Cable insulation defect detection method and system based on infrared imaging

Abstract

The invention belongs to the technical field of cable detection, and discloses a cable insulation defect detection method and system based on infrared imaging. Comprising the following modules: a control test module used for applying an electric excitation sequence test of constant current step boosting and constant voltage step current boosting to a cable; according to the invention, the states of the cable under different operation conditions are actively simulated by setting a constant current step boosting and constant voltage step current rising double-sequence electric excitation test method, resistance type defect heating can be effectively excited by a current rising sequence for a light-load cable, medium type defect heating can be effectively excited by a boosting sequence for various cables, and the test efficiency is improved. By analyzing the response of the defect in the dynamic excitation process and utilizing a defect classification model to analyze the comprehensive feature vector, the defects of different properties such as partial discharge, dielectric loss and overlarge contact resistance can be clearly distinguished, and the classification of cable defect types is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable detection, more particularly, the present application relates to a cable insulation defect detection method and system based on infrared imaging. BACKGROUND

[0002] Power cable is the main artery of power transmission and distribution network, and its insulation state is directly related to the safe and stable operation of the entire power grid. Cable insulation will age over time due to factors such as electricity, heat, mechanical stress, and environmental stress, resulting in defects such as partial discharge, increased dielectric loss, damp insulation, and oxidized connection points. These defects can cause local overheating, and if not detected and addressed in a timely manner, they can lead to insulation breakdown, causing significant economic losses and social impact.

[0003] Therefore, it is necessary to detect the insulation performance of the cable and identify insulation defects. The current detection method mainly involves taking thermal images to determine the temperature of the cable when it is powered on, and then determining the insulation defects of the cable. However, the existing detection method based on infrared imaging has obvious limitations. First, this method relies heavily on the operating load of the cable. When the load current is low, the heat generated by the defect is small, and the temperature rise is not obvious, which can easily lead to missed detection and cannot meet the early warning requirements. Second, the existing method mainly relies on the highest temperature or relative temperature difference in the thermal image obtained by single shooting to make judgments, lacks analysis of the dynamic characteristics of the defect, and cannot effectively distinguish between defect types. SUMMARY

[0004] To solve the problems in the background art, the present application provides a cable insulation defect detection method and system based on infrared imaging.

[0005] To achieve the above-mentioned purposes, the present application provides the following technical solutions: a cable insulation defect detection system based on infrared imaging, comprising the following modules: A control test module for applying a constant current step-up and a constant voltage step-up electrical excitation sequence test to the cable; A data acquisition module for acquiring thermal imaging images of the cable during the electrical excitation sequence test, and obtaining preliminary feature parameters of the abnormal temperature section of the cable according to the thermal imaging images; A defect evaluation module for generating a defect score based on the preliminary feature parameters, and dividing the defect level of the abnormal temperature section according to the defect score; A defect feature module for obtaining a comprehensive feature vector of the abnormal temperature section based on the preliminary feature parameters of the abnormal temperature section; A defect classification module for obtaining historical defect type data, constructing a defect classification model based on the historical defect type data, and classifying the defects of the abnormal temperature section based on the defect classification model.

[0006] Further, the control test module comprises a programmable excitation power supply unit and a sequence control unit; The programmable excitation power supply unit can independently and precisely control the amplitude of the output voltage and current; The sequence control unit is configured to perform the following two independent test sequences: The first sequence is constant current step-up: control the excitation power supply to output a constant basic current value, and then step up the output voltage by a first preset step size until the rated voltage is reached; The second sequence is constant voltage step-up current: control the excitation power supply to output a rated voltage, and then step up the output current by a second preset step size until the rated current is reached.

[0007] Further, the data acquisition module comprises an infrared thermal imager, an image processing unit and a preliminary feature extraction unit; The infrared thermal imager is used to collect infrared radiation data of the cable surface and generate a temperature field distribution image after each excitation step reaches a thermal steady state; The image processing unit is used to filter and denoise the temperature field distribution image, calibrate the temperature, and stitch the image to generate a panoramic thermal image of the cable; The preliminary feature extraction unit is used to analyze the panoramic thermal image, automatically identify and locate one or more abnormal temperature segment areas, and obtain preliminary feature parameters of each abnormal temperature segment area, including the maximum temperature value, the average temperature value, the temperature difference relative to the normal segment, the temperature rise rate, and the area of the abnormal temperature segment area; In the identified abnormal temperature segment area, the maximum value of all pixel temperature values is the maximum temperature value, which directly reflects the extreme heating condition of the abnormal temperature segment area; The arithmetic mean of all pixel temperature values in the abnormal temperature segment area is the average temperature value, which reflects the overall heating level of the abnormal area and represents the total heat power of the defect; On the same cable, select a segment far away from the abnormal temperature segment area and without any abnormal performance, calculate its average temperature to obtain the normal segment average temperature value, and then subtract the normal segment average temperature value from the average temperature value of the abnormal temperature segment area to obtain the temperature difference relative to the normal segment, which directly represents the temperature rise caused by the defect itself; Under the excitation step, record the temperature change curve with time from the start of excitation to the temperature reaching a steady state, and the temperature rise rate is the average slope or first derivative of the curve in the main rising stage, which represents the activity and severity of the defect; The boundary of the abnormal temperature region is determined by an image segmentation algorithm, the total number of pixels within the boundary is calculated, and the pixel area is converted into the actual physical area by spatial calibration according to the distance between the infrared thermal imager and the cable, the focal length of the lens and other parameters, that is, the area of the region is obtained, which reflects the spatial scale of the abnormal temperature section region.

[0008] Further, the process of generating a defect score according to the preliminary characteristic parameters includes: The defect score P is:

[0009] In the formula, , , , and are weight coefficients, which are obtained by training a large amount of historical data, is the maximum temperature value of the abnormal temperature section region, is the maximum temperature reference value, which is the maximum allowable temperature of the cable insulation material or the typical maximum value in the historical data, is the average temperature value, is the average temperature reference value, is the temperature difference value relative to the normal section, is the maximum allowable temperature rise, which is a safety threshold value set according to standards or experience, is the temperature rise rate, is the maximum temperature rise rate reference value, which can be set according to the maximum rate observed in the historical data, is the area of the region, is the maximum area reference value.

[0010] Further, the process of dividing the defect levels of the abnormal temperature section according to the defect score includes: Set the defect score range, and divide each abnormal temperature section into the corresponding defect score range; When P≤X, it is determined that the level of the abnormal temperature section is normal; When X When Y When P>Z, it is determined that the level of the abnormal temperature section is serious.

[0011] Further, the process of obtaining the comprehensive characteristic vector of the abnormal temperature section according to the preliminary characteristic parameters of the abnormal temperature section includes: All potential abnormal temperature section regions are automatically framed in panoramic thermograms of the cable, and defect scores of all abnormal temperature section regions are obtained according to a defect scoring formula, and defect levels of the abnormal temperature section regions are obtained according to the defect scores, for each abnormal temperature section region of each abnormal level or above, a specific excitation ladder has been executed on the abnormal temperature section region and has been kept for a sufficient time to reach thermal stability, and a corresponding single-step feature vector of the abnormal temperature section region is obtained; The single-step feature vector refers to a set of preliminary feature parameters describing the current thermal state of a specific abnormal temperature section region under a specific and stable electrical excitation condition; The preliminary feature parameters further include a current excitation voltage value and a current excitation current value of the abnormal temperature section region in the excitation ladder; The control test module completely executes all excitation ladders of the two sequences of constant current ladder voltage and constant voltage ladder current, and there are i excitation ladders in total; For an abnormal temperature section region, the single-step feature vector of the abnormal temperature section region under the i-th excitation ladder is :

[0012] In the formula, is a temperature difference value of the relative normal section of the i-th excitation ladder, is a temperature rise rate of the i-th excitation ladder, is an area of the region of the i-th excitation ladder, is a current excitation voltage value of the i-th excitation ladder, is a current excitation current value of the i-th excitation ladder; For the same abnormal temperature section region, the corresponding single-step feature vector of the abnormal temperature section region under each excitation ladder is , , ,..., ; According to the time sequence of the excitation ladders, the i single-step feature vectors are spliced head to tail to obtain a comprehensive feature vector of the abnormal temperature section region; The comprehensive feature vector refers to a longer and higher-dimensional vector formed by splicing the single-step feature vectors obtained under each excitation ladder in sequence according to the excitation sequence after the same abnormal temperature section region experiences all different electrical excitation ladders; The comprehensive feature vector is:

[0013] In the formula, is a temperature difference value of the relative normal section of the first excitation ladder, a temperature rise rate of the first excitation step, an area of the first excitation step, a current excitation voltage value of the first excitation step, a current excitation current value of the first excitation step; The comprehensive feature vector constitutes a digital representation of the dynamic response characteristics of the abnormal temperature section area under multi-step electrical excitation.

[0014] Further, the process of acquiring historical defect type data and constructing a defect classification model according to the historical defect type data includes: The historical defect type data is historical case data of the cable, and each case data includes a comprehensive feature vector of an abnormal temperature section area and a defect type label corresponding thereto confirmed through laboratory disassembly; The historical defect type data of the cable is acquired, and for each cable defect section, i.e., an abnormal temperature section area, a comprehensive feature vector thereof is extracted, all extracted comprehensive feature vectors are preprocessed, and the preprocessed comprehensive feature vectors are combined with corresponding defect type labels to form a training data set, the preprocessed comprehensive feature vectors are taken as input, and the defect type labels are taken as output, a gradient boosting decision tree algorithm is used for model training, and a defect classification model is obtained.

[0015] Further, the comprehensive feature vector of the abnormal and above level cable abnormal temperature section area is extracted, the extracted comprehensive feature vector is preprocessed in the same way as in the training stage, the preprocessed comprehensive feature vector is input into the defect classification model, and the output of the defect classification model is obtained, i.e., the defect type classification result of the abnormal temperature section area is obtained; The defect type classification result includes: a dielectric loss type defect, a partial discharge type defect, a resistance type contact defect, and an armored layer defect.

[0016] The cable insulation defect detection method based on infrared imaging includes the following steps: S1: performing a constant current step-up sequence, fixing the current at a lower base value, gradually increasing the voltage to the rated value, then performing a constant voltage step-up sequence, fixing the voltage at the rated value, and gradually increasing the current to the rated value, and waiting for thermal stabilization at each excitation step; S2: in the thermal stabilization state of each excitation step, using an infrared thermal imager to collect panoramic thermal images of the cable, identifying and locating abnormal temperature section areas through image processing algorithms, and calculating preliminary feature parameters, including the highest temperature, the average temperature, the temperature difference relative to the normal section, the temperature rise rate, the area, the current excitation voltage value, and the current excitation current value; S3: Based on the extracted preliminary feature parameters, substitute into the defect scoring formula for calculation to the defect score P, compare with the preset defect score range, and preliminarily divide the defect level; S4: For the abnormal temperature segment area rated as abnormal and above, the single-order feature vector corresponding to the abnormal temperature segment area is obtained according to the preliminary feature parameters, the single-order feature vectors under all excitation steps are spliced in the test order to form a high-dimensional comprehensive feature vector; S5: The constructed comprehensive feature vector is input into the pre-trained gradient boosting decision tree classification model, and the most possible type classification result of the defect of the abnormal temperature segment area is output.

[0017] The technical effects and advantages of the cable insulation defect detection method and system based on infrared imaging are: (1) By setting constant current step-up and constant voltage step-up dual sequence electric excitation test method, the state of the cable under different operating conditions is simulated actively, for light load cable, the current sequence can effectively stimulate the heating of resistance type defects, for various cables, the voltage sequence can effectively stimulate the heating of medium type defects, by analyzing the response of the defects in this dynamic excitation process, and using the defect classification model to analyze the comprehensive feature vector, different nature defects such as partial discharge, dielectric loss and excessive contact resistance can be clearly distinguished, and the classification of cable defect types is realized.

[0018] (2) By setting the defect score, the preliminary feature data collected in the panoramic thermal image is analyzed, the preliminary feature data includes the maximum temperature value, the average temperature value, the temperature difference value relative to the normal segment, the temperature rise rate and the area, the corresponding defect scoring formula is obtained according to the preliminary feature data, the preliminary feature parameters of the cable abnormal temperature segment area are substituted into the defect scoring formula, and the corresponding defect score is obtained, and the defect level of the abnormal temperature segment area is graded according to the defect score. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The system structure diagram of the present application is shown. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0021] Referring to Figure 1 , the cable insulation defect detection system based on infrared imaging includes the following modules: The control test module is configured to apply a constant current step-up voltage and a constant voltage step-up current to the cable. The data acquisition module is configured to acquire thermal imaging images of the cable during the electrical excitation sequence test, and obtain preliminary characteristic parameters of the abnormal temperature section of the cable according to the thermal imaging images. The defect evaluation module is configured to generate a defect score according to the preliminary characteristic parameters, and divide the defect level of the abnormal temperature section according to the defect score. The defect feature module is configured to obtain a comprehensive feature vector of the abnormal temperature section according to the preliminary characteristic parameters of the abnormal temperature section. The defect classification module is configured to acquire historical defect type data, construct a defect classification model according to the historical defect type data, and classify the abnormal temperature section according to the defect classification model.

[0022] It should be further explained that, in the specific implementation process, the control test module includes a programmable excitation power supply unit and a sequence control unit. The programmable excitation power supply unit can independently and precisely control the amplitude of the output voltage and current. The sequence control unit is configured to perform the following two independent test sequences: The first sequence is constant current step-up voltage: control the excitation power supply to output a constant basic current value, and then step up the output voltage by a first preset step size until the rated voltage is reached. The basic current value in the first sequence is 5% to 20% of the rated current value of the cable. The second sequence is constant voltage step-up current: control the excitation power supply to output a rated voltage, and then step up the output current by a second preset step size until the rated current is reached. The first preset step size and the second preset step size are 5%, 10% or 20% of the rated value.

[0023] It should be further explained that, in the specific implementation process, the data acquisition module includes an infrared thermal imager, an image processing unit and a preliminary feature extraction unit. The infrared thermal imager is configured to acquire infrared radiation data of the surface of the cable and generate a temperature field distribution image after each excitation step reaches a thermal steady state. The image processing unit is configured to filter and denoise the temperature field distribution image, calibrate the temperature, and stitch the image to generate a panoramic thermal image of the cable. The preliminary feature extraction unit is configured to analyze the panoramic thermal image, automatically identify and locate one or more abnormal temperature segment regions, and obtain preliminary feature parameters of each abnormal temperature segment region, the preliminary feature parameters including a highest temperature value, an average temperature value, a temperature difference value relative to a normal segment, a temperature rise rate, and a region area of the abnormal temperature segment region. In the identified abnormal temperature segment region, the maximum value of temperature values of all pixel points is the highest temperature value, which directly reflects the extreme heating condition of the abnormal temperature segment region. The highest temperature value is a key indicator for calculating the temperature rise and judging whether it exceeds the material safety threshold. Resistance-type defects usually result in a very high highest temperature value. The average temperature value is obtained by calculating the arithmetic mean of temperature values of all pixel points in the abnormal temperature segment region. The average temperature value reflects the overall heating level of the abnormal region and represents the total heat power of the defect. In the case of similar areas, the higher the average temperature value, the more heat the defect generates, and it is usually more serious. On the same cable, select a segment far from the abnormal temperature segment region and without any abnormal performance, calculate its average temperature to obtain a normal segment average temperature value, and then subtract the normal segment average temperature value from the average temperature value of the abnormal temperature segment region to obtain the temperature difference value relative to the normal segment. The temperature difference value relative to the normal segment directly represents the temperature rise caused by the defect itself. Under the excitation step, record the temperature change curve with time during the process from the start of excitation to the temperature reaching stability. The temperature rise rate is the average slope or first derivative of the curve in the main rising stage. The temperature rise rate represents the activity and severity of the defect. A serious or sharp defect will quickly generate heat, resulting in a large temperature rise rate value, while a slowly developing defect has a slower temperature rise rate. The boundary of the abnormal temperature region is determined by the image segmentation algorithm, the total number of pixels within the boundary is calculated, and the pixel area is converted into the actual physical area according to the distance between the infrared thermal imager and the cable, the focal length of the lens, and other parameters for spatial calibration. The region area is obtained, which reflects the spatial scale of the abnormal temperature segment region.

[0024] It should be further noted that in the specific implementation process, the process of generating a defect score according to the preliminary feature parameters includes: The defect score P is:

[0025] In the formula, , , , and are weight coefficients, which are 0.2, 0.2, 0.3, 0.1 and 0.2, respectively. is the highest temperature value of the abnormal temperature section area, is the highest temperature reference value, taking the highest allowable temperature of the cable insulation material or the typical maximum value in historical data, specifically 100, is the average temperature value, is the average temperature reference value, specifically 80, is the temperature difference value relative to the normal section, is the maximum allowable temperature rise, a safety threshold value set according to standards or experience, specifically 20, is the temperature rise rate, is the maximum temperature rise rate reference value, which can be set according to the maximum rate observed in historical data, specifically 2, is the area of the region, is the maximum area reference value, specifically 200; Suppose the data of a certain abnormal temperature section area are: is 65, is 55, is 15, is 0.8, S is 50, then P is 0.5825.

[0026] It needs to be further explained that in the specific implementation process, the process of dividing the defect level of the abnormal temperature section according to the defect score includes: setting the defect score range, dividing each abnormal temperature section into the corresponding defect score range; when P≤X, determining that the level of the abnormal temperature section is normal level; when X when Y when P>Z, determining that the level of the abnormal temperature section is serious level; when P is 0.5825, and X is 0.5 and Y is 0.6, the abnormal temperature section is determined to be abnormal level, and the abnormal temperature section of the cable is marked.

[0027] It needs to be further explained that in the specific implementation process, the process of obtaining the comprehensive feature vector of the abnormal temperature section according to the preliminary characteristic parameters of the abnormal temperature section includes: automatically framing all potential abnormal temperature section areas in the panoramic thermal image of the cable, obtaining the defect scores of all abnormal temperature section areas according to the defect score formula, obtaining the defect levels of the abnormal temperature section areas according to the defect scores, for each abnormal temperature section area of each abnormal level and above, a specific excitation ladder has been executed and has been kept for enough time to reach thermal stability, so as to obtain the corresponding single-step feature vector thereof; The single-step feature vector refers to a set of preliminary feature parameters describing the current thermal state of a specific abnormal temperature segment region under a specific, stable electrical excitation condition; The preliminary feature parameters further include the current excitation voltage value and the current excitation current value of the abnormal temperature segment region in the excitation ladder; The control test module completely executes all excitation ladders of the two sequences of constant current step-up and constant voltage step-up, assuming a total of i excitation ladders; For an abnormal temperature segment region, its single-step feature vector under the i-th excitation ladder is ,

[0028] In the formula, is the temperature difference of the relative normal segment of the i-th excitation ladder, is the temperature rise rate of the i-th excitation ladder, is the area of the i-th excitation ladder, is the current excitation voltage value of the i-th excitation ladder, is the current excitation current value of the i-th excitation ladder; For the same abnormal temperature segment region, its corresponding single-step feature vector under each excitation ladder is , , ,..., ; According to the time sequence of the excitation ladders, the i single-step feature vectors are spliced head-to-tail to obtain the comprehensive feature vector of the abnormal temperature segment region; The comprehensive feature vector refers to a longer, high-dimensional vector formed by splicing the single-step feature vectors obtained under each excitation ladder in sequence according to the excitation sequence, for the same abnormal temperature segment region after experiencing all different electrical excitation ladders; The comprehensive feature vector is:

[0029] In the formula, is the temperature difference of the relative normal segment of the first excitation ladder, is the temperature rise rate of the first excitation ladder, is the area of the first excitation ladder, is the current excitation voltage value of the first excitation ladder, is the current excitation current value of the first excitation ladder; The comprehensive feature vector constitutes a digital representation of the dynamic response characteristics of the abnormal temperature segment region under multi-step electrical excitation. each There are 5 features, a total of i excitation steps are tested, then is dimensional vector, for example, 10 steps are 50-dimensional; It is not a static value, but a dynamic curve that records the heat characteristics of the abnormal temperature segment area and how it changes with the change of external electrical excitation (voltage / current).

[0030] It needs to be further explained that in the specific implementation process, the process of acquiring historical defect type data and constructing a defect classification model according to the historical defect type data includes: The historical defect type data is historical case data of the cable, and each case data includes a comprehensive feature vector of an abnormal temperature segment area and a defect type label corresponding thereto confirmed by laboratory disassembly; Acquire historical defect type data of the cable, for each cable defect segment, i.e. abnormal temperature segment area, extract its comprehensive feature vector, perform data preprocessing on all extracted comprehensive feature vectors, and combine them with their corresponding defect type labels to form a training data set., the preprocessed comprehensive feature vector is used as input, and the defect type label is used as output, and the gradient boosting decision tree algorithm is used for model training to obtain a defect classification model.

[0031] It needs to be further explained that in the specific implementation process, the process of classifying defects according to the defect classification model for the abnormal temperature segment includes: Extract the comprehensive feature vector of the cable abnormal temperature segment area of the abnormal and above level, perform the same data preprocessing on the extracted comprehensive feature vector as in the training stage, input the preprocessed comprehensive feature vector into the defect classification model, and obtain the output of the defect classification model., i.e. obtain the defect type classification result of the abnormal temperature segment area; The defect type classification result includes: dielectric loss type defect, partial discharge type defect, resistance type contact defect, and armored layer defect, wherein: Resistance type contact defect: its ΔT will significantly increase in the current rise sequence, and almost unchanged in the voltage rise sequence; Dielectric loss type defect: its ΔT will significantly increase in the voltage rise sequence, and change little in the current rise sequence.

[0032] The cable insulation defect detection method based on infrared imaging includes the following steps: S1: Perform a constant current step voltage rise sequence, fix the current at a lower base value, gradually increase the voltage to the rated value, and then perform a constant voltage step current rise sequence, fix the voltage at the rated value, and gradually increase the current to the rated value. Wait for thermal stabilization at each excitation step; S2: In the thermal steady state of each excitation step, a panoramic thermal image of the cable is collected by using an infrared thermal imager, an abnormal temperature section area is identified and located by an image processing algorithm, and preliminary characteristic parameters are calculated, including the highest temperature, the average temperature, the temperature difference relative to the normal section, the temperature rise rate, the area, the current excitation voltage value and the current excitation current value; S3: Based on the extracted preliminary characteristic parameters, the defect score formula is substituted for calculation to the defect score P, and the defect is preliminarily classified according to the preset defect score range; S4: For the abnormal temperature section area rated as abnormal and above, the single-step characteristic vector of the corresponding abnormal temperature section area is obtained according to the preliminary characteristic parameters, and the single-step characteristic vectors of all excitation steps are spliced in the test order to form a high-dimensional comprehensive characteristic vector; S5: The constructed comprehensive characteristic vector is input into the pre-trained gradient boosting decision tree classification model, and the most possible type classification result of the defect of the abnormal temperature section area is output.

[0033] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0034] Finally: The above is only a preferred embodiment of the present application and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A system for detecting defects in the insulation of a cable based on infrared imaging, characterized in that, The application relates to a cable defect detection system, which comprises the following modules: a control test module for applying a constant current step-up voltage and a constant voltage step-up current electric excitation sequence test to a cable; a data acquisition module for acquiring a thermal imaging image of the cable during the electric excitation sequence test, and obtaining preliminary characteristic parameters of an abnormal temperature section of the cable according to the thermal imaging image; a defect evaluation module for generating a defect score according to the preliminary characteristic parameters, and dividing a defect level of the abnormal temperature section according to the defect score; a defect feature module for obtaining a comprehensive feature vector of the abnormal temperature section according to the preliminary characteristic parameters; a defect classification module for acquiring historical defect type data, constructing a defect classification model according to the historical defect type data, and classifying defects of the abnormal temperature section according to the defect classification model.

2. The infrared imaging based cable insulation defect detection system of claim 1, wherein, The control test module comprises a programmable excitation power supply unit and a sequence control unit; the programmable excitation power supply unit can independently and precisely control the amplitude of output voltage and current; the sequence control unit is configured to perform the following two independent test sequences: a first sequence is constant current step-up voltage: the excitation power supply outputs a constant basic current value, and then the output voltage is stepwise increased by a first preset step until the rated voltage is reached; a second sequence is constant voltage step-up current: the excitation power supply outputs the rated voltage, and then the output current is stepwise increased by a second preset step until the rated current is reached.

3. The infrared imaging based cable insulation defect detection system of claim 2, wherein, The data acquisition module comprises an infrared thermal imager, an image processing unit and a preliminary feature extraction unit; the infrared thermal imager is used for acquiring infrared radiation data of the cable surface and generating a temperature field distribution image after each excitation step reaches a thermal steady state; the image processing unit is used for filtering and denoising the temperature field distribution image, temperature calibration and image stitching to generate a panoramic thermal image of the cable; the preliminary feature extraction unit is used for analyzing the panoramic thermal image, automatically identifying and locating one or more abnormal temperature section regions, and obtaining preliminary characteristic parameters of each abnormal temperature section region, the preliminary characteristic parameters comprising a maximum temperature value, an average temperature value, a temperature difference value relative to a normal section, a temperature rise rate and a region area of the abnormal temperature section region; in the identified abnormal temperature section region, the maximum value of the temperature values of all pixel points is the maximum temperature value, which directly reflects the extreme heating condition of the abnormal temperature section region; the arithmetic average of the temperature values of all pixel points in the abnormal temperature section region is the average temperature value, which reflects the overall heating level of the abnormal region and represents the total heat power of the defect; on the same cable, a section far away from the abnormal temperature section region and without any abnormal performance is selected, the average temperature of the section is calculated to obtain a normal section average temperature value, and then the average temperature value of the abnormal temperature section region is subtracted from the normal section average temperature value to obtain the temperature difference value relative to the normal section, which directly represents the temperature rise caused by the defect itself. Under the excitation step, the temperature change curve with time is recorded from the beginning of the excitation to the temperature reaching a steady state, and the temperature rise rate is the average slope or first-order derivative of the curve in the main rising stage, which represents the activity and severity of the defect; The boundary of the abnormal temperature area is determined by an image segmentation algorithm, the total number of pixels within the boundary is calculated, and the pixel area is converted into the actual physical area by spatial calibration according to the distance between the infrared thermal imager and the cable, the focal length of the lens and other parameters, that is, the area of the region is obtained, which reflects the spatial scale of the abnormal temperature area.

4. The infrared imaging based cable insulation defect detection system of claim 3, wherein, The process of generating a defect score according to the preliminary characteristic parameters includes: The defect score P is: In the formula, , , , and are weight coefficients, which are obtained by training a large amount of historical data, is the highest temperature value of the abnormal temperature section area, is the highest temperature reference value, which is the highest allowable temperature of the cable insulation material or the typical maximum value in the historical data, is the average temperature value, is the average temperature reference value, is the temperature difference value relative to the normal section, is the maximum allowable temperature rise, which is a safety threshold value set according to standards or experience, is the temperature rise rate, is the maximum temperature rise rate reference value, which can be set according to the maximum rate observed in the historical data, is the area of the region, is the maximum area reference value.

5. The infrared imaging based cable insulation defect detection system of claim 4, wherein, The process of classifying the defect level of the abnormal temperature section according to the defect score includes: Set the defect score range, and divide each abnormal temperature section into the corresponding defect score range; When P≤X, it is determined that the level of the abnormal temperature section is normal; When X When Y When P>Z, it is determined that the level of the abnormal temperature section is serious.

6. The infrared imaging based cable insulation defect detection system of claim 5, wherein, The process of obtaining the comprehensive characteristic vector of the abnormal temperature section according to the preliminary characteristic parameters of the abnormal temperature section includes: All potential abnormal temperature section areas are automatically framed in the panoramic thermal image of the cable, the defect scores of all abnormal temperature section areas are obtained according to the defect score formula, the defect levels of the abnormal temperature section areas are obtained according to the defect scores, and for each abnormal and above level of the abnormal temperature section area, a specific excitation step has been performed and has been maintained for a sufficient time to reach thermal stability, so that the corresponding single-step characteristic vector is obtained; The single-step characteristic vector refers to a set of preliminary characteristic parameters describing the current thermal state of a specific abnormal temperature section area under a specific, stable electrical excitation condition; The preliminary characteristic parameters further include the current excitation voltage value and the current excitation current value of the abnormal temperature section area in the excitation step; The control test module completely performs all excitation steps of the two sequences of constant current step-up and constant voltage step-up, and there are i excitation steps in total; For an abnormal temperature section region, its single-step eigenvector under the i-th excitation step is: ​ wherein is the temperature difference of the relative normal section of the i-th excitation step, is the temperature rise rate of the i-th excitation step, is the area of the i-th excitation step, is the current excitation voltage value of the i-th excitation step, is the current excitation current value of the i-th excitation step; For the same abnormal temperature section area, its corresponding single step feature vector under each excitation step is: , , ,..., ; According to the time sequence of the excitation steps, the i single-step characteristic vectors are spliced head-to-tail to obtain the comprehensive characteristic vector of the abnormal temperature section area; The comprehensive characteristic vector refers to a longer, high-dimensional vector formed by splicing the single-step characteristic vectors obtained under each excitation step in sequence after the same abnormal temperature section area has experienced all different electrical excitation steps; Synthetic feature vector is: wherein is the temperature difference of the relative normal section of the first excitation step, is the temperature rise rate of the first excitation step, is the area of the first excitation step, is the current excitation voltage value of the first excitation step, is the current excitation current value of the first excitation step; The comprehensive characteristic vector constitutes a digital representation of the dynamic response characteristics of the abnormal temperature section area under multi-step electrical excitation.

7. The infrared imaging based cable insulation defect detection system of claim 6, wherein, The process of obtaining historical defect type data and constructing a defect classification model according to the historical defect type data includes: The historical defect type data is historical case data of the cable, and each case data includes a comprehensive characteristic vector of an abnormal temperature section area and a defect type label confirmed by laboratory disassembly and corresponding to the abnormal temperature section area; The historical defect type data of the cable is acquired, a comprehensive feature vector of each cable defect section, i.e., an abnormal temperature section area, is extracted, data preprocessing is performed on all the extracted comprehensive feature vectors, and the preprocessed comprehensive feature vectors are combined with corresponding defect type labels to form a training data set, the preprocessed comprehensive feature vectors are taken as input, and the defect type labels are taken as output, a gradient boosting decision tree algorithm is used for model training, and a defect classification model is obtained.

8. The infrared imaging based cable insulation defect detection system of claim 7, wherein, The comprehensive feature vector of the abnormal temperature section area of the cable is extracted, the same data preprocessing as in the training stage is performed on the extracted comprehensive feature vector, the preprocessed comprehensive feature vector is input into the defect classification model, and the output of the defect classification model is obtained, i.e., the defect type classification result of the abnormal temperature section area is obtained. The defect type classification result includes: a dielectric loss type defect, a partial discharge type defect, a resistance type contact defect, and an armored layer defect.

9. A method for detecting defects in cable insulation based on infrared imaging, implemented based on the system for detecting defects in cable insulation based on infrared imaging according to any one of claims 1-8, characterized in that, The method comprises the following steps: S1: a constant current step-up sequence is performed, the current is fixed at a lower base value, the voltage is gradually increased to the rated value, a constant voltage step-up sequence is performed, the voltage is fixed at the rated value, the current is gradually increased to the rated value, and heat stabilization is required at each excitation step; S2: in the heat stabilization state of each excitation step, a panoramic thermograph of the cable is collected by using an infrared thermal imager, an abnormal temperature section area is identified and located by using an image processing algorithm, and preliminary feature parameters are calculated, the preliminary feature parameters include the highest temperature, the average temperature, the temperature difference relative to the normal section, the temperature rise rate, the area, the current excitation voltage value, and the current excitation current value; S3: based on the extracted preliminary feature parameters, the defect score P is calculated by substituting the defect score formula, and the defect is preliminarily classified according to the preset defect score range; S4: for the abnormal temperature section area rated as abnormal and above, a single-step feature vector of the corresponding abnormal temperature section area is obtained according to the preliminary feature parameters, the single-step feature vectors of the abnormal temperature section area under all excitation steps are spliced in the test order to form a high-dimensional comprehensive feature vector; S5: the constructed comprehensive feature vector is input into the pre-trained gradient boosting decision tree classification model, and the most possible type classification result of the defect of the abnormal temperature section area is output.

Images