Composite insulator infrared temperature measurement distance and visual angle compensation calibration method based on RBF algorithm

Abstract

The invention discloses a composite insulator infrared temperature measurement distance and visual angle compensation calibration method based on an RBF algorithm. The method comprises the following steps: acquiring infrared temperature measurement data of a composite insulator under 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 RBF neural network parameters, performing deep learning training, checking a training effect, and obtaining a trained RBF neural network model; and inputting infrared temperature measurement data to be predicted into the trained RBF neural network model, and checking a correction result of the RBF neural network model. Through the method, the precision of infrared inspection of the unmanned aerial vehicle can be improved, and the insulator fault grade judgment is more accurate.

Description

Technical Field

[0001] The invention relates to the technical field of infrared temperature measurement, and in particular to a composite insulator infrared temperature measurement distance and viewing angle compensation calibration method based on an RBF algorithm. Background Art

[0002] The composite insulators of power transmission lines are subjected to the combined effects of electrical, mechanical and environmental factors for a long time. It is easy for acid corrosion and electrocorrosion to cause partial discharge heating inside the sheath, which leads to aging and deterioration of the core rod, and brittleness, thereby reducing the mechanical strength of the insulator. In severe working conditions, string breakage and line drop accidents may also occur, resulting in inestimable consequences. In recent years, infrared temperature measurement technology has been increasingly used in the inspection of power transmission lines. During the operation and maintenance of line insulators, drones equipped with infrared imagers are usually used for inspection. The distance of 3-10m is maintained during shooting, and the connection between the infrared lens and the insulator is at an angle of 30° to the conductor. According to the principle of infrared temperature measurement, the distance and shooting angle of view will affect the temperature measurement accuracy, so it is difficult to obtain the true temperature of the composite insulator very accurately. A slight measurement deviation of the temperature may cause misjudgment of the state of the composite insulator. Therefore, it is of great significance to study how to eliminate the influence of distance and angle of view on the infrared temperature measurement results and correct the temperature measurement accuracy for accurately evaluating the state of the insulator.

[0003] In recent years, some scholars at home and abroad have conducted research on the correction of infrared temperature measurement accuracy, mainly focusing on improving temperature measurement accuracy and reliability. The methods include using advanced infrared imaging technology, optimizing image processing algorithms, and improving temperature compensation models. Although some progress has been made, there are still some shortcomings. The main problems include: the temperature measurement accuracy is greatly affected by environmental factors, especially the changes in viewing angle and distance have a significant impact on the accuracy of temperature measurement; the existing calibration methods are highly complex and have limited applicability, and it is difficult to maintain consistent measurement results in different working environments; in addition, the real-time and automation levels need to be improved. These problems restrict the widespread application of infrared temperature measurement technology in composite insulator detection, and further research and optimization are urgently needed.

[0004] Literature 1: He Zhenpeng, Wei Xing, Li Baichun, et al. Compensation research on the influence of distance and fog on infrared temperature measurement accuracy [J]. Infrared Technology, 2024, 46(06): 681-690. In the technical scheme recorded, the influence of distance, relative humidity and fog on the temperature measurement accuracy of infrared thermal imagers was studied, and an error compensation model was proposed to improve the temperature measurement accuracy of infrared thermal imagers in foggy weather.

[0005] Literature 2: Mi Songtao, Zhang Yucun, Fu Xianbin, et al. Infrared temperature measurement compensation method based on emissivity correction model [J]. Journal of Metrology, 2024, 45(02): 164-169. In the technical scheme recorded, the influence of surface roughness on the accuracy of infrared temperature measurement was studied, and an infrared temperature measurement compensation method based on the emissivity correction model was proposed. Experimental verification showed that this method can significantly improve the accuracy of infrared temperature measurement.

[0006] Literature 3: Jin Yang. Research on dual-spectrum infrared temperature measurement compensation technology [D]. Xi'an University of Technology, 2024. In the recorded technical solution, 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 proves that the model is accurate and reliable.

[0007] Literature 4: Zhang Pipei, Shi Wei, Guo Chenrui, et al. Research on the surface temperature difference correction method of UHV voltage heating equipment under the influence of environmental factors [J]. Infrared Technology, 2023, 45(03): 328-334. In the recorded technical scheme, the influence of environmental factors such as light intensity, ambient temperature and wind speed on the surface temperature difference of UHV voltage heating equipment was studied, and a temperature difference correction method under non-ideal environmental conditions was proposed.

[0008] Infrared temperature measurement correction algorithm based on data regression prediction of radial basis function neural network (RBF). RBF is a feedforward neural network with good nonlinear fitting ability and fast convergence speed, which is suitable for processing complex nonlinear data. The algorithm constructs an RBF network, maps the input data to a high-dimensional space, and uses radial basis functions such as Gaussian functions to fit the data, thereby achieving accurate correction of infrared temperature measurement. Summary of the invention

[0009] In order to solve the problems existing in the prior art that the temperature measurement accuracy is significantly affected by the distance and viewing angle, the calibration method is complicated and has limited applicability, the present invention provides a composite insulator infrared temperature measurement distance and viewing angle compensation calibration method based on RBF algorithm, by which the accuracy of UAV infrared inspection can be improved, and the judgment of insulator fault level can be made more accurate.

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

[0011] The infrared temperature measurement distance and viewing angle compensation calibration method of composite insulator based on RBF algorithm includes the following steps:

[0012] Step 1: Obtain infrared temperature measurement data of composite insulators 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 RBF neural network parameters, perform deep learning training, test the training effect, and obtain the trained RBF neural network model;

[0015] Step 4: Input the infrared temperature measurement data to be predicted into the trained RBF neural network model and verify the correction result of the RBF neural network model.

[0016] In step 1, a composite insulator temperature rise test platform is built. Figure 2 As 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 measurement distances and viewing angles. The composite insulator temperature rise test platform includes a high-voltage power supply system and a high-voltage measurement system;

[0017] The high voltage power supply system includes a voltage regulator, a transformer and a protective resistor; the AC220 power supply is connected to the voltage regulator, the voltage regulator is connected to the transformer, and the transformer is connected to the protective resistor;

[0018] In addition, the composite insulator temperature rise test 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;

[0019] In the composite insulator temperature rise test platform, the RC voltage divider is connected to the transformer to divide the high voltage signal into a voltage range suitable for measurement to achieve high voltage measurement;

[0020] The pressure device is connected to the transformer, and the high voltage output in the test is controlled by adjusting the voltage to ensure that the test conditions are met;

[0021] The composite insulator suspension vehicle is connected to the composite insulator and is responsible for suspending and fixing the composite insulator, simulating the actual working environment and ensuring its correct position;

[0022] The optical fiber thermometer is connected to the temperature sensor on the composite insulator through the CH1, CH2, CH3, and CH4 channels to record the temperature rise data in real time and provide accurate temperature detection;

[0023] Infrared thermometers and infrared thermal imagers are mainly used for non-contact temperature monitoring and visualization of temperature distribution, so as to analyze the relationship between the measured value and the true value, and compare the results with the fiber optic thermometer to verify the accuracy of infrared temperature measurement.

[0024] In step 3, a radial basis function neural network (RBF) is used to perform data regression prediction and error correction.

[0025] First, determine the structure of the RBF neural network, including the input layer, hidden layer and output layer, where the number of neurons in the input layer is equal to the number of input variables, the number of neurons in the hidden layer is determined according to the complexity of the problem, and the number of neurons in the output layer is equal to the number of output variables;

[0026] Then, the center, width and weight of the RBF neural network are initialized, and the training data is clustered by the k-means clustering algorithm to determine the center of the hidden layer radial basis function. The selection of the center and the clustering result are achieved by minimizing the objective function:

[0027]

[0028] In the formula, k is the number of clusters, C i is the i-th cluster, x j is the jth data point, μ i is the center point of the ith cluster.

[0029] And calculate the width of each radial basis function. The width is calculated as:

[0030]

[0031] In the formula, σ i is the width of the ith radial basis function, d max is the maximum distance between cluster centers, and M is the number of neurons in the hidden layer.

[0032] Calculate the hidden layer output, that is, the mapping result of the input data through the radial basis function. The output of the radial basis function can be expressed by a Gaussian function:

[0033]

[0034] In the formula, h i (x) is the output of the ith radial basis function, x is the input data, c i is the center of the ith radial basis function.

[0035] The least squares method is used to determine the weight of the output layer, and the output of the output layer is a linear combination of the outputs of the hidden layer:

[0036]

[0037] In the formula, y is the output result, w i is the weight of the i-th hidden layer neuron; M is the number of neurons in the hidden layer.

[0038] By inputting test data, the network parameters are adjusted until the output error meets the requirements.

[0039] Finally, input variables such as distance x1, viewing angle x2, and infrared temperature measurement value x3, calculate the neuron outputs of the hidden layer and the output layer through the RBF neural network, and obtain the corrected infrared temperature measurement value y'. Then, the weights and biases are continuously adjusted until the preset accuracy requirements are met, and the optimized RBF neural network model is output.

[0040] The present invention provides a composite insulator infrared temperature measurement distance and viewing angle compensation calibration method based on RBF algorithm, and the technical effects are as follows:

[0041] 1) The present invention solves the problems existing in the prior art that the temperature measurement accuracy is significantly affected by the distance and viewing angle, the calibration method is complicated and has limited applicability. By constructing a radial basis function neural network (RBF) and utilizing its nonlinear fitting ability and fast convergence speed, the present invention realizes the accurate correction of infrared temperature measurement data.

[0042] 2) This method can not only effectively eliminate the influence of distance and viewing angle on the infrared temperature measurement results, but also maintain consistent measurement results in different working environments, significantly improve the real-time and automation level of UAV infrared inspection, ensure accurate assessment of the status of composite insulators, and thus improve the safety and operational reliability of transmission lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0044] Figure 1 This is the flow chart for insulator temperature measurement correction.

[0045] Figure 2 This is the wiring schematic diagram of the infrared temperature measurement test;

[0046] Among them: 1- voltage regulator, 2- transformer, 3- protection resistor, 4- fiber optic temperature meter, 5- composite insulator, 6- computer.

[0047] Figure 3 Figure 1 shows a dirty insulator.

[0048] Figure 4 Take a posture picture for the infrared temperature measuring instrument.

[0049] Figure 5(a) shows the temperature measurement result of the infrared thermometer;

[0050] Figure 5(b) shows the temperature measurement result of the infrared thermal imager.

[0051] Figure 6 This is the RBF network structure diagram.

[0052] Figure 7 This is the RBF correction flowchart.

[0053] Figure 8(a) is a graph of the model iteration error;

[0054] Figure 8(b) is a graph showing the error during the training process;

[0055] Figure 8(c) is a comparison chart of the correction effects.

[0056] Fig. 9 This is a comparison chart of the correction results of the infrared thermometer.

[0057] Fig.10 This is a comparison chart of the positive results of the infrared imager.

[0058] Fig.11 This is the infrared temperature measurement correction diagram. DETAILED DESCRIPTION

[0059] The infrared temperature measurement distance and viewing angle compensation calibration method of composite insulator based on RBF algorithm includes the following steps:

[0060] Step 1: Build a composite insulator temperature rise test platform and obtain infrared temperature measurement data through composite insulator infrared temperature measurement experiment;

[0061] Step 2: Organize the external temperature measurement experimental data, create an infrared temperature measurement data set, and divide the data set into a training set and a validation set at a ratio of 8:2;

[0062] Step 3: Configure RBF neural network parameters, perform deep learning training, test the training effect, and obtain the trained RBF neural network model;

[0063] Step 4: Input the data to be predicted, test the modified results of the RBF neural network model, and compare it with other models.

[0064] In step 1, in order to explore the influence of distance and viewing angle on infrared temperature measurement and obtain an infrared temperature measurement data set, the present invention builds a composite insulator temperature rise test platform, tests 110kV composite insulators, and studies the relationship between the measured value and the true value of the infrared thermal imager and infrared thermometer at different measurement distances and viewing angles. As shown in Figure 5(a) and Figure 5(b), as the temperature measurement distance of the temperature measuring instrument increases, the temperature difference between the measured temperature and the accurate temperature increases, the larger the pitch angle, the larger the temperature measurement error, and the maximum error of the infrared thermometer is 1.6°C. The infrared thermal imager is 1.1°C.

[0065] The composite insulator temperature rise test platform used in this test mainly includes a high-voltage power supply system and a high-voltage measurement system. The high-voltage power supply system includes 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). The composite insulator temperature rise test platform mainly includes a pressurizing device, a composite insulator suspension vehicle, a 110kV composite insulator, an optical fiber thermometer, an infrared thermometer, and an infrared thermal imager. In order to eliminate the interference of other factors such as temperature, humidity, and wind speed on the test, this test was carried out in the laboratory, so the wind speed was 0m / s, and the time for infrared temperature measurement was short, so the temperature and humidity can be regarded as constant.

[0066] In the step 2, infrared temperature measurement results of several commonly used types of composite insulators at different distances and viewing angles using an infrared thermometer and an infrared thermal imager are obtained.

[0067] In step 3, in order to improve the accuracy of infrared temperature measurement, the present invention uses radial basis function neural network (RBF) for data regression prediction and error correction. RBF neural network can effectively process complex temperature measurement data and improve temperature measurement accuracy through its nonlinear mapping and fast convergence ability.

[0068] First, determine the structure of the RBF neural network, including the input layer, hidden layer and output layer, where the number of neurons in the input layer is equal to the number of input variables, the number of neurons in the hidden layer is determined according to the complexity of the problem, and the number of neurons in the output layer is equal to the number of output variables.

[0069] Then, the center, width and weight of the RBF neural network are initialized, and the training data is clustered by the k-means clustering algorithm to determine the center of the hidden layer radial basis function. The selection of the center and the clustering result are achieved by minimizing the objective function:

[0070]

[0071] In the formula, k is the number of clusters, C i is the i-th cluster, x j is the jth data point, μ i is the center point of the ith cluster.

[0072] Calculate the width of each radial basis function. The width is calculated as:

[0073]

[0074] In the formula, σ i is the width of the ith radial basis function, d max is the maximum distance between cluster centers, and M is the number of neurons in the hidden layer.

[0075] Calculate the hidden layer output, that is, the mapping result of the input data through the radial basis function. The output of the radial basis function can be expressed by a Gaussian function:

[0076]

[0077] In the formula, h i (x) is the output of the ith radial basis function, x is the input data, c i is the center of the ith radial basis function.

[0078] The least squares method is used to determine the weight of the output layer, and the output of the output layer is a linear combination of the outputs of the hidden layer:

[0079]

[0080] In the formula, y is the output result, w i is the weight of the i-th hidden layer neuron. By inputting test data, adjust the network parameters until the output error meets the requirements.

[0081] Finally, input variables such as distance x1, viewing angle x2, and infrared temperature measurement value x3, calculate the neuron outputs of the hidden layer and the output layer through the RBF neural network, and obtain the corrected infrared temperature measurement value y'. Then, the weights and biases are continuously adjusted until the preset accuracy requirements are met, and the optimized RBF neural network model is output.

[0082] Through these steps, the RBF neural network can efficiently and accurately correct the infrared temperature measurement value and improve the temperature measurement accuracy and reliability.

[0083] In step 4, after the training is completed, the temperature measurement test is carried out again, the infrared temperature measurement results are corrected by correction models of different methods, and the accurate temperature value is recorded to obtain the correction results of infrared temperature measurement. In addition to using the radial basis function neural network (RBF) proposed in the present invention, the currently commonly used statistical regression models are used, including linear, nonlinear and exponential function fitting, to correct the infrared temperature measurement data respectively. The infrared temperature measurement data after statistical correction are compared and analyzed to compare the correction accuracy of the RBF infrared temperature measurement correction model proposed in the present invention and other models.

[0084] 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 RBF algorithm. First, infrared temperature measurement is performed on the composite insulator in the test to obtain an infrared temperature measurement data set. Then the data set is trained to establish a correction model, and finally the correction effect of the algorithm proposed by the present invention is verified by experiments. The experiment shows that before correction, the average temperature difference between the infrared thermometer and the thermal imager is 1.65°C and 1.32°C, respectively, and the average errors are 17.06% and 9.07%. After correction by the RBF algorithm, the average temperature difference is reduced to 0.19°C and 0.27°C, respectively, and the average errors are reduced to 2.07% and 3.69%, respectively. The correction results show that the infrared temperature measurement correction method for composite insulators based on RBF proposed by the present invention has a significant correction effect and effectively improves the infrared temperature measurement accuracy of composite insulators. This method is a comprehensive correction method, which can add other influencing factors for correction and is suitable for different working environments. This method can be applied to the field of UAV infrared inspection of power transmission lines, 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 power transmission line safety.

[0085] Verification example:

[0086] The present invention uses a WT-YH-400 optical fiber temperature measuring device to measure the temperature of the composite insulator surface. The temperature measurement accuracy is ±0.5°C. The temperature value obtained by the optical fiber temperature measuring device can be used as an accurate value to judge the measurement accuracy of other temperature measurement methods, and as a data set input for subsequent infrared temperature measurement correction methods. The present invention uses two types of infrared temperature measuring instruments, namely infrared laser thermometers and infrared thermal imagers. Infrared laser thermometers mainly use lasers to perform point temperature measurement and have no image display function, while infrared thermal imagers can generate infrared images to display the specific temperature distribution of the object being measured. The results of these two types of infrared temperature measurements will be corrected later. The specific parameters of the three temperature measuring instruments are detailed in Table 1:

[0087] Table 1 Infrared temperature measuring instrument parameters

[0088]

[0089] The samples used in the test were three 110kV FXBW4-110 / 70 composite insulators with equalizing rings, one of which was used to simulate the operating conditions of normal insulators, and the other two were used to simulate the operating conditions of contaminated insulators. In accordance with the requirements of GB / T4585-2004, the surfaces of the three normal insulators were cleaned with tap water (heated to 50°C) containing trisodium phosphate or other cleaning agents, and then rinsed with tap water. Then, two composite insulators with surface contamination were artificially constructed using the solid coating method. 40g of loess, 1000g of tap water and an appropriate amount of NaCl were mixed, and the contaminated liquid was covered on the two cleaned and dried insulators by pouring, and then placed in the environment to dry naturally for 24 hours. Photos of contaminated insulators are shown in the figure. Figure 3 shown.

[0090] According to the above test steps, the infrared temperature measurement results of several commonly used types of composite insulators were obtained by infrared thermometer and infrared thermal imager at different distances and viewing angles. The specific results are shown in Figure 5(a) and Figure 5(b). It can be seen from the temperature measurement results that as the temperature measurement distance of the infrared thermometer and infrared thermal imager increases, the temperature difference between the measured temperature and the accurate temperature basically shows a trend of increasing, and the temperature measurement accuracy will also be affected by changing the pitch angle of the temperature measurement. The larger the pitch angle, the greater the temperature measurement error. The maximum temperature difference of the infrared thermometer is 1.6℃. The infrared thermal imager is 1.1℃. It can be seen from the infrared temperature measurement principle and experiments 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, a certain temperature compensation method must be used to correct the influence of these influencing factors on infrared temperature measurement.

[0091] In order to verify the effectiveness of the proposed temperature measurement correction algorithm, the present invention will re-conduct the verification test and use other commonly used correction algorithms to compare with the algorithm of the present invention. First, an infrared temperature rise test is carried out to obtain a data set for infrared temperature measurement correction, and then the data set is trained by statistical regression, machine learning and deep learning different correction algorithms. After the training is completed, the correction model is derived. At the same time, 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 the RBF deep learning network correction, the most commonly used data correction methods are also used, including the statistical regression correction method of linear, nonlinear and exponential fitting correction, and the machine learning correction method of ELM and PLS. The infrared thermometer and infrared thermal imager are corrected for temperature measurement, and the correction effects of different algorithms on the two infrared temperature measuring instruments are compared.

[0092] The commonly used statistical regression models currently include linear, nonlinear, exponential and logarithmic function fitting. However, due to the negative value of the input viewing angle value, the logarithmic and power function fitting is 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.

[0093] Table 2 Fitting correction results

[0094]

[0095]

[0096] In Table 2, R 2 represents the correlation coefficient of the fitting curve. The closer it is to 1, the stronger the model explanatory ability and the better the fitting effect. 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.

[0097] The model training and verification of machine learning are carried out in Windows 11 system, the CPU is Intel (R) Core (TM) i5-13490F 4.80GHz, the GPU is NVIDIA GeForce RTX 3050, and the programming software is Matlab R2021a. The specific correction steps of the deep learning correction algorithm are as follows: 1) First, make a data set. The data set obtained in the experiment is preprocessed, including data cleaning, missing value processing, etc., and then the data set is divided, and the infrared temperature measurement data obtained in the experiment is divided into a training set and a verification set according to 8:2; 2) Set training parameters. Set the number of hidden layer nodes, maximum number of iterations, and error threshold of deep learning and machine learning; 3) Model training. The data set is trained using different algorithms, and the trained model is exported at the same time; 4) Training effect test. Obtain training measurement indicators, including root mean square error and accuracy, to check the training effect. The error change curve of the RBF algorithm training process used in the present invention and the comparison of the correction effect of the verification set are shown in Figures 8(a), 8(b), and 8(c).

[0098] After the training is completed, the temperature measurement test is carried out again, the infrared temperature measurement results are corrected through the correction models of different methods, and the accurate temperature values ​​are recorded to obtain the corrected results of infrared temperature measurement as shown in the figure. Fig. 9 shown.

[0099] Fig.10 This is a comparison chart of the positive results of the infrared imager. Fig.11This is the infrared temperature measurement correction diagram. Fig.10 and Fig.11 As shown in the figure, after the infrared temperature measurement data is corrected by the RBF algorithm, the temperature difference and error of the infrared thermometer and infrared thermal imager are significantly reduced. Before correction, the temperature measurement errors of the two instruments are large. After correction, the RBF algorithm shows the best correction effect compared with the linear, nonlinear and exponential fitting methods. The temperature measurement error before correction is 7.20%, and the error after correction is reduced to 1.61%, which significantly improves the accuracy of temperature measurement.

[0100] The specific correction data are shown in Table 3.

[0101] Table 3 Comparison of correction results

[0102]

[0103] It can be seen from the correction results that before correction, the temperature difference between the infrared thermometer and the infrared thermal imager was large. After correction by different algorithms, the temperature difference of infrared temperature measurement was significantly improved. Therefore, the correction effects of different algorithms on different temperature measuring instruments are also different. Linear, nonlinear and exponential statistical regression correction methods have a certain correction effect on infrared temperature measurement, but the overall correction effect is poor, especially when processing complex data, the generalization ability of statistical regression correction is weak. In contrast, the machine learning model performs better. The RBF model is the most outstanding in the overall correction effect. It not only performs well at individual points, but also has strong overall stability, which is better than other correction methods. Specifically, after the infrared thermometer is corrected by the RBF model, the average temperature difference is only 0.19℃, and the average error is 2.07%, while the average temperature difference of the infrared thermal imager is 0.27℃, and the average error is 3.69%. These results show that the RBF model performs particularly well in infrared temperature measurement correction and is suitable for high-precision infrared temperature measurement correction applications. The specific correction effects are shown in Table 3. The RBF model performs best among all correction methods.

Claims

1. The infrared temperature measurement distance and viewing angle compensation calibration method of composite insulator based on RBF algorithm is characterized by The following steps are involved: Step 1: Obtain infrared temperature measurement data of composite insulators 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 RBF neural network parameters, perform deep learning training, test the training effect, and obtain the trained RBF neural network model; Step 4: Input the infrared temperature measurement data to be predicted into the trained RBF neural network model and verify the correction result of the RBF neural network model.

2. The method for calibrating the distance and viewing angle compensation of infrared temperature measurement of composite insulators based on RBF algorithm according to claim 1 is characterized in that: In the step 1, a composite insulator temperature rise test platform was built, and a systematic experiment was conducted on 110kV composite insulators 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 composite insulator temperature rise test platform includes a high-voltage power supply system and a high-voltage measurement system; The high voltage power supply system comprises a voltage regulator (1), a transformer (2) and a protective resistor (3); the AC220 power supply is connected to the voltage regulator (1), the voltage regulator (1) is connected to the transformer (2), and the transformer (2) is connected to the protective resistor (3); In addition, the composite insulator temperature rise test platform is also equipped with a pressurizing device, a composite insulator suspension vehicle, an optical fiber thermometer (4), an infrared thermometer and an infrared thermal imager.

3. The method for calibrating the distance and viewing angle compensation of infrared temperature measurement of composite insulators based on RBF algorithm according to claim 2 is characterized in that: In the composite insulator temperature rise test platform, a resistance-capacitance voltage divider is connected to a transformer (2) and is used to divide a high voltage signal into a voltage range suitable for measurement, so as to achieve high voltage measurement; The pressure device is connected to the transformer (2) to control the high voltage output in the test by adjusting the voltage to ensure that the test conditions are met; the composite insulator suspension vehicle is connected to the composite insulator (5) to suspend and fix the composite insulator (5) to simulate the actual working environment and ensure its correct position; The optical fiber thermometer is connected to the temperature sensor on the composite insulator (5) through the CH1, CH2, CH3 and CH4 channels to record the temperature rise data in real time and provide accurate temperature detection; Infrared thermometers and infrared thermal imagers are used for non-contact temperature monitoring and visualization of temperature distribution, so as to analyze the relationship between the measured value and the true value, and compare the results with the optical fiber thermometer (4) to verify the accuracy of infrared temperature measurement.

4. The method for calibrating the distance and viewing angle compensation of infrared temperature measurement of composite insulators based on RBF algorithm according to claim 1 is characterized in that: In step 3, a radial basis function neural network (RBF) is used to perform data regression prediction and error correction. First, determine the structure of the RBF neural network, including the input layer, hidden layer and output layer, where the number of neurons in the input layer is equal to the number of input variables, the number of neurons in the hidden layer is determined according to the complexity of the problem, and the number of neurons in the output layer is equal to the number of output variables; Then, the center, width and weight of the RBF neural network are initialized, and the training data is clustered by the k-means clustering algorithm to determine the center of the hidden layer radial basis function. The selection of the center and the clustering result are achieved by minimizing the objective function: In the formula, k is the number of clusters, C i is the i-th cluster, x j is the jth data point, μ i is the center point of the i-th cluster; And calculate the width of each radial basis function. The width is calculated as: In the formula, σ i is the width of the ith radial basis function, d max is the maximum distance between cluster centers, M is the number of neurons in the hidden layer; Calculate the hidden layer output, that is, the mapping result of the input data through the radial basis function. The output of the radial basis function can be expressed by a Gaussian function: In the formula, h i (x) is the output of the ith radial basis function, x is the input data, c i is the center of the ith radial basis function; The least squares method is used to determine the weight of the output layer, and the output of the output layer is a linear combination of the outputs of the hidden layer: In the formula, y is the output result, w i is the weight of the i-th hidden layer neuron; M is the number of neurons in the hidden layer; By inputting test data, adjust the network parameters until the output error meets the requirements; Finally, input the distance x1, viewing angle x2, and infrared temperature measurement value x3 variables, calculate the neuron outputs of the hidden layer and the output layer through the RBF neural network, obtain the corrected infrared temperature measurement value y', and continuously adjust the weights and biases until the preset accuracy requirements are met, and output the optimized RBF neural network model.

Images