Composite insulator infrared inspection calibration method based on PLS algorithm

Abstract

The invention discloses a composite insulator infrared inspection calibration method based on a PLS algorithm. The method comprises the following steps: acquiring infrared temperature measurement data at different distances and visual angles; preprocessing the infrared temperature measurement data, making an infrared temperature measurement data set, and dividing the infrared temperature measurement data set into a training set and a verification set according to a certain proportion; configuring PLS algorithm model parameters, carrying out PLS algorithm model training, checking a training effect, and obtaining a trained PLS algorithm model; and inputting infrared temperature measurement data to be predicted into the trained PLS algorithm model, and checking a correction result of the PLS algorithm model. Through the method, the precision of infrared inspection of the unmanned aerial vehicle can be improved, so that the fault level of the insulator can be judged more accurately.

Description

Technical Field

[0001] The invention relates to the technical field of infrared temperature measurement, and in particular to an infrared inspection and calibration method for composite insulators based on a PLS algorithm. Background Art

[0002] For a long time, composite insulators of transmission lines have been exposed to multiple factors such as electrical, mechanical and environmental factors, which can easily lead to aging of insulators and local discharge heating inside the sheath, which in turn causes aging and deterioration of the core rod and reduces the mechanical strength. Under extreme working conditions, it may even cause faults such as broken strings and dropped lines, with serious consequences. In order to improve the efficiency and accuracy of inspections, drones equipped with infrared imagers have been widely used for inspections in recent years. However, changes in distance and viewing angle during shooting will significantly affect the temperature measurement accuracy, making it difficult to accurately obtain the true temperature of the composite insulator, which in turn affects the judgment of the fault level.

[0003] In recent years, domestic and foreign scholars have conducted a lot of research on the accuracy correction of infrared temperature measurement. The main goal is to improve the accuracy and reliability of temperature measurement. The methods used include advanced infrared imaging technology, optimized image processing algorithms, and improved temperature compensation models. Although some progress has been made, there are still some shortcomings. The main problems include: the accuracy of temperature measurement is significantly affected by environmental factors, especially the changes in viewing angle and distance have a great impact on the accuracy of temperature measurement; the existing calibration methods are complex and have limited applicability, and it is difficult to maintain consistent measurement results in different environments; in addition, the real-time and automation levels need to be improved. These problems limit the application of infrared temperature measurement technology in composite insulator detection and need further research and optimization.

[0004] Reference [1]: Peng Zelong, Li Quan. Infrared human body temperature measurement accuracy compensation system based on binocular ranging and YOLOv5s [J]. Information and Computer (Theoretical Edition), 2022, 34(07): 80-83. A human body temperature measurement method based on binocular ranging and YOLOv5s is proposed. By real-time measuring the target depth information and performing three-dimensional position data compensation, the accuracy and precision of infrared temperature measurement are improved. However, the results of the infrared thermal imager are not corrected.

[0005] Reference [2]: Li Ying, Wu Ji, Wang Zengbin, et al. Research on the design and characterization of GIS infrared temperature sensing system [J]. High Voltage Electrical Appliances, 2021, 57(10): 18-24. An infrared temperature measurement system was designed and manufactured. The infrared temperature measurement results were compensated by temperature fitting, which improved the reliability and accuracy of temperature measurement at key locations of GIS equipment. However, it was not compared with other models and the accuracy could not be verified.

[0006] Reference [3]: Jin Yang. Research on dual-spectrum infrared temperature measurement compensation technology [D]. Xi'an University of Technology, 2024. The target boundary matching algorithm is combined with the dual-spectrum system to conduct ranging experiments, and a ranging model and a ranging error compensation model are established. The experimental data prove that the model is accurate and reliable. However, only two factors are corrected separately, and a comprehensive correction model is not established.

[0007] Reference [4]: ​​Cao Heyang. Key technologies and implementation of high-precision infrared temperature measurement [D]. Nanjing University of Science and Technology, 2021. An infrared temperature measurement algorithm based on adaptive correction without blackbody source is proposed, and the temperature measurement accuracy and anti-interference performance are improved by compensating for ambient temperature and test distance. However, only fitting correction is considered, and the method is relatively simple.

[0008] Partial least squares regression (PLS) is a commonly used machine learning algorithm that improves the stability and predictive ability of the model by converting the original independent variables into a set of new comprehensive variables and then using these principal components for regression analysis. The PLS algorithm has the advantage of simplifying multivariate problems into low-dimensional problems and can effectively handle multicollinearity and noisy data. Summary of the invention

[0009] Aiming at the problem that the temperature measurement accuracy of composite insulators is reduced due to the influence of distance and viewing angle changes during infrared temperature measurement, the present invention proposes a composite insulator infrared temperature measurement distance and viewing angle compensation calibration method based on partial least squares regression algorithm (PLS). This method can improve the accuracy of UAV infrared inspection, thereby achieving more accurate judgment of the fault level of insulators.

[0010] The technical solution adopted by the present invention is:

[0011] The infrared inspection and calibration method of composite insulator based on PLS algorithm includes the following steps:

[0012] Step 1: Obtain infrared temperature measurement data at different distances and viewing angles;

[0013] Step 2: Preprocess the infrared temperature measurement data to create an infrared temperature measurement data set, and divide the infrared temperature measurement data set into a training set and a validation set according to a certain ratio;

[0014] Step 3: Configure the PLS algorithm model parameters, perform PLS algorithm model training, test the training effect, and obtain the trained PLS algorithm model;

[0015] Step 4: Input the infrared temperature measurement data to be predicted into the trained PLS algorithm model to verify the correction result of the PLS algorithm model.

[0016] In step 1, a composite insulator infrared temperature measurement experimental platform was built. Figure 2As shown in the figure, a systematic experiment was conducted on 110kV composite insulators to analyze the relationship between the measured values ​​and the true values ​​of the infrared thermal imager at different measuring distances and viewing angles.

[0017] The composite insulator infrared temperature measurement experimental platform includes a booster, a suspension vehicle, a 110kV composite insulator, a fiber optic temperature meter, an infrared thermal imager, and an infrared temperature meter;

[0018] The boost device mainly includes a high-voltage power supply system and a high-voltage measurement system;

[0019] The high voltage power supply system includes a voltage regulator, a step-up transformer (150kV, 50kVA) and a protection resistor (10kΩ, 10kW); the AC220 power supply is connected to the voltage regulator, the voltage regulator is connected to the step-up transformer, and the step-up transformer is connected to the protection resistor;

[0020] The high voltage measurement system includes a voltmeter (voltage division ratio 1:1000);

[0021] The suspension vehicle is used to suspend 110kV composite insulators. The ground wire of the pressurized equipment is connected to the top suspension insulator of the suspension vehicle;

[0022] The infrared thermal imager model is FLUKET Ti100 3, which is used to measure the infrared temperature of 110kV composite insulators;

[0023] Fiber optic thermometer, used to measure the actual temperature of 110kV composite insulators;

[0024] The 110kV composite insulator is connected to the primary side of the step-up transformer through a protective resistor, the secondary side of the step-up transformer is connected to the voltage regulator, and a voltmeter is connected in parallel to the secondary side of the step-up transformer;

[0025] The model of the infrared thermometer is FLUKE 566 laser thermometer, which is also used to perform infrared temperature measurement on composite insulators and compare the temperature measurement results with those of the infrared thermal imager.

[0026] In step 3, a partial least squares regression (PLS) algorithm model is used for data regression prediction and error correction. The PLS algorithm model converts the original independent variables into a set of new comprehensive variables, and then uses these principal components for regression analysis, specifically including the following steps:

[0027] S3.1. Determine the principal components: Select an appropriate number of principal components and extract the main linear components related to the infrared temperature measurement values; determine the appropriate number of principal components through cross-validation or other standards, and select principal components that can effectively explain the independent variables and have a strong correlation with the infrared temperature measurement values. Use the variance contribution rate to evaluate the contribution of each principal component and screen out the main linear components with a high correlation with the infrared temperature measurement values:

[0028]

[0029] S3.2. Establish a PLS regression model: Use the principal component regression method to calculate the regression coefficients and establish a PLS regression model.

[0030] The specific steps are as follows:

[0031] Calculate the score matrix T and loading matrix P:

[0032] T = XW (2);

[0033] P=X T T (3);

[0034] Calculate the regression coefficient B:

[0035] B=(T T T) -1 T T Y (4);

[0036] Where X is the original data matrix, W is the weight matrix, and Y is the infrared temperature measurement value matrix; T is the transpose of the original matrix, T T is the transpose of the score matrix.

[0037] S3.3, Model verification: Input the verification data into the PLS regression model, calculate the corrected infrared temperature measurement value, and evaluate the correction effect of the model.

[0038] Corrected infrared temperature measurement value for:

[0039]

[0040] The present invention provides a composite insulator infrared inspection and calibration method based on PLS algorithm, and the technical effects are as follows:

[0041] 1) This paper proposes a distance and viewing angle compensation calibration method based on the PLS algorithm, which aims to eliminate the influence of distance and viewing angle in the temperature measurement process, improve the accuracy of infrared temperature measurement, and thus achieve accurate assessment of the fault status of insulators. This method not only improves the accuracy of drone infrared inspection, but also provides more reliable technical support for the operation and maintenance of power transmission lines, and has important application value and prospects.

[0042] 2) The present invention applies the PLS algorithm to infrared temperature measurement correction. Through the PLS algorithm, factors affecting infrared temperature measurement such as distance and angle are corrected and compensated, thereby achieving accurate temperature measurement of composite insulators and providing strong technical support for the inspection of composite insulators on transmission lines.

[0043] 3) By introducing the PLS algorithm model, the present invention can accurately compensate and calibrate the infrared temperature measurement data under different measurement distances and viewing angles. This method not only simplifies the complex calibration process, but also significantly improves the accuracy and reliability of temperature measurement. In the infrared inspection of drones, the method of the present invention can effectively reduce the interference of environmental factors on the temperature measurement results, realize accurate monitoring of the state of composite insulators and fault judgment, thereby improving the safety and operation efficiency of transmission lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0045] Figure 1 This is the flow chart for insulator temperature measurement correction.

[0046] Figure 2 This is the wiring schematic diagram of the infrared temperature measurement test;

[0047] Among them: 1-110kV composite insulator, 2-fiber optic thermometer, 3-infrared thermal imager, 4-voltage regulator, 5-step-up transformer, 6-protective resistor, 7-voltmeter, 8-laptop computer, 9-switch.

[0048] Figure 3 Figure 1 shows a dirty insulator.

[0049] Figure 4 Capture attitude images for thermal imaging cameras.

[0050] Figure 5 This is the temperature measurement result of the infrared thermal imager.

[0051] Figure 6 Flowchart for PLS correction.

[0052] Figure 7 This is a comparison chart of the correction effect.

[0053] Figure 8 This is a comparison chart of the correction results for infrared imagers.

[0054] Fig. 9 This is the infrared temperature measurement correction diagram. DETAILED DESCRIPTION

[0055] The infrared temperature measurement distance and viewing angle compensation calibration method of composite insulator based on PLS algorithm includes the following steps:

[0056] Step 1: Build an experimental platform and conduct infrared temperature measurement experiments on 110kV composite insulators to obtain infrared temperature measurement data at different distances and viewing angles.

[0057] Step 2: Organize and preprocess the experimental data, create an infrared temperature measurement data set, and divide the data set into a training set and a validation set in a ratio of 8:2.

[0058] Step 3: Configure the PLS algorithm model parameters, perform PLS algorithm model training, test the training effect, and obtain the trained PLS algorithm model.

[0059] Step 4: Input the data to be predicted into the trained PLS model, test the modified results of the model, and compare them with the results of other models to verify the effect of the modification.

[0060] In step 1, in order to study the influence of distance and viewing angle on infrared temperature measurement and obtain infrared temperature measurement data set, the present invention builds a composite insulator infrared temperature measurement experimental platform, such as Figure 2 As shown in Figure 1, a systematic experiment was conducted on 110kV composite insulators. The experiment aims to analyze the relationship between the measured value and the true value of the infrared thermal imager at different measurement distances and viewing angles. The specific measurement results are shown in Figure 1. Figure 5 As shown in the figure, from the temperature measurement results, it can be seen that as the temperature distance of the temperature measuring instrument increases, the temperature difference between the measured temperature and the accurate temperature basically shows a trend of increasing, and the larger the pitch angle, the larger the temperature measurement error, and the maximum error is 1.1℃.

[0061] The composite insulator infrared temperature measurement experimental platform includes a high-voltage power supply system and a high-voltage measurement system. The high-voltage power supply system consists of a voltage regulator, a transformer (150kV, 50kVA) and a protective resistor (10kΩ, 10kW); the high-voltage measurement system includes a resistor-capacitor voltage divider (voltage divider ratio 1:1000). In addition, the platform is also equipped with a pressurizing device, a composite insulator suspension vehicle, a fiber optic thermometer, an infrared thermometer and an infrared thermal imager. In order to eliminate the interference of factors such as temperature, humidity and wind speed, the experiment was carried out in an indoor environment with a wind speed of 0m / s, and the temperature measurement time was short to ensure that the temperature and humidity can be considered constant.

[0062] In the step 2, the temperature measurement data of various types of composite insulators at different distances and viewing angles using an infrared thermometer and an infrared thermal imager are obtained through experiments, and these data are sorted to form an infrared temperature measurement result data set.

[0063] In step 3, in order to improve the accuracy of infrared temperature measurement, the present invention uses partial least squares regression (PLS) algorithm for data regression prediction and error correction. The PLS algorithm converts the original independent variables into a set of new comprehensive variables, and then uses these principal components for regression analysis, thereby improving the stability and predictive ability of the model. The following are the specific steps of the PLS algorithm:

[0064] Data preprocessing: Clean and normalize the infrared temperature measurement data to ensure data consistency and model stability.

[0065] Divide the dataset: Divide the dataset into training set and validation set to ensure the generalization ability of the model.

[0066] Determine the principal components: Select an appropriate number of principal components and extract the main linear components related to the infrared temperature measurement values.

[0067] Establish a regression model: Use the principal component regression method to calculate the regression coefficient and establish a PLS regression model. The specific steps are as follows:

[0068] Calculate the score matrix T and loading matrix P:

[0069] T = XW (2);

[0070] P=X T T (3);

[0071] Calculate the regression coefficient B:

[0072] B=(T T T) -1 T T Y (4);

[0073] Where X is the original data matrix, W is the weight matrix, and Y is the infrared temperature measurement value matrix.

[0074] Model verification: Input the verification data into the PLS model, calculate the corrected infrared temperature measurement value, and evaluate the correction effect of the model. Corrected infrared temperature measurement value for:

[0075]

[0076] Through the above steps, the PLS algorithm can effectively correct the infrared temperature measurement value and improve the temperature measurement accuracy and reliability.

[0077] In step 4, after completing the training, the infrared temperature measurement test is performed again, and the temperature measurement data are corrected by different correction models, and the corrected accurate temperature values ​​are recorded and compared to obtain the final infrared temperature measurement correction results. In addition to the partial least squares regression (PLS) algorithm proposed in the present invention, other commonly used statistical regression models are also used, including linear, nonlinear and exponential function fitting. These models are used to correct the infrared temperature measurement data, and the correction accuracy and effect of the PLS infrared temperature measurement correction model proposed in the present invention and other models are evaluated by statistics and comparison of the corrected infrared temperature measurement data.

[0078] The present invention studies the infrared temperature measurement correction method of composite insulators running under power in the laboratory, and proposes an infrared temperature measurement correction method based on the PLS algorithm. First, the infrared temperature measurement data set of the composite insulator is obtained through experiments. Then the data set is trained to establish the PLS algorithm model. Finally, the correction effect of the model is verified through experiments, and compared with the correction effects of other commonly used correction models. The experimental results show that the average temperature difference of the infrared thermal imager before correction is 1.32°C, and the average error is 9.07%. After correction by the PLS algorithm model, the average temperature difference is reduced to 0.15°C, and the average error is reduced to 1.62%. The correction results show that the infrared temperature measurement correction method of composite insulators based on PLS proposed by the present invention has a significant correction effect and can effectively improve the infrared temperature measurement accuracy of composite insulators. This method can not only comprehensively consider influencing factors such as distance and viewing angle, but also expand to include other potential influencing factors, so as to adapt to different working environments. This method can be widely used in the field of transmission line UAV infrared inspection, greatly improving the accuracy of UAV infrared temperature measurement and the accuracy of insulator status determination, and providing strong technical support for regular UAV inspection of composite insulators and maintenance of transmission line safety.

[0079] The present invention aims at the problem of infrared temperature measurement accuracy of composite insulators and proposes a correction method based on PLS algorithm. The infrared temperature measurement data is systematically corrected by PLS (partial least squares regression) algorithm, so as to solve the temperature measurement error caused by the change of measurement distance and viewing angle in the traditional method. First, the present invention conducts infrared temperature measurement test on composite insulators by designing an experimental platform to collect data at different distances and viewing angles. Subsequently, the PLS algorithm is used to train these data and establish a correction model for specific temperature measurement scenarios. The experimental results show that the average temperature difference of the uncorrected infrared thermometer and thermal imager is 1.65℃ and 1.32℃, respectively, and the measurement errors are 17.06% and 9.07%, respectively. After applying PLS correction, the temperature difference is significantly reduced to 0.12℃ and 0.08℃, and the measurement error is reduced to 1.36% and 0.97%. This method can not only greatly improve the temperature measurement accuracy of composite insulators, but also has strong adaptability and expansibility, and can be adjusted and optimized according to different working environments and conditions. By adopting the PLS algorithm, the infrared temperature measurement results can be accurately corrected. This method can be applied to UAV infrared inspection of transmission lines, providing reliable technical support for temperature monitoring in practical applications.

[0080] Verification example:

[0081] The present invention uses an optical fiber temperature measuring device to measure the temperature of the composite insulator surface, and the temperature measurement accuracy reaches ±0.5°C. The temperature value of the optical fiber temperature measuring device can be regarded as an accurate value, which is used to evaluate the accuracy of other temperature measurement methods and as a data set input for the subsequent infrared temperature measurement correction method. The present invention also selects an infrared thermal imager for infrared temperature measurement. The instrument can generate an infrared image to display the temperature distribution of the measured object. The temperature measurement results of the infrared thermal imager will be corrected later. The specific parameters of the two temperature measuring instruments are shown in Table 1:

[0082] Table 1 Infrared temperature measuring instrument parameters

[0083]

[0084] Three composite insulators with grading rings, model 110kV FXBW4-110 / 70, were used in the test. One of them was used to simulate normal operating conditions, and the other two were used to simulate operation under dirty conditions. In accordance with GB / T 4585-2004 standard, the three normal insulators were cleaned with 50°C hot tap water with trisodium phosphate or other detergents, and then rinsed with tap water. Subsequently, two insulators with surface contamination were artificially constructed by the solid coating method. 40g of loess, 1000g of tap water and an appropriate amount of NaCl were mixed, and the dirty liquid was covered on the two cleaned and dried insulators by the pouring method, and then naturally dried in the environment for 24 hours. Photos of the contaminated insulators are shown in Figure 1. Figure 3 shown.

[0085] According to the above steps, the infrared temperature measurement results of several common composite insulators at different distances and viewing angles were obtained. The specific results are as follows: Figure 4 As shown. The temperature measurement results show that as the measuring distance of the infrared thermal imager increases, the temperature difference between the measured temperature and the accurate temperature tends to increase. In addition, changing the pitch angle of the temperature measurement will also affect the temperature measurement accuracy. The larger the pitch angle, the greater the temperature measurement error. The maximum temperature measurement error is 1.1°C. It can be seen from the infrared temperature measurement principle and test results that distance and viewing angle are important factors affecting the accuracy of infrared temperature measurement. In order to ensure the effectiveness of the infrared temperature measurement value, these influencing factors must be corrected through appropriate temperature compensation methods.

[0086] In order to verify the effectiveness of the proposed temperature measurement correction algorithm, the present invention will re-conduct the verification test and compare it with other commonly used correction algorithms. First, an infrared temperature rise test is carried out to obtain a data set for infrared temperature measurement correction. Then, the data set is trained by different correction algorithms such as statistical regression, machine learning and deep learning, and the correction model is derived after the training is completed. Then the insulator temperature rise test is carried out again, the infrared temperature measurement image of the test insulator is taken, and the infrared temperature measurement data as well as the distance and viewing angle are recorded. The recorded data is input into the trained correction model for infrared temperature measurement correction and compared with the accurate temperature measured in the verification test. In addition to using PLS (partial least squares regression) for correction, other most commonly used data correction methods are also used, including statistical regression methods with linear, nonlinear and exponential fitting corrections. These methods are used to perform temperature measurement correction on infrared thermometers and infrared thermal imagers, and compare the correction effects of different algorithms on the two infrared temperature measuring instruments. 。

[0087] The commonly used statistical regression models currently include linear, nonlinear, exponential, logarithmic and power function fitting. However, since the input viewing angle value has negative values, the logarithmic and power function fitting are not considered in the present invention. The present invention uses the linear, nonlinear and exponential function fitting methods to perform fitting analysis on the infrared temperature measurement data set. The fitting results are shown in Table 2. 2 represents the correlation coefficient of the fitting curve. r1 is the corrected temperature value of the infrared thermal imager, d is the distance, θ is the viewing angle, T m2 is the temperature value measured by the infrared thermal imager, T r2 is the corrected temperature value of the infrared thermometer, T m2 It is the temperature value measured by infrared thermometer.

[0088] Table 2 Fitting correction results

[0089]

[0090] The machine learning model training and verification were performed in Windows 11 system, with the hardware configuration of Intel(R) Core(TM) i5-13490F 4.80GHz CPU and NVIDIA GeForce RTX 3050GPU, and the programming software used was MatlabR2021a. The specific steps of the deep learning correction algorithm are as follows:

[0091] 1) Dataset preparation: The dataset obtained in the experiment was preprocessed, including data cleaning and missing value processing, and then the dataset was divided into a training set and a validation set in a ratio of 8:2.

[0092] 2) Set training parameters: Configure relevant parameters of deep learning and machine learning, including the number of hidden layer nodes, maximum number of iterations, and error threshold.

[0093] 3) Model training: Use different algorithms to train the data set and export the trained model.

[0094] 4) Training effect test: obtain training metrics, such as root mean square error and accuracy, to check the training effect. Figure 7 shown.

[0095] After the training is completed, the temperature measurement test is repeated, the infrared temperature measurement results are corrected through the correction models of different methods, and the accurate temperature values ​​are recorded. Fig. 9 As shown, the specific correction data is shown in Table 3:

[0096] Table 3 Comparison of correction results

[0097]

[0098] It can be seen from the correction results that before correction, the temperature difference of the infrared thermal imager was large. After correction by different algorithms, the temperature difference of the infrared temperature measurement was significantly reduced. There are differences in the correction effects of different algorithms on temperature measuring instruments. The statistical regression methods of linear, nonlinear and exponential fitting have certain correction effects on infrared temperature measurement, but the overall effect is poor, especially when processing complex data, the generalization ability of the statistical regression method is weak. In contrast, the machine learning model performs better. In the present invention, the PLS model shows more outstanding performance and stability. The PLS model not only has the best correction effect on individual points, but also outperforms other models in overall performance. Specifically, after correction by the PLS model, the average temperature difference of the infrared thermal imager is 0.15°C, and the average error is 1.62%. These results show that the performance of the PLS model in infrared temperature measurement correction is particularly outstanding, and it is suitable for high-precision infrared temperature measurement correction applications.

Claims

1. The infrared inspection and calibration method of composite insulator based on PLS algorithm is characterized by The following steps are involved: Step 1: Obtain infrared temperature measurement data at different distances and viewing angles; Step 2: Preprocess the infrared temperature measurement data to create an infrared temperature measurement data set, and divide the infrared temperature measurement data set into a training set and a validation set; Step 3: Configure the PLS algorithm model parameters, perform PLS algorithm model training, test the training effect, and obtain the trained PLS algorithm model; Step 4: Input the infrared temperature measurement data to be predicted into the trained PLS algorithm model to verify the correction result of the PLS algorithm model.

2. The composite insulator infrared inspection and calibration method based on the PLS algorithm according to claim 1 is characterized in that: In the step 1, a composite insulator infrared temperature measurement experimental platform is built, and a systematic experiment is carried out on the 110 kV composite insulator (1) to analyze the relationship between the measured value and the true value of the infrared thermal imager (3) at different measurement distances and viewing angles; The composite insulator infrared temperature measurement experimental platform includes a booster device, a suspension vehicle, a 110kV composite insulator (1), an optical fiber temperature meter (2), an infrared thermal imager (3), and an infrared temperature meter; The voltage boosting device includes a high voltage power supply system and a high voltage measurement system; The high-voltage power supply system comprises a voltage regulator (4), a step-up transformer (5), and a protective resistor (6); the AC220 power supply is connected to the voltage regulator (4), the voltage regulator (4) is connected to the step-up transformer (5), and the step-up transformer (5) is connected to the protective resistor (6); The high voltage measurement system comprises a voltmeter (7); A suspension vehicle is used to suspend a 110 kV composite insulator (1), and a ground wire of a pressurized device is connected to the top suspension insulator of the suspension vehicle; An infrared thermal imager (3) for measuring the infrared temperature of a 110 kV composite insulator (1); An optical fiber thermometer (2) is used to measure the actual temperature of the 110 kV composite insulator (1); The 110 kV composite insulator (1) is connected to the primary side of the step-up transformer (5) through a protective resistor (6), the secondary side of the step-up transformer (5) is connected to a voltage regulator (4), and a voltmeter (7) is connected in parallel to the secondary side of the step-up transformer (5); The infrared thermometer is used to perform infrared temperature measurement on a 110 kV composite insulator (1) and compare the temperature measurement result with that of an infrared thermal imager (3).

3. The infrared inspection and calibration method for composite insulators based on the PLS algorithm according to claim 1 is characterized in that: In step 3, a PLS algorithm model is used for data regression prediction and error correction; the PLS algorithm model converts the original independent variables into a set of new comprehensive variables, and then uses these principal components for regression analysis, specifically including the following steps: S3.

1. Use the variance contribution rate to evaluate the contribution of each principal component and screen out the main linear components with high correlation with infrared temperature measurement values: S3.

2. Use the principal component regression method to calculate the regression coefficient and establish a PLS regression model. The specific steps are as follows: Calculate the score matrix T and loading matrix P: T = XW (2); P=X T T (3); Calculate the regression coefficient B: B=(T T T) -1 T T Y (4); Where X is the original data matrix, W is the weight matrix, and Y is the infrared temperature measurement value matrix; T is the transpose of the original matrix, T T is the transpose of the score matrix; S3.3, input the verification data into the PLS regression model, calculate the corrected infrared temperature measurement value, and evaluate the correction effect of the model; Corrected infrared temperature measurement value for:

Images