Multispectral composite insulator defect detection method

Abstract

The invention discloses a multispectral composite insulator defect detection method, and relates to the technical field of big data analysis. By setting the defect trend judgment threshold and predicting the reflectivity of the composite insulator, the abnormal change of the reflectivity is effectively identified, operation and maintenance personnel are helped to take maintenance measures in time, the risk of sudden equipment failures is reduced, the maintenance and operation cost is reduced, the service life of the composite insulator is prolonged, and the overall safety of a power grid system is improved. The defect trend chart is established based on historical data, the change condition of the reflectivity of the composite insulator along with time is visually displayed, reference data can be dynamically adjusted, data support is provided for subsequent detection, the detection accuracy is ensured, and the equipment management efficiency is improved. A dynamic reflectivity compensation mechanism is adopted, the detection process is optimized, errors caused by environmental factor changes are effectively solved, the precision of detection data is improved, the field operation complexity is reduced, and the efficiency and reliability of the detection process are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a multi-spectrum composite insulator defect detection method. Background Art

[0002] Composite insulators are of great significance in ultra-high voltage transmission lines because they have excellent insulation performance, corrosion resistance and anti-aging ability, which can effectively ensure the safe and stable operation of transmission lines. Compared with traditional porcelain or glass insulators, composite insulators are lighter and have better anti-pollution flashover performance, especially in harsh environments such as high pollution and high humidity. Therefore, composite insulators are widely used in long-distance transmission and complex environmental conditions. They are an indispensable and important component in modern power transmission systems, ensuring the efficient and continuous operation of transmission lines.

[0003] During the use of composite insulators, regular inspection is the key to ensure their normal function and safety. However, the inspection results are often significantly affected by the ambient humidity. Humidity changes not only cause fluctuations in the reflectivity characteristic value, thereby affecting the accuracy of defect detection, but also make the inspection process more complicated and time-consuming. This detection characteristic affected by humidity means that in actual applications, relevant technicians must spend more time on data collection and analysis, increasing the cost of inspection. Summary of the invention

[0004] The object of the present invention is to provide a multi-spectrum composite insulator defect detection method to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a multi-spectrum composite insulator defect detection method, the defect detection method comprising the following steps:

[0006] Step S1: within the humidity detection standard range, using unoperated composite insulators to simulate defect detection, setting the humidity to the minimum humidity, the maximum humidity and the median humidity for simulated detection, respectively, obtaining image data of simulated detection at each humidity to construct a specific humidity image data set;

[0007] Step S1-1, determining the standard range of humidity detection through the composite insulator defect detection industry standard;

[0008] Step S1-2: According to the standard range of humidity detection, unoperated composite insulators are used to simulate defect detection, and the humidity conditions for simulated defect detection are set by limiting the standard range of humidity detection, wherein the humidity conditions include minimum humidity, maximum humidity and median humidity, and image data for simulated detection is obtained, and a specific humidity image data set is constructed based on the obtained image data.

[0009] The composite insulator defect detection industry standard is used to ensure the standardization, effectiveness and reliability of the detection process. It is formulated by the state or industry competent authorities and covers the performance requirements, detection methods, data processing and evaluation criteria of composite insulators.

[0010] By conducting simulated defect detection on unoperated composite insulators within the standard range of humidity detection and setting the minimum, maximum and median humidity values ​​to construct a specific humidity image dataset, it is helpful to accurately evaluate the impact of humidity on the reflectivity of composite insulators, establish comprehensive reflectivity benchmark data, and improve the accuracy and reliability of defect detection.

[0011] Step S2, extracting the reflectivity of the simulated detected image data according to the specific humidity image data set, and generating compensation reference data;

[0012] Step S2-1, obtaining the reflectivity of the composite insulator in different bands according to the specific humidity image data set, wherein the reflectivity is calculated using the following formula:

[0013]

[0014] Where R(y) is the reflectivity of the composite insulator under the specific wavelength y; L target It is expressed as the reflection brightness of the composite insulator at a specific wavelength y; L rference It is expressed as the reflected brightness of the reference standard white plate at the same specific wavelength y;

[0015] Step S2-2: Perform reflectivity analysis and record on the image data of the simulated detection, store the reflectivities corresponding to the humidity conditions of the minimum humidity, the maximum humidity and the median humidity, and generate compensation reference data.

[0016] By extracting the reflectivity of the specific humidity image data set, the reflectivity of the composite insulator in different bands is calculated using the formula, and the reflectivity data under various humidity conditions is analyzed and recorded to generate compensation benchmark data.

[0017] Step S3, setting a reflection dynamic threshold, obtaining composite insulator image data of the actual detection point, and calculating the reflectivity of the composite insulator image data of the detection point by compensating the reference data in combination with the Lagrange interpolation method, and recording the calculated reflectivity as compensated reflectivity data, and dynamically setting the compensation reference data by combining the compensated reflectivity data with the reflection dynamic threshold;

[0018] Step S3-1, determine the specific location of the composite insulator to be detected, obtain the image data and environmental humidity data of the composite insulator through the sensor, and record the detected environmental humidity of the composite insulator as the real-time detection humidity C actual ;

[0019] Step S3-2: Use the compensation reference data in combination with the Lagrange interpolation method to perform reflectivity compensation calculation on the composite insulator image data of the detection point. The reflectivity compensation calculation uses the following formula:

[0020]

[0021] Where, R(K) represents the reflectivity of the composite insulator when the humidity value is K; K represents the ambient humidity value of the composite insulator; n represents the number of data for compensating the benchmark data; L i (K) is expressed as Lagrangian basis function;

[0022] The Lagrangian basis function calculation formula is as follows:

[0023]

[0024] In the formula, K1, K2 and K3 represent the minimum humidity, median humidity and maximum humidity respectively;

[0025] The detection location is usually at a key point of a power transmission line, such as a substation or a crossing point of a high-voltage line.

[0026] By setting the dynamic reflection threshold, the image data of the composite insulator at the actual detection point is obtained, and the reflectivity compensation calculation is performed by combining the compensation benchmark data with the Lagrange interpolation method to obtain the compensated reflectivity data under real-time detection humidity conditions, thereby dynamically adjusting the compensation benchmark data to ensure accurate defect judgment of the composite insulator.

[0027] The specific process of dynamically setting the compensation reference data by combining the compensation reflectivity data with the characteristic dynamic threshold is as follows: Setting the reflection dynamic threshold F max , obtain the ambient temperature C of the composite insulator at the detection location when it was last detected last , according to the real-time detection humidity C actual Combined reflection dynamic threshold F max The compensation reference data is dynamically set, and after the reflectivity compensation calculation is completed for the composite insulator image data of the detection point, the compensation reference data is restored to the reflectivity corresponding to the minimum humidity, the maximum humidity and the median humidity;

[0028] When C actual -F max ≥0, use the ambient temperature C at the time of the last detection last The corresponding reflectivity replaces the reflectivity corresponding to the lowest humidity in the compensation benchmark data. After the compensation benchmark data is dynamically set, the set compensation benchmark data is used to perform compensation calculation on the reflectivity of the composite insulator at the actual detection point;

[0029] When C actual-F max When <0, the ambient temperature C of the last detection is used last The corresponding reflectivity replaces the reflectivity corresponding to the highest humidity in the compensation benchmark data. After the compensation benchmark data is dynamically set, the set compensation benchmark data is used to perform compensation calculation on the reflectivity of the composite insulator at the actual detection point.

[0030] By combining the compensated reflectivity data with the characteristic dynamic threshold, after setting the reflection dynamic threshold, the compensated benchmark data is dynamically adjusted according to the real-time detected humidity and the ambient temperature of the last detection: when the real-time detected humidity minus the reflection dynamic threshold is greater than or equal to zero, the reflectivity of the lowest humidity is replaced by the reflectivity corresponding to the ambient temperature at the last detection; when the real-time detected humidity minus the reflection dynamic threshold is less than zero, the reflectivity of the highest humidity is replaced. The adjusted compensated benchmark data is used for the reflectivity compensation calculation of the actual detection point to ensure the accuracy of the detection result.

[0031] Step S4, setting a defect reflectivity threshold, and establishing a detection archive set, performing defect judgment on the image data after compensation calculation, and recording and storing the defect judgment result in the detection archive set;

[0032] Step S4-1, obtain historical data of composite insulator defect detection, and set the defect reflectivity threshold Q according to the historical data max , defect reflectivity threshold Q max Set the calculation to use the following formula:

[0033] Q max =Zc*N;

[0034] In the formula, Q max It is represented by the reflectivity threshold of composite insulator detection; Z is represented by the average value of normal reflectivity data of composite insulator historical defect detection; c is represented by the defect fluctuation adjustable coefficient of composite insulator defect detection, and c can usually be selected from 1.5 to 3, depending on your requirements for false alarm rate and missed alarm rate; N is represented by the standard deviation of normal reflectivity data of composite insulator historical defect detection;

[0035] Step S4-2: Establish a detection archive set, which includes a unique detection position identifier, detection time, compensated reflectivity, compensated reflectivity and defect judgment result, and perform defect judgment on the image data of the composite insulator at the detection point. The judgment process is as follows:

[0036] When R(K)≥Q max When the composite insulator at the detection point is judged to be defective, the composite insulator at the detection point is marked as degraded and stored in the detection archive set;

[0037] When R(K)<Qmax When the composite insulator at the detection point is judged as having no defects, the composite insulator at the detection point is marked as normal and stored in the detection archive set.

[0038] By setting the defect reflectivity threshold and establishing a detection archive collection, defect judgment is performed on the composite insulator image data after compensation calculation, and the defect reflectivity threshold is calculated based on historical data to achieve effective detection and management of composite insulators.

[0039] Step S5, extracting historical data at the same position from the detection archive set, analyzing and extracting the compensation reflectivity data of each detection, and constructing a defect trend graph based on the compensation reflectivity data of each detection;

[0040] Step S5-1, read the unique detection location identifier in the detection archive set, use the location data of the composite insulator at the detection point to perform a matching search, and obtain the historical data of the detection point;

[0041] Step S5-2: extract the compensated reflectivity data of each test according to the historical data of the test point, and construct a defect trend graph according to the compensated reflectivity data of each test and the corresponding test time data.

[0042] By extracting historical data of the same location from the inspection archive collection, analyzing and extracting the compensated reflectivity data of each inspection, and constructing a defect trend graph, it is possible to intuitively display the change trend of the reflectivity of the composite insulator over time, which helps to timely discover potential degradation or abnormalities and improve the accuracy of inspection and the efficiency of preventive maintenance.

[0043] Step S6, setting a defect trend judgment threshold, predicting the compensation reflectivity of the composite insulator at the corresponding position for the next inspection based on the defect trend graph, and using the predicted compensation reflectivity to calculate the growth rate slope of the defect trend graph, and judging the trend growth rate slope of the defect trend graph in combination with the defect trend.

[0044] Set the defect trend judgment threshold P max , the reflectivity of the composite insulator to be tested next time is predicted based on the compensated reflectivity data at each data point in the defect trend graph. The calculation formula used for the composite insulator reflectivity prediction is as follows:

[0045] B next =a*RR(K)+(1-a))*B previous ;

[0046] In the formula, B next It is expressed as the predicted compensating reflectivity of the composite insulator for the next test; a is the smoothing coefficient, ranging from 0 to 1; B previousIt is expressed as the compensated reflectivity of the composite insulator detected last time;

[0047] The predicted compensation reflectivity is used to calculate the growth rate slope of the defect trend graph, and the growth rate slope is calculated using the following formula:

[0048]

[0049] Where S represents the slope of the defect trend graph growth rate calculated based on the predicted compensation reflectivity; T next Indicates the time of the next inspection; T(K) indicates the inspection time of the current composite insulator; T next -T(K) represents the time difference between two inspections of the composite insulator;

[0050] When P max When >S, the predicted compensation reflectivity growth rate slope in the defect trend diagram is normal, indicating that there will be no defects in the next composite insulator inspection result and it is in a normal state;

[0051] When P max When ≤S, the growth rate slope of the predicted compensation reflectivity in the defect trend diagram is abnormal, indicating that the next inspection result of the composite insulator will be defective and in a degraded state, and an early warning signal that the composite insulator at the inspection point is about to become abnormal is issued.

[0052] By setting the defect trend judgment threshold, and combining the trend chart to predict the reflectivity of the composite insulator and calculate the growth rate slope, the abnormal change of reflectivity can be identified in advance, and the degradation of the composite insulator can be warned. Timely maintenance measures can be taken to reduce the risk of equipment failure, improve operational reliability and safety, and optimize the detection process and improve detection efficiency.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. The present invention can effectively identify abnormal changes in reflectivity by setting a defect trend judgment threshold and predicting the reflectivity of the composite insulator. By predicting the compensation reflectivity and combining the growth rate slope in the defect trend graph, the system can warn of the degradation of the composite insulator before the problem occurs. The early warning mechanism can help operation and maintenance personnel take maintenance measures in a timely manner, thereby reducing sudden equipment failures, reducing emergency maintenance and operation costs, further extending the service life of composite insulators, and improving the overall safety of the power grid system.

[0055] 2. The present invention establishes a defect trend chart based on historical data to intuitively display the change of the reflectivity of the composite insulator over time. Through the time series analysis of the reflectivity and the dynamically updated defect trend chart, not only can the benchmark data be dynamically adjusted, but also data support can be provided for subsequent detection. This makes the detection more coherent, can capture the potential degradation trend of the composite insulator, ensure the accuracy of the detection, and help formulate a more effective preventive maintenance plan to improve the efficiency of equipment management.

[0056] 3. The present invention optimizes the detection process by setting up a dynamic reflectivity compensation mechanism. Combining real-time humidity data, historical environmental data and compensation benchmark data, the Lagrange interpolation method is used to perform reflectivity compensation calculations to achieve accurate detection of composite insulators under different environmental conditions. This effectively solves the errors caused by changes in environmental factors, improves the accuracy of detection data, reduces the complexity of on-site operations, and greatly improves the efficiency and reliability of the detection process. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic diagram of a multi-spectrum composite insulator defect detection method of the present invention; DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Embodiment 1: Figure 1 As shown, the present invention provides a technical solution, a multi-spectrum composite insulator defect detection method, the defect detection method comprises the following steps:

[0060] Step S1: within the humidity detection standard range, using unoperated composite insulators to simulate defect detection, setting the humidity to the minimum humidity, the maximum humidity and the median humidity for simulated detection, respectively, obtaining image data of simulated detection at each humidity to construct a specific humidity image data set;

[0061] Step S1-1, determining the standard range of humidity detection through the composite insulator defect detection industry standard;

[0062] Step S1-2: According to the standard range of humidity detection, unoperated composite insulators are used to simulate defect detection, and the humidity conditions for simulated defect detection are set by limiting the standard range of humidity detection, wherein the humidity conditions include minimum humidity, maximum humidity and median humidity, and image data for simulated detection is obtained, and a specific humidity image data set is constructed based on the obtained image data.

[0063] Step S2, extracting the reflectivity of the simulated detected image data according to the specific humidity image data set, and generating compensation reference data;

[0064] Step S2-1, obtaining the reflectivity of the composite insulator in different bands according to the specific humidity image data set, wherein the reflectivity is calculated using the following formula:

[0065]

[0066] Where R(y) is the reflectivity of the composite insulator under the specific wavelength y; L target It is expressed as the reflection brightness of the composite insulator at a specific wavelength y; L rference It is expressed as the reflected brightness of the reference standard white plate at the same specific wavelength y;

[0067] Step S2-2: Perform reflectivity analysis and record on the image data of the simulated detection, store the reflectivities corresponding to the humidity conditions of the minimum humidity, the maximum humidity and the median humidity, and generate compensation reference data.

[0068] Step S3, setting a reflection dynamic threshold, obtaining composite insulator image data of the actual detection point, and calculating the reflectivity of the composite insulator image data of the detection point by compensating the reference data in combination with the Lagrange interpolation method, and recording the calculated reflectivity as compensated reflectivity data, and dynamically setting the compensation reference data by combining the compensated reflectivity data with the reflection dynamic threshold;

[0069] Step S3-1, determine the specific location of the composite insulator to be detected, obtain the image data and environmental humidity data of the composite insulator through the sensor, and record the detected environmental humidity of the composite insulator as the real-time detection humidity C actual ;

[0070] Step S3-2: Use the compensation reference data in combination with the Lagrange interpolation method to perform reflectivity compensation calculation on the composite insulator image data of the detection point. The reflectivity compensation calculation uses the following formula:

[0071]

[0072] Where, R(K) represents the reflectivity of the composite insulator when the humidity value is K; K represents the ambient humidity value of the composite insulator; n represents the number of data for compensating the benchmark data; L i (K) is expressed as Lagrangian basis function;

[0073] The Lagrangian basis function calculation formula is as follows:

[0074]

[0075] Where K1, K2 and K3 represent the minimum humidity, median humidity and maximum humidity respectively.

[0076] The specific process of dynamically setting the compensation reference data by combining the compensation reflectivity data with the characteristic dynamic threshold is as follows: Setting the reflection dynamic threshold F max , obtain the ambient temperature C of the composite insulator at the detection location when it was last detected last , according to the real-time detection humidity C actual Combined reflection dynamic threshold F max The compensation reference data is dynamically set, and after the reflectivity compensation calculation is completed for the composite insulator image data of the detection point, the compensation reference data is restored to the reflectivity corresponding to the minimum humidity, the maximum humidity and the median humidity;

[0077] When C actual -F max ≥0, use the ambient temperature C at the time of the last detection last The corresponding reflectivity replaces the reflectivity corresponding to the lowest humidity in the compensation benchmark data. After the compensation benchmark data is dynamically set, the set compensation benchmark data is used to perform compensation calculation on the reflectivity of the composite insulator at the actual detection point;

[0078] When C actual -F max When <0, the ambient temperature C of the last detection is used last The corresponding reflectivity replaces the reflectivity corresponding to the highest humidity in the compensation benchmark data. After the compensation benchmark data is dynamically set, the set compensation benchmark data is used to perform compensation calculation on the reflectivity of the composite insulator at the actual detection point.

[0079] Step S4, setting a defect reflectivity threshold, and establishing a detection archive set, performing defect judgment on the image data after compensation calculation, and recording and storing the defect judgment result in the detection archive set;

[0080] Step S4-1, obtain historical data of composite insulator defect detection, and set the defect reflectivity threshold Q according to the historical data max , defect reflectivity threshold Q max Set the calculation to use the following formula:

[0081] Q max =Zc*N;

[0082] In the formula, Q max It represents the reflectivity threshold of composite insulator detection; Z represents the average value of normal reflectivity data of historical defect detection of composite insulator; c represents the defect fluctuation adjustable coefficient of composite insulator defect detection; N represents the standard deviation of normal reflectivity data of historical defect detection of composite insulator;

[0083] Step S4-2: Establish a detection archive set, which includes a unique detection position identifier, detection time, compensated reflectivity, compensated reflectivity and defect judgment result, and perform defect judgment on the image data of the composite insulator at the detection point. The judgment process is as follows:

[0084] When R(K)≥Q max When the composite insulator at the detection point is judged to be defective, the composite insulator at the detection point is marked as degraded and stored in the detection archive set;

[0085] When R(K)<Q max When the composite insulator at the detection point is judged as having no defects, the composite insulator at the detection point is marked as normal and stored in the detection archive set.

[0086] Step S5, extracting historical data at the same position from the detection archive set, analyzing and extracting the compensation reflectivity data of each detection, and constructing a defect trend graph based on the compensation reflectivity data of each detection;

[0087] Step S5-1, read the unique detection location identifier in the detection archive set, use the location data of the composite insulator at the detection point to perform a matching search, and obtain the historical data of the detection point;

[0088] Step S5-2: extract the compensated reflectivity data of each test according to the historical data of the test point, and construct a defect trend graph according to the compensated reflectivity data of each test and the corresponding test time data.

[0089] Step S6, setting a defect trend judgment threshold, predicting the compensation reflectivity of the composite insulator at the corresponding position for the next inspection based on the defect trend graph, and using the predicted compensation reflectivity to calculate the growth rate slope of the defect trend graph, and judging the trend growth rate slope of the defect trend graph in combination with the defect trend.

[0090] Set the defect trend judgment threshold P max, the reflectivity of the composite insulator to be tested next time is predicted based on the compensated reflectivity data at each data point in the defect trend graph. The calculation formula used for the composite insulator reflectivity prediction is as follows:

[0091] B next =a*RR(K)+(1-a)*B previous ;

[0092] In the formula, B next It is expressed as the predicted compensating reflectivity of the composite insulator for the next test; a is the smoothing coefficient, ranging from 0 to 1; B previous It is expressed as the compensated reflectivity of the composite insulator detected last time;

[0093] The predicted compensation reflectivity is used to calculate the growth rate slope of the defect trend graph, and the growth rate slope is calculated using the following formula:

[0094]

[0095] Where S represents the slope of the defect trend graph growth rate calculated based on the predicted compensation reflectivity; T next Indicates the time of the next inspection; T(K) indicates the inspection time of the current composite insulator; T next -T(K) represents the time difference between two inspections of the composite insulator;

[0096] When P max When >S, the predicted compensation reflectivity growth rate slope in the defect trend diagram is normal, indicating that there will be no defects in the next composite insulator inspection result and it is in a normal state;

[0097] When P max When ≤S, the growth rate slope of the predicted compensation reflectivity in the defect trend diagram is abnormal, indicating that the next inspection result of the composite insulator will be defective and in a degraded state, and an early warning signal that the composite insulator at the inspection point is about to become abnormal is issued.

[0098] For example, the composite insulator test point is point A, and the humidity test standard range is 20% to 80%, the minimum humidity is 20%, the maximum humidity is 80%, and the median humidity is 50%. The reflection brightness of the composite insulator is at wavelength y = 600nm, L target =500. The reflective brightness of a standard white plate at wavelength y = 600nm, L rference = 600. The current humidity value C of the composite insulator test point A actual =45. The humidity value of the composite insulator test point A during the last test is C last=50. Defect fluctuation adjustable coefficient c = 1.5. The historical detection average reflectivity Z of composite insulator detection point A = 0.8, the historical detection standard deviation N = 0.05, the smoothing coefficient a = 0.8, and the next detection time of composite insulator detection point A is T away from the current detection time. next -T(K) = 10 days, defect trend judgment threshold P max =0.002.

[0099] Substitute the known data into the formula to calculate the reflectivity of the composite insulator at the composite insulator detection point A. The calculation process is as follows:

[0100] R(y)=500 / 600=0.8333;

[0101] It can be obtained that the reflectivity of the composite insulator at a wavelength y of 600nm is 0.8333;

[0102] The calculation results are combined with the Lagrange difference method to compensate the reflectivity, and the known data are brought into the formula for calculation. The calculation process is as follows:

[0103] R(45)=0.75*0.00972+0.8*-0.972+0.85*-0.0708=-0.7649;

[0104] According to calculations, when the humidity is 45%, the reflectivity after compensation calculation is about 0.7649;

[0105] Bring in known data to calculate the defect reflectivity threshold. The calculation process is as follows:

[0106] Q max =0.8+1.5*1.05=0.875;

[0107] According to calculations, the defect reflectivity threshold for judging composite insulator defects is 0.875;

[0108] Get the compensation reflectivity B of the composite insulator detection point A during the last detection previous =0.7, calculate the reflectivity after the next detection compensation calculation, the calculation process is as follows:

[0109] B next =0.8*0.7+0.2*0.7649=0.71298;

[0110] Bring in known data to calculate the slope of the defect trend growth rate in the defect trend graph. The calculation process is as follows:

[0111] S=(0.71298-0.7) / 10=0.001298;

[0112] The growth rate slope of the defect trend graph can be obtained as S = 0.001298;

[0113] Since 0.001298<0.002, we get S<P max The predicted growth rate slope of the compensation reflectivity in the defect trend diagram is normal, indicating that there will be no defects in the next inspection result of the composite insulator and it is in a normal state.

[0114] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A multi-spectral composite insulator defect detection method, characterized in that: The defect detection method comprises the following steps: Step S1: within the humidity detection standard range, using unoperated composite insulators to simulate defect detection, setting the humidity to the minimum humidity, the maximum humidity and the median humidity for simulated detection, respectively, obtaining image data of simulated detection at each humidity to construct a specific humidity image data set; Step S2, extracting the reflectivity of the simulated detected image data according to the specific humidity image data set, and generating compensation reference data; Step S3, setting a reflection dynamic threshold, obtaining composite insulator image data of the actual detection point, and calculating the reflectivity of the composite insulator image data of the detection point by compensating the reference data in combination with the Lagrange interpolation method, and recording the calculated reflectivity as compensated reflectivity data, and dynamically setting the compensation reference data by combining the compensated reflectivity data with the reflection dynamic threshold; Step S4, setting a defect reflectivity threshold, and establishing a detection archive set, performing defect judgment on the image data after compensation calculation, and recording and storing the defect judgment result in the detection archive set; Step S5, extracting historical data at the same position from the detection archive set, analyzing and extracting the compensation reflectivity data of each detection, and constructing a defect trend graph based on the compensation reflectivity data of each detection; Step S6, setting a defect trend judgment threshold, predicting the compensation reflectivity of the composite insulator at the corresponding position for the next inspection based on the defect trend graph, and using the predicted compensation reflectivity to calculate the growth rate slope of the defect trend graph, and judging the trend growth rate slope of the defect trend graph in combination with the defect trend.

2. A multi-spectrum composite insulator defect detection method according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1, determining the standard range of humidity detection; Step S1-2: According to the standard range of humidity detection, unoperated composite insulators are used to simulate defect detection, and the humidity conditions for simulated defect detection are set by limiting the standard range of humidity detection, wherein the humidity conditions include minimum humidity, maximum humidity and median humidity, and image data for simulated detection is obtained, and a specific humidity image data set is constructed based on the obtained image data.

3. A multi-spectrum composite insulator defect detection method according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1, obtaining the reflectivity of the composite insulator in different bands according to the specific humidity image data set, wherein the reflectivity is calculated using the following formula: Where R(y) is the reflectivity of the composite insulator under the specific wavelength y; L target It is expressed as the reflection brightness of the composite insulator at a specific wavelength y; L rference It is expressed as the reflected brightness of the reference standard white plate at the same specific wavelength y; Step S2-2: Perform reflectivity analysis and record on the image data of the simulated detection, store the reflectivities corresponding to the humidity conditions of the minimum humidity, the maximum humidity and the median humidity, and generate compensation reference data.

4. A multi-spectrum composite insulator defect detection method according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1, determine the specific location of the composite insulator to be detected, obtain the image data and environmental humidity data of the composite insulator through the sensor, and record the detected environmental humidity of the composite insulator as the real-time detection humidity C actual ; Step S3-2: Use the compensation reference data in combination with the Lagrange interpolation method to perform reflectivity compensation calculation on the composite insulator image data of the detection point. The reflectivity compensation calculation uses the following formula: Where, R(K) represents the reflectivity of the composite insulator when the humidity value is K; K represents the ambient humidity value of the composite insulator; n represents the number of data for compensating the benchmark data; L i (K) is expressed as Lagrangian basis function; The Lagrangian basis function calculation formula is as follows: Where K1, K2 and K3 represent the minimum humidity, median humidity and maximum humidity respectively.

5. A multi-spectrum composite insulator defect detection method according to claim 4, characterized in that: In step S3, the specific process of dynamically setting the compensation reference data by combining the compensation reflectivity data with the characteristic dynamic threshold is as follows: Setting the reflection dynamic threshold F max , obtain the ambient temperature C of the composite insulator at the detection location when it was last detected last , according to the real-time detection humidity C actual Combined reflection dynamic threshold F max The compensation reference data is dynamically set, and after the reflectivity compensation calculation is completed for the composite insulator image data of the detection point, the compensation reference data is restored to the reflectivity corresponding to the minimum humidity, the maximum humidity and the median humidity; When C actual -F max ≥0, use the ambient temperature C at the time of the last detection last The corresponding reflectivity replaces the reflectivity corresponding to the lowest humidity in the compensation benchmark data. After the compensation benchmark data is dynamically set, the set compensation benchmark data is used to perform compensation calculation on the reflectivity of the composite insulator at the actual detection point; When C actual -F max When <0, the ambient temperature C of the last detection is used last The corresponding reflectivity replaces the reflectivity corresponding to the highest humidity in the compensation benchmark data. After the compensation benchmark data is dynamically set, the set compensation benchmark data is used to perform compensation calculation on the reflectivity of the composite insulator at the actual detection point.

6. A multi-spectrum composite insulator defect detection method according to claim 5, characterized in that: The specific steps of step S4 are as follows: Step S4-1, obtain historical data of composite insulator defect detection, and set the defect reflectivity threshold Q according to the historical data max , defect reflectivity threshold Q max Set the calculation to use the following formula: Q max =Z-c*N; In the formula, Q max It represents the reflectivity threshold of composite insulator detection; Z represents the average value of normal reflectivity data of historical defect detection of composite insulator; c represents the defect fluctuation adjustable coefficient of composite insulator defect detection; N represents the standard deviation of normal reflectivity data of historical defect detection of composite insulator; Step S4-2: Establish a detection archive set, which includes a unique detection position identifier, detection time, compensated reflectivity, compensated reflectivity and defect judgment result, and perform defect judgment on the image data of the composite insulator at the detection point. The judgment process is as follows: When R(K)≥Q max When the composite insulator at the detection point is judged to be defective, the composite insulator at the detection point is marked as degraded and stored in the detection archive set; When R(K)<Q max When the composite insulator at the detection point is judged as having no defects, the composite insulator at the detection point is marked as normal and stored in the detection archive set.

7. A multi-spectrum composite insulator defect detection method according to claim 6, characterized in that: The specific steps of step S5 are as follows: Step S5-1, read the unique detection location identifier in the detection archive set, use the location data of the composite insulator at the detection point to perform a matching search, and obtain the historical data of the detection point; Step S5-2: extract the compensated reflectivity data of each test according to the historical data of the test point, and construct a defect trend graph according to the compensated reflectivity data of each test and the corresponding test time data.

8. The multi-spectrum composite insulator defect detection method according to claim 7, characterized in that: In step S6, the defect trend judgment threshold P is set max , the reflectivity of the composite insulator to be tested next time is predicted based on the compensated reflectivity data at each data point in the defect trend graph. The calculation formula used for the composite insulator reflectivity prediction is as follows: B next =a*R(K)+(1-a)*B previous ; In the formula, B next It is expressed as the predicted compensating reflectivity of the composite insulator for the next test; a is the smoothing coefficient, ranging from 0 to 1; B previous It is expressed as the compensated reflectivity of the composite insulator detected last time; The predicted compensation reflectivity is used to calculate the growth rate slope of the defect trend graph, and the growth rate slope is calculated using the following formula: Where S represents the slope of the defect trend graph growth rate calculated based on the predicted compensation reflectivity; T next Indicates the time for the next test; T(K) represents the current detection time of the composite insulator; T next -T(K) represents the time difference between two inspections of the composite insulator; When P max When >S, the predicted compensation reflectivity growth rate slope in the defect trend diagram is normal, indicating that there will be no defects in the next composite insulator inspection result and it is in a normal state; When P max When ≤S, the growth rate slope of the predicted compensation reflectivity in the defect trend diagram is abnormal, indicating that the next inspection result of the composite insulator will be defective and in a degraded state, and an early warning signal that the composite insulator at the inspection point is about to become abnormal is issued.

Images