well as environmental data such as temperature, precipitation, and topography. Use the labeled data set to train the selected algorithm to learn the correlation between the defect cause and the cable usage parameters and environmental data, thereby building a cause analysis model.

[0131] Optionally, you can evaluate the performance of the model through methods such as cross-validation to ensure that it maintains a high prediction accuracy even on unseen data.

[0132] In one embodiment, after the cause analysis model is constructed, environmental data and cable usage parameters around the cable to be tested are obtained in real time or periodically, and the obtained environmental data and cable usage parameters are input into the pre-constructed cause analysis model. The cause analysis model infers based on the input data and the previously learned associations to predict the most likely cause of the defect.

[0133] Optionally, a corresponding maintenance and repair plan can be formulated based on the prediction results to reduce or eliminate the impact of defects on cable performance. The actual processing results are fed back to the model to further optimize the performance and accuracy of the model.

[0134] Figure 5 The module structure block diagram of a cable defect detection system based on image recognition provided by an embodiment of the present invention. The system is used to execute the cable defect detection method based on image recognition provided by the above embodiment, and has the corresponding functional modules and beneficial effects of the execution method. Figure 5 As shown, the system specifically includes:

[0135] An image acquisition module 301 is used to acquire an image of the cable to be tested;

[0136] A first parameter determination module 302, configured to determine a defect parameter to be evaluated of the cable to be tested based on the image of the cable to be tested, wherein the defect parameter to be evaluated is used to evaluate the performance and health status of the cable to be tested;

[0137] A second parameter determination module 303 is used to determine the cable usage parameter associated with the defect parameter to be evaluated based on the historical data of the cable to be tested;

[0138] A parameter value recording module 304 is used to collect parameter information values ​​of the cable usage parameters based on a preset frequency and record changes in the parameter information values;

[0139] The result determination module 305 is used to determine the defect detection result of the cable to be tested based on the change of the parameter information value.

[0140] In one embodiment, the first parameter determination module 302 is specifically used to:

[0141] Identify the image of the cable to be tested based on an image processing algorithm to obtain characteristic information of the cable to be tested;

[0142] Comparing the characteristic information with preset defect characteristic information to determine the defect type of the cable to be tested;

[0143] Determine the defect parameter to be evaluated corresponding to the defect type based on the preset association relationship.

[0144] In one embodiment, the second parameter determination module 303 is specifically used to:

[0145] Collecting historical defect data of the cable to be tested within a preset historical time period, wherein the historical defect data includes defect type, defect parameter, occurrence time, occurrence location, and historical usage parameters;

[0146] Filter out historical usage parameters that meet preset change trends within a preset time period before and after the occurrence of historical defects;

[0147] When the number of times the historical usage parameter is screened reaches a preset threshold, an association relationship is established between the historical usage parameter and the defect type, defect parameter, occurrence time, and occurrence location;

[0148] Determine the cable usage parameter associated with the defect parameter to be evaluated based on the association relationship;

[0149] Wherein, when the defect parameter to be evaluated is insulation resistance, the associated cable usage parameter is the actual voltage of the cable to be tested;

[0150] When the defect parameter to be evaluated is partial discharge, the associated cable usage parameter is the load current of the cable to be tested;

[0151] When the defect parameter to be evaluated is thermal stability, the associated cable usage parameters are the length, cross section and surface temperature of the cable to be tested;

[0152] When the defect parameter to be evaluated is compressive strength, the associated cable usage parameter is the pressure value of the cable to be tested;

[0153] When the defect parameter to be evaluated is the bending degree, the associated cable usage parameter is the bending radius of the cable to be tested.

[0154] In one embodiment, the cable defect detection system based on image recognition further includes an abnormal parameter recognition module, which is used to:

[0155] Calculating the difference between the usage parameter values ​​of the cable to be tested collected at adjacent time points;

[0156] When the difference is greater than a preset threshold, the usage parameter value of the cable to be tested collected at the corresponding time point is eliminated.

[0157] In one embodiment, the result determination module 305 is specifically used to:

[0158] Recording the usage parameter values ​​of the cable to be tested in chronological order to form time series data;

[0159] Draw a change curve of the usage parameter of the cable to be tested based on the time series data;

[0160] Obtaining the fluctuation amplitude of the usage parameter of the cable to be tested based on the change curve;

[0161] When the similarity between the fluctuation amplitude and the fluctuation amplitude of the preset defect meets a preset threshold, it is determined that the cable to be tested has a defect.

[0162] In one embodiment, the cable defect detection system based on image recognition further includes a defect cause prediction module, which is used to:

[0163] The cable usage parameters and environmental data of known defect causes are obtained in advance, and the defect causes are used as labels, and the cable usage parameters and environmental data are used as input variables for supervised training to learn the correlation between the defect causes and the environmental data and the cable usage parameters, thereby constructing a cause analysis model; wherein the environmental data includes temperature data, precipitation data and landform data;

[0164] Obtain environmental data around the cable under test;

[0165] A cause analysis model is pre-constructed using the acquired environmental data and the cable usage parameter input values, and the cause analysis model predicts the cause of the defect based on the pre-learned correlation between the defect cause and the environmental data and the usage parameters.

[0166] It can be seen from the above scheme that an image of the cable to be tested is obtained, and defect parameters to be evaluated of the cable to be tested are determined based on the image of the cable to be tested, and the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested; the cable usage parameters associated with the defect parameters to be evaluated are determined based on the historical data of the cable to be tested, and the parameter information values ​​of the cable usage parameters are collected based on a preset frequency, and the changes in the parameter information values ​​are recorded; the defect detection results of the cable to be tested are determined based on the changes in the parameter information values. This scheme performs cable defect detection through cable usage parameters, which solves the technical problem that existing detection methods often find it difficult to give accurate evaluation results for non-obvious defects or parameters for which there is no comparison basis. It can intelligently use cable usage parameters to predict and evaluate cable defects, thereby improving the accuracy and efficiency of cable defect detection.

[0167] Figure 6 A schematic diagram of the structure of a cable defect detection device based on image recognition provided by an embodiment of the present invention is shown in FIG. Figure 6 As shown, the device includes a processor 401, a memory 402, an input device 403 and an output device 404; the number of processors 401 in the device can be one or more. Figure 4 A processor 401 is taken as an example; the processor 401, memory 402, input device 403 and output device 404 in the device can be connected by a bus or other means. Figure 6The example of connecting through a bus is taken. The memory 402, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions or modules corresponding to the cable defect detection method based on image recognition in the embodiment of the present invention. The processor 401 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 402, that is, realizes the above-mentioned cable defect detection method based on image recognition. The input device 403 can be used to receive input digital or character information, and generate key signal input related to user settings and function control of the device. The output device 404 may include a display device such as a display screen.

[0168] An embodiment of the present invention further provides a storage medium containing computer executable instructions, wherein the computer executable instructions are used to execute a cable defect detection method based on image recognition when executed by a computer processor, the method comprising:

[0169] Acquire an image of the cable to be tested, and determine defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested, wherein the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested;

[0170] Determine the cable usage parameters associated with the defect parameters to be evaluated based on the historical data of the cable to be tested, collect parameter information values ​​of the cable usage parameters based on a preset frequency, and record changes in the parameter information values;

[0171] The defect detection result of the cable to be tested is determined based on the change of the parameter information value.

[0172] It is worth noting that in the above-mentioned embodiment of the cable defect detection method system based on image recognition, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present invention.

[0173] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the embodiments of the present invention are not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the embodiments of the present invention. Therefore, although the embodiments of the present invention are described in more detail through the above embodiments, the embodiments of the present invention are not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the embodiments of the present invention, and the scope of the embodiments of the present invention is determined by the scope of the appended claims.

Claims

1. A cable defect detection method based on image recognition, characterized in that: include: Acquire an image of the cable to be tested, and determine defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested, wherein the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested; Determine the cable usage parameters associated with the defect parameters to be evaluated based on the historical data of the cable to be tested, collect parameter information values ​​of the cable usage parameters based on a preset frequency, and record changes in the parameter information values; The defect detection result of the cable to be tested is determined based on the change of the parameter information value.

2. The cable defect detection method based on image recognition according to claim 1 is characterized in that: The step of determining the defect parameter to be evaluated of the cable to be tested based on the image of the cable to be tested includes: Identify the image of the cable to be tested based on an image processing algorithm to obtain characteristic information of the cable to be tested; Comparing the characteristic information with preset defect characteristic information to determine the defect type of the cable to be tested; Determine the defect parameter to be evaluated corresponding to the defect type based on the preset association relationship.

3. The cable defect detection method based on image recognition according to claim 1 is characterized in that: The determining the cable usage parameter associated with the defect parameter to be evaluated based on the historical data of the cable to be tested includes: Collecting historical defect data of the cable to be tested within a preset historical time period, wherein the historical defect data includes defect type, defect parameter, occurrence time, occurrence location, and historical usage parameters; Filter out historical usage parameters that meet preset change trends within a preset time period before and after the occurrence of historical defects; When the number of times the historical usage parameter is screened reaches a preset threshold, an association relationship is established between the historical usage parameter and the defect type, defect parameter, occurrence time, and occurrence location; The cable usage parameter associated with the defect parameter to be evaluated is determined based on the association relationship.

4. The cable defect detection method based on image recognition according to claim 3 is characterized in that: The determining, based on the association relationship, the cable usage parameter associated with the defect parameter to be evaluated comprises: When the defect parameter to be evaluated is insulation resistance, the associated cable usage parameter is the actual voltage of the cable to be tested; When the defect parameter to be evaluated is partial discharge, the associated cable usage parameter is the load current of the cable to be tested; When the defect parameter to be evaluated is thermal stability, the associated cable usage parameters are the length, cross section and surface temperature of the cable to be tested; When the defect parameter to be evaluated is compressive strength, the associated cable usage parameter is the pressure value of the cable to be tested; When the defect parameter to be evaluated is the bending degree, the associated cable usage parameter is the bending radius of the cable to be tested.

5. The cable defect detection method based on image recognition according to claim 4 is characterized in that: After collecting the parameter information value of the cable usage parameter based on the preset frequency, the method further includes: Calculating the difference between the usage parameter values ​​of the cable to be tested collected at adjacent time points; When the difference is greater than a preset threshold, the usage parameter value of the cable to be tested collected at the corresponding time point is eliminated.

6. The cable defect detection method based on image recognition according to claim 1, characterized in that: The determining the defect detection result of the cable to be tested based on the change of the parameter information value includes: Recording the usage parameter values ​​of the cable to be tested in chronological order to form time series data; Draw a change curve of the usage parameter of the cable to be tested based on the time series data; Obtaining the fluctuation amplitude of the usage parameter of the cable to be tested based on the change curve; When the similarity between the fluctuation amplitude and the fluctuation amplitude of the preset defect meets a preset threshold, it is determined that the cable to be tested has a defect.

7. The cable defect detection method based on image recognition according to claim 6 is characterized in that: After determining that the tested cable has a defect when the similarity between the fluctuation amplitude and the fluctuation amplitude of the preset defect meets a preset threshold, the method further includes: The cable usage parameters and environmental data of known defect causes are obtained in advance, and the defect causes are used as labels, and the cable usage parameters and environmental data are used as input variables for supervised training to learn the correlation between the defect causes and the environmental data and the cable usage parameters, thereby constructing a cause analysis model; wherein the environmental data includes temperature data, precipitation data and landform data; Obtain environmental data around the cable under test; A cause analysis model is pre-constructed using the acquired environmental data and the cable usage parameter input values, and the cause analysis model predicts the cause of the defect based on the pre-learned correlation between the defect cause and the environmental data and the usage parameters.

8. A cable defect detection system based on image recognition, characterized in that: include: An image acquisition module, used for acquiring an image of the cable to be tested; A first parameter determination module, used to determine the defect parameters to be evaluated of the cable to be tested based on the image of the cable to be tested, wherein the defect parameters to be evaluated are used to evaluate the performance and health status of the cable to be tested; A second parameter determination module, configured to determine the cable usage parameter associated with the defect parameter to be evaluated based on the historical data of the cable to be tested; A parameter value recording module, used to collect parameter information values ​​of the cable usage parameters based on a preset frequency, and record changes in the parameter information values; The result determination module is used to determine the defect detection result of the cable to be tested based on the change of the parameter information value.

9. A cable defect detection device in combination with cable usage parameters, the device comprising: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the cable defect detection method based on image recognition as described in any one of claims 1-7.

10. A storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the cable defect detection method based on image recognition as described in any one of claims 1 to 7 when executed by a computer processor.

Images