Insulator pollution on-line monitoring system and monitoring method based on light diffraction

Abstract

The invention discloses an insulator contamination online monitoring system based on light diffraction, which comprises a main control unit, the main control unit is electrically connected with an optical measurement module, a wind speed and direction sensor, a rainfall sensor, a temperature and humidity sensor and a communication module through cables, and the communication module is further electrically connected with a monitoring center through a cable. The wind speed and direction sensor is fixedly installed on a tower, the rainfall sensor is fixedly installed on a cross arm at the top of a telegraph pole, the temperature and humidity sensor is fixedly installed below a tower cross arm, and the optical measurement module is arranged on a tower of a power transmission line. The problems that in the prior art, time and labor are wasted in the manual detection process, the manual detection process is greatly influenced by the technology and experience of detection personnel, and a non-manual detection method does not fully consider the influence of contamination components and content on ESDD and NSDD of the insulator, so that the detection result cannot accurately represent the contamination degree of the insulator are solved. The invention further discloses an insulator contamination online monitoring method based on light diffraction.

Description

Technical Field

[0001] The present invention belongs to the technical field of insulator contamination online monitoring, and relates to an insulator contamination online monitoring system based on light diffraction. The present invention also relates to an insulator contamination online monitoring method based on light diffraction. Background Art

[0002] Insulators play a crucial role in supporting and insulating transmission lines and are among the most widely used pieces of electrical equipment. Suspended particles in the atmosphere settle on the insulator surface, which can easily cause surface flashover under conditions such as fog, haze, and rain. Therefore, research is needed on the mechanisms of pollution flashover and anti-pollution measures.

[0003] Currently, the degree of insulator contamination is primarily characterized by equivalent salt density (ESDD) and ash density (NSDD). Traditional ESDD and NSDD detection methods are primarily manual, requiring the insulator to be disassembled, wiped, surface contamination dissolved, the solution conductivity measured, and the insoluble matter dried and weighed. This entire process is time-consuming and labor-intensive, and significantly influenced by the skill and experience of the inspector. Furthermore, some non-manual detection methods fail to fully consider the impact of contaminant composition and content on the insulator's ESDD and NSDD, resulting in test results that cannot accurately represent the insulator's contamination level. Therefore, it is necessary to provide an efficient and reliable method for detecting insulator contamination to effectively guide insulator anti-contamination efforts. Summary of the Invention

[0004] The purpose of the present invention is to provide an online monitoring system for insulator contamination based on light diffraction, which solves the problems in the prior art that the manual detection process is time-consuming and labor-intensive and is greatly affected by the skills and experience of the detection personnel, and that the non-manual detection method does not fully consider the impact of the contamination composition and content on the ESDD and NSDD of the insulator, thereby making the detection results unable to accurately characterize the contamination degree of the insulator.

[0005] Another object of the present invention is to provide an on-line monitoring method for insulator contamination based on light diffraction.

[0006] The technical solution adopted by the present invention is an online monitoring system for insulator contamination based on light diffraction, which includes a main control unit, which is electrically connected to an optical measurement module, a wind speed and direction sensor, a rain sensor, a temperature and humidity sensor, and a communication module through cables. The communication module is also electrically connected to a monitoring center through cables. The wind speed and direction sensor is fixedly mounted on a tower, the rain sensor is fixedly mounted on a crossarm at the top of a utility pole, the temperature and humidity sensor is fixedly mounted below the crossarm of the tower, and the optical measurement module is arranged on the tower of the transmission line.

[0007] Furthermore, the optical measurement module includes a high-voltage generator and an X-ray tube connected in sequence, as well as a disk collection unit fixed on the tower of the transmission line. The cross-sectional area of the disk collection unit is equal to the surface area of the insulator, and the material is the same as the insulator shed. After the high-voltage generator applies high voltage to the X-ray tube, the X-rays generated by the X-ray tube are irradiated onto the contamination of the disk collection unit. The module also includes a detector and a goniometer. The detector and the goniometer are also electrically connected to an electronic processing unit through a cable. When the angle between the contaminated surface of the disk collection unit and the incident X-ray ray is the Bragg angle of a certain crystal plane, the detector receives the diffraction line of the crystal plane, and the diffraction angle of the diffraction line is measured by the goniometer.

[0008] Furthermore, when the detector scans within the plane of the goniometer circle, the disk collection unit and the detector are linked at a speed of 1:2. The diffraction lines generated at the angular position that meets the Bragg condition are recorded in sequence by the detector and converted into electrical pulse signals, and the diffraction spectrum is obtained through the electronic processing unit.

[0009] The second technical solution adopted by the present invention is an online monitoring method for insulator contamination based on optical diffraction. This method utilizes the above-mentioned online monitoring system for insulator contamination based on optical diffraction. Specifically, a high-voltage generator applies high voltage to an X-ray tube. The X-rays generated by the X-ray tube irradiate the contamination on a disk collection unit. When the angle between the contamination surface and the incident X-ray line is the Bragg angle of a certain crystal plane, a detector receives diffraction lines from that crystal plane. The diffraction angle of these diffraction lines is measured by a goniometer. As the detector scans within the plane of the goniometer circle, the disk collection unit and detector operate in tandem at a ratio of 1:2. Diffraction lines generated at angular positions that meet the Bragg condition are sequentially recorded by the detector and converted into electrical pulse signals. A diffraction spectrum is generated by the electronic processing module. The main control unit packages the diffraction spectrum with data collected by the wind speed and direction sensor, rainfall sensor, and temperature and humidity sensor, and transmits the data to the monitoring center via the communication module. The monitoring center analyzes and calculates the received data to obtain the insulator's ESDD and NSDD.

[0010] Furthermore, the monitoring center obtains the contamination components, the proportion of each component, and the thickness of the contamination layer on the disc collection unit based on the diffraction spectrum; The monitoring center has a built-in equivalent model of the disk collection unit and the thickness of the pollution layer on the insulator surface. The equivalent model of the disk collection unit and the thickness of the pollution layer on the insulator surface equates the thickness of the pollution layer on the disk collection unit surface to the thickness of the pollution layer on the insulator surface; The monitoring center has a built-in trained insulator surface contamination detection model. The insulator surface contamination detection model takes the thickness of the insulator surface contamination layer, the contamination components and the proportion of each component as input, and outputs the corresponding contamination degree, that is, a neural network model built and trained to output the insulator ESDD and NSDD values.

[0011] Furthermore, the monitoring center obtained the specific pollution components based on the diffraction spectrum: Step 1.1, use MDI JADE software to smooth the diffraction spectrum, subtract the background, and subtract deal with; Step 1.2, according to whether the mathematical "second-order derivative" is 0, the diffraction spectrum processed in step 1.1 is searched for peaks, and the 2nd order derivative of the diffraction peak is obtained. , and then calculate the interplanar spacing d : (1) in, d is the interplanar spacing; is the X-ray wavelength; It is Bragg Corner; Step 1.3, according to the interplanar spacing d Calculate relative diffraction intensity I : (2) in, I is the relative diffraction intensity; F is the structure factor; P It is a multifactorial one; Step 1.4, the relative diffraction intensity of the highest peak I max Set to 100, use 100 I / I max Each diffraction peak in the diffraction spectrum is normalized to obtain the corresponding normalized relative intensity 100 I / I max ; Step 1.5, normalize the relative intensity to 100 I / I max and the corresponding interplanar spacing d Compare with the PDF card to obtain the contamination components.

[0012] Furthermore, the monitoring center obtains the specific proportion of the pollutant components based on the diffraction spectrum: For by n The filth composed of filthy ingredients, j The relationship between the proportion of the polluting components and the relative diffraction intensity is: (3) in, I j It is j The relative diffraction intensity of the pollutant components; C is a physical constant; Kj is the intensity factor; is the density of filth; V j and W j They are j Volume fraction and weight fraction of the polluting components; is the mass absorption coefficient; The monitoring center obtains the thickness of the contamination layer on the disk collection unit based on the diffraction spectrum as follows: The crystal structure of the contamination layer is determined based on the diffraction pattern, and then the thickness of the contamination layer on the disk collection unit is calculated from the lattice constant: (4) In the above formula, t is the thickness of the dirt layer; is the X-ray wavelength; h is the lattice spacing; θ is the Bragg angle corresponding to the acquisition; is the angle of incidence, that is, the angle between the X-ray and the surface normal of the disk collection unit.

[0013] Furthermore, the equivalent model of the disk collection unit and the thickness of the pollution layer on the insulator surface is obtained according to the following steps: Take several disk collection units on the tower, use the goniometer and detector to collect diffraction intensity information at different diffraction angles, and record the generated diffraction spectrum, and obtain the temperature collected by the wind speed and direction sensor, rainfall sensor, temperature and humidity sensor when collecting the corresponding diffraction spectrum. T , relative humidity H , wind speed v , Windward Angle ,rainfall R , with temperature T , relative humidity H , wind speed v , Windward Angle ,rainfall R As the input of the neural network, under the same environment, the collision conditions of particles on the disk collection unit and the insulator surface are obtained respectively. According to the definition of local collision coefficient, the local collision coefficient of the disk collection unit is obtained. and the local collision coefficient of the insulator surface , to calculate the ratio of the local collision coefficients at the corresponding positions of the insulator surface and the disk collection unit surface under the same external conditions As output, through at least 500 iterations, establish the external environment and The relationship between particles, disk collection units and insulators under different environments is obtained. Under the same environment, the equivalent model of the thickness of dirt accumulation on the disc collection unit and the insulator surface is obtained: According to formula (4), the thickness of the contamination layer on the surface of the disk collection unit is obtained, and then combined with the ratio of the local collision coefficient , which is equivalent to the insulator surface, and then the optical detection results of the disk collection unit surface are equivalent to the insulator surface, specifically: The thickness of the dirt layer on the surface of the disc collection unit t 0=[ t 01 t 02 ... t 0i ... t 0m ] is equivalent to the thickness of the dirt layer on the surface of the insulator t I =[ t I1 t I2 ... t Ii ... t Im ],Right now (5) in, t Ii Is a disc collection unit i The thickness of the pollution layer is equivalent to the thickness of the pollution layer on the surface of the insulator; t 0i Is a disc collection unit i Thickness of the pollution layer at yes i The local collision coefficient of the insulator surface corresponds to Is a disc collection unit i The local collision coefficient at m The disc collection unit and the insulator surface are divided into corresponding m local area, and then according to the local collision coefficient, the disk collection unit m The pollution thickness of each area is converted into the corresponding m The thickness of the region; The local collision coefficient between the insulator surface and the disk collection unit surface is calculated according to the following formula: (6) in, dS 0 is the area of the triangle formed by three adjacent particles at an infinite distance from the insulator or disk collection unit; dS It is the micro area enclosed by three particles that are eventually captured by the insulator or disk collection unit; v0 is the initial velocity of the particle; v 1. v 2. v 3 are the velocities of the three captured particles; According to formula (6), the local collision coefficient of the insulator surface at the corresponding position of the insulator and the disk collection unit surface is calculated: and the local collision coefficient of the disk collection unit surface .

[0014] Furthermore, the speed of the three captured particles is obtained as follows: Use RNG The model analyzes the airflow field outside the insulator and the disk collection unit, and obtains the discrete phase Lagrangian equation of the spherical particle motion: (7) in, F D ( v - v p ) is the drag force per unit mass of the pollution particle; v p is the speed of movement of the dirt particles; t p It is the movement time of the dirt particles; v is the fluid velocity; G It is the gravity exerted on the dirt particles themselves; in, F D Expressed as: (8) In the above formula, is the fluid density; is the density of dirt particles; d p is the diameter of the dirt particle; C D is the resistance rate of the dirt particles; Re is the Reynolds number; is the fluid dynamic viscosity; 、 、 is the empirical coefficient; Then, by performing integration operations on formula (7) at discrete time steps, the velocity of the pollution particle at each position on the pollution particle's trajectory can be obtained.

[0015] Furthermore, the training process of the insulator surface contamination detection model is as follows: Step 2.1, obtaining a variety of pollution samples from the area to be measured; Step 2.2: According to the solid pollution layer method, the pollution sample is evenly coated on the surface of the insulator and the thickness of the pollution layer is measured; Step 2.3, cleaning the insulator surface contaminated in step 2.2 with deionized water, measuring the conductivity of the filtrate and calculating the ESDD value of the insulator by referring to the surface area of the insulator; Step 2.4: Dry and weigh the insoluble matter in step 2.3, and calculate the NSDD value of the insulator based on the surface area; Step 2.5: Perform at least five experiments on the same contaminated sample in steps 2.2-2.4 to obtain the corresponding contamination layer thickness, insulator ESDD, and insulator NSDD values; In step 2.6, the data obtained in step 2.5 is used as a training set. Each set of training data includes the pollution components, proportions, pollution layer thickness, and corresponding pollution degrees, that is, the ESDD value and NSDD value of the insulator. The pollution components, proportions, and pollution layer thickness are used as inputs, and the corresponding pollution degrees are used as outputs to train a neural network to obtain an insulator surface pollution degree detection model, that is, to obtain the conversion relationship between the pollution components, proportions, pollution layer thickness, and ESDD and NSDD.

[0016] The beneficial effects of the present invention are: The light diffraction-based online monitoring system for insulator contamination of the present invention is equipped with a disk collection unit with the same surface area and material as the insulator to facilitate the collection of light detection data. It adopts a non-contact detection method to avoid the human errors caused by traditional detection. It is suitable for harsh conditions such as high and low temperatures, strong electricity, and strong corrosion, and has good stability.

[0017] The present invention's online insulator contamination monitoring method based on light diffraction combines contamination composition and content to characterize the insulator contamination degree, improving the accuracy of insulator status assessment. Based on the local collision coefficient, the equivalence between the disk collection unit and the insulator during the contamination accumulation process is achieved. Through extensive experiments and a BP neural network, the optical detection results are converted into traditional ESDD and NSDD. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a structural diagram of the insulator contamination online monitoring system based on light diffraction of the present invention; Figure 2 It is a schematic diagram of local collision in the on-line monitoring method for insulator contamination based on light diffraction of the present invention; Figure 3 The present invention is a flowchart of an artificial contamination test in an on-line monitoring method for insulator contamination based on light diffraction.

[0019] In the figure: 1. Main control unit, 2. Optical measurement module, 2-1. High-voltage generator, 2-2. X-ray tube, 2-3. Disk acquisition unit, 2-4. Goniometer, 2-5. Detector, 2-6. Electronic processing unit, 3. Wind speed and direction sensor, 4. Rainfall sensor, 5. Temperature and humidity sensor, 6. Communication module, 7. Monitoring center. DETAILED DESCRIPTION

[0020] The following describes it in detail with reference to specific implementation methods.

[0021] Example 1 The present invention is based on the light diffraction insulator pollution online monitoring system, such as Figure 1 As shown, it includes a main control unit 1, which is electrically connected to an optical measurement module 2, a wind speed and direction sensor 3, a rain sensor 4, a temperature and humidity sensor 5, and a communication module 6 through cables. The communication module 6 is also electrically connected to a monitoring center 7 through cables. The wind speed and direction sensor 3 is fixedly installed on the tower, the rain sensor 4 is fixedly installed on the cross arm at the top of the pole, and the temperature and humidity sensor 5 is fixedly installed below the cross arm of the tower, away from the heating components such as the conductor and insulator. 1.5 meters, the optical measurement module 2 is set on the tower of the transmission line.

[0022] The optical measurement module 2 includes a high-voltage generator 2-1 and an X-ray tube 2-2 connected in sequence, as well as a disk collection unit 2-3 fixed to the tower of the transmission line. The cross-sectional area of the disk collection unit 2-3 is equal to the surface area of the insulator, and the material is the same as the insulator shed. After the high-voltage generator 2-1 applies high voltage to the X-ray tube 2-2, the X-rays generated by the X-ray tube 2-2 are irradiated onto the contamination of the disk collection unit 2-3. The module also includes a detector 2-5 and a goniometer 2-4. The detector 2-5 and the goniometer 2-4 are also electrically connected to an electronic processing unit 2-6 via a cable. When the angle between the contaminated surface of the disk collection unit 2-3 and the incident X-ray ray is the Bragg angle of a certain crystal plane, the detector 2-5 receives the diffraction line of the crystal plane, and the diffraction angle of the diffraction line is measured by the goniometer 2-4.

[0023] When the detector 2-5 scans within the plane where the goniometer 2-4 circle is located, the disk collection unit 2-3 and the detector 2-5 are linked at a speed of 1:2. The diffraction lines generated at the angular position that meets the Bragg condition are recorded in sequence by the detector 2-5 and converted into electrical pulse signals, and the diffraction spectrum is obtained through the electronic processing unit 2-6.

[0024] The goniometer has two coaxial large and small turntables, which are linked by a gear transmission system with a transmission ratio of 1:2. The disc collection unit is fixed to the center of the small turntable by a clamp, and the detector is installed on the edge of the large turntable by a bracket. Its rotation trajectory is an arc with the goniometer rotation axis as the center. When the small turntable drives the disc collection unit to rotate When the angle is set, the large turntable drives the detector to rotate 2 horn.

[0025] Example 2 The present invention provides an online monitoring method for insulator contamination based on light diffraction, using the online monitoring system for insulator contamination based on light diffraction in Example 1. Specifically, a high-voltage generator 2-1 applies high voltage to an X-ray tube 2-2, and the X-rays generated by the X-ray tube 2-2 irradiate the contamination of the disk collection unit 2-3. When the angle between the contamination surface and the incident X-ray line is the Bragg angle of a certain crystal plane, the detector 2-5 receives the diffraction line of the crystal plane. The diffraction angle of the diffraction line is measured by the goniometer 2-4. When the detector 2-5 scans within the plane where the goniometer 2-4 is located, the disk The acquisition unit 2-3 and the detector 2-5 are linked at a speed of 1:2. The diffraction lines generated at the angular position that meets the Bragg condition are recorded in sequence by the detector 2-5 and converted into electrical pulse signals. The diffraction spectrum is obtained through the electronic processing module 2-6. The main control unit 1 packages the diffraction spectrum and the data collected by the wind speed and direction sensor 3, the rainfall sensor 4, and the temperature and humidity sensor 5, and sends them to the monitoring center 7 through the communication module 6 (such as a 4G communication module). The monitoring center 7 analyzes and calculates the received data to obtain the ESDD and NSDD of the insulator.

[0026] The monitoring center 7 obtains the contamination components, the proportion of each component, and the thickness of the contamination layer on the disc collection unit based on the diffraction spectrum; The monitoring center 7 has a built-in equivalent model of the disk collection unit and the thickness of the pollution layer on the insulator surface. The equivalent model of the disk collection unit and the thickness of the pollution layer on the insulator surface equates the thickness of the pollution layer on the surface of the disk collection unit 2-3 to the thickness of the pollution layer on the insulator surface. The monitoring center 7 has a built-in trained insulator surface contamination detection model, which is a model that converts the optical detection results of the insulator surface (contamination components, proportions, and accumulated contamination thickness) into contamination degree (ESDD, NSDD). The insulator surface contamination detection model takes the thickness of the insulator surface contamination layer, the contamination components, and the proportions of each component as inputs, and outputs the corresponding contamination degree, that is, a neural network model built and trained for the output of the insulator ESDD and NSDD values.

[0027] Example 3 Based on Example 2, the monitoring center 7 obtains the following pollution components according to the diffraction spectrum: Step 1.1, use MDI JADE software to smooth the diffraction spectrum, subtract the background, and subtract deal with; Step 1.2, according to whether the mathematical "second-order derivative" is 0, the diffraction spectrum processed in step 1.1 is searched for peaks, and the 2nd order derivative of the diffraction peak is obtained. , and then calculate the interplanar spacingd : (1) in, d is the interplanar spacing; is the X-ray wavelength; It is Bragg Corner; Step 1.3, according to the interplanar spacing d Calculate relative diffraction intensity I : (2) in, I is the relative diffraction intensity; F is the structure factor; P It is a multifactorial one; Step 1.4, the relative diffraction intensity of the highest peak I max Set to 100, use 100 I / I max Each diffraction peak in the diffraction spectrum is normalized to obtain the corresponding normalized relative intensity 100 I / I max ; Step 1.5, normalize the relative intensity to 100 I / I max and the corresponding interplanar spacing d Compare with the PDF card to obtain the contamination components.

[0028] Example 4 Based on Example 3, the monitoring center 7 obtains the specific proportion of the contamination components according to the diffraction spectrum: For n The filth composed of filthy ingredients, j The relationship between the proportion of the polluting components and the relative diffraction intensity is: (3) in, I j It is j After obtaining the pollution components, the relative diffraction intensity of each component can be found according to the PDF card; C is a physical constant; K j is the intensity factor; is the density of filth; V j and W j They are jVolume fraction and weight fraction of the polluting components; is the mass absorption coefficient; The monitoring center 7 obtains the thickness of the contamination layer on the disk collection unit based on the diffraction spectrum as follows: The crystal structure of the contamination layer is determined based on the diffraction pattern, and then the thickness of the contamination layer on the disk collection unit is calculated from the lattice constant: (4) In the above formula, t is the thickness of the dirt layer; is the X-ray wavelength; h is the lattice spacing; θ is the Bragg angle corresponding to the acquisition; is the angle of incidence, that is, the angle between the X-ray and the surface normal of the disk collection unit.

[0029] Example 5 Based on Example 4, the equivalent model of the disk collection unit and the thickness of the pollution layer on the insulator surface is obtained according to the following steps: Take several disk collection units on the tower, use the goniometer and detector to collect diffraction intensity information at different diffraction angles, and record the generated diffraction spectrum. Also, obtain the temperature collected by wind speed and direction sensor 3, rainfall sensor 4, and temperature and humidity sensor 5 when collecting the corresponding diffraction spectrum. T , relative humidity H , wind speed v , Windward Angle ,rainfall R , with temperature T , relative humidity H , wind speed v , Windward Angle ,rainfall R As the input of the neural network, under the same environment, the collision conditions of particles on the disk collection unit and the insulator surface are obtained respectively. According to the definition of local collision coefficient, the local collision coefficient of the disk collection unit is obtained. and the local collision coefficient of the insulator surface , to calculate the ratio of the local collision coefficients at the corresponding positions of the insulator surface and the disk collection unit surface under the same external conditions As output, through at least 500 iterations, establish the external environment and The relationship between particles, disk collection units and insulators under different environments is obtained. Under the same environment, the equivalent model of the thickness of dirt accumulation on the disc collection unit and the insulator surface is obtained: According to formula (4), the thickness of the contamination layer on the surface of the disk collection unit is obtained, and then combined with the ratio of the local collision coefficient , which is equivalent to the insulator surface, and then the optical detection results of the disk collection unit surface are equivalent to the insulator surface, specifically: The thickness of the dirt layer on the surface of the disc collection unit t 0=[ t 01 t 02 ... t 0i ... t 0m ] is equivalent to the thickness of the dirt layer on the surface of the insulator t I =[ t I1 t I2 ... t Ii ... t Im ],Right now (5) in, t Ii Is a disc collection unit i The thickness of the pollution layer is equivalent to the thickness of the pollution layer on the surface of the insulator; t 0i Is a disc collection unit i Thickness of the pollution layer at yes i The local collision coefficient of the insulator surface corresponds to Is a disc collection unit i The local collision coefficient at m The disc collection unit and the insulator surface are divided into corresponding m local area, and then according to the local collision coefficient, the disk collection unit m The pollution thickness of each area is converted into the corresponding m The thickness of the region; like Figure 2 As shown in the figure, the local collision coefficient between the insulator surface and the disk collection unit surface is calculated according to the following formula: (6) in, dS 0 is the area of the triangle formed by three adjacent particles at an infinite distance from the insulator or disk collection unit; dS It is the micro area enclosed by three particles that are eventually captured by the insulator or disk collection unit; v 0 is the initial velocity of the particle; v 1. v 2. v 3 are the velocities of the three captured particles; According to formula (6), the local collision coefficient of the insulator surface at the corresponding position of the insulator and the disk collection unit surface is calculated: and the local collision coefficient of the disk collection unit surface .

[0030] The velocities of the three captured particles are obtained as follows: Use RNG The model analyzes the airflow field outside the insulator and the disk collection unit, and obtains the discrete phase Lagrangian equation of the spherical particle motion: (7) in, F D ( v - v p ) is the drag force per unit mass of the pollution particle; v p is the speed of movement of the dirt particles; t p It is the movement time of the dirt particles; v is the fluid velocity; G It is the gravity exerted on the dirt particles themselves; in, F D Expressed as: (8) In the above formula, is the fluid density; is the density of dirt particles; d p is the diameter of the dirt particle; C D is the resistance rate of the dirt particles; Re is the Reynolds number; is the fluid dynamic viscosity; 、 、 is the empirical coefficient; Then, by performing integration operations on formula (7) at discrete time steps, the velocity of the contamination particle at each position on the contamination particle's trajectory can be obtained, and then the particle trajectory equation can be obtained.

[0031] Example 6 Based on Example 5, the training process of the insulator surface contamination detection model is as follows: Step 2.1, obtaining a variety of pollution samples from the area to be measured; Step 2.2: According to the solid pollution layer method, the pollution sample is evenly coated on the surface of the insulator and the thickness of the pollution layer is measured; Step 2.3, cleaning the insulator surface contaminated in step 2.2 with deionized water, measuring the conductivity of the filtrate and calculating the ESDD value of the insulator by referring to the surface area of the insulator; Step 2.4: Dry and weigh the insoluble matter in step 2.3, and calculate the NSDD value of the insulator based on the surface area; Step 2.5: Perform at least five experiments on the same contaminated sample in steps 2.2-2.4 to obtain the corresponding contamination layer thickness, insulator ESDD, and insulator NSDD values; In step 2.6, the data obtained in step 2.5 is used as a training set. Each set of training data includes the pollution components, proportions, pollution layer thickness, and corresponding pollution degrees, that is, the ESDD value and NSDD value of the insulator. The pollution components, proportions, and pollution layer thickness are used as inputs, and the corresponding pollution degrees are used as outputs to train a neural network to obtain an insulator surface pollution degree detection model, that is, to obtain the conversion relationship between the pollution components, proportions, pollution layer thickness, and ESDD and NSDD.

[0032] Example 7 On the basis of Example 6, the method of obtaining a plurality of pollution samples from the area to be measured is as follows: Figure 3 As shown, based on the surrounding environment of the measurement area, mixed salts with different pollution components and proportions are prepared as pollution test samples and thoroughly stirred. Then, one of the mixed salts is selected and evenly coated on the insulator surface according to the solid pollution layer method, and the pollution layer thickness is measured. The contaminated insulator surface is then cleaned with deionized water, and the conductivity of the filtrate is measured and compared with the insulator surface area to calculate the insulator ESDD value. The insoluble matter is filtered, dried, weighed, and the insulator NSDD value is calculated based on the surface area. A large number of pollution tests are repeated by varying the amount and preparation of the mixed salt. At least five valid tests are performed with the same mixed salt and mixed salt amount, and the average value is taken. Based on the above large amount of test data, a neural network is constructed with the pollution components, proportions, and pollution layer thickness as inputs and the corresponding pollution degree (insulator ESDD and NSDD values) as outputs to obtain the insulator surface ESDD and NSDD from the optical detection results.

Claims

1. The insulator contamination online monitoring system based on light diffraction is characterized by: The invention comprises a main control unit (1), wherein the main control unit (1) is electrically connected to an optical measurement module (2), a wind speed and direction sensor (3), a rain sensor (4), a temperature and humidity sensor (5), and a communication module (6) through cables, and the communication module (6) is also electrically connected to a monitoring center (7) through cables, wherein the wind speed and direction sensor (3) is fixedly mounted on a tower, the rain sensor (4) is fixedly mounted on a cross arm at the top of a power pole, the temperature and humidity sensor (5) is fixedly mounted below the cross arm of the pole tower, and the optical measurement module (2) is arranged on a pole tower of a power transmission line.

2. The insulator contamination online monitoring system based on light diffraction according to claim 1 is characterized in that: The optical measurement module (2) includes a high-voltage generator (2-1) and an X-ray tube (2-2) connected in sequence, and also includes a disk collection unit (2-3) fixed on a tower of a power transmission line. The cross-sectional area of the disk collection unit (2-3) is equal to the surface area of the insulator, and the material is the same as the insulator shed. After the high-voltage generator (2-1) applies high voltage to the X-ray tube (2-2), the X-rays generated by the X-ray tube (2-2) are irradiated onto the contamination of the disk collection unit (2-3). The optical measurement module (2) also includes a detector (2-5) and a goniometer (2-4). The detector (2-5) and the goniometer (2-4) are also electrically connected to an electronic processing unit (2-6) through a cable. When the angle between the contamination surface of the disk collection unit (2-3) and the incident X-ray ray is the Bragg angle of a certain crystal plane, the detector (2-5) receives the diffraction line of the crystal plane, and the diffraction angle of the diffraction line is measured by the goniometer (2-4).

3. The insulator contamination online monitoring system based on light diffraction according to claim 2 is characterized in that: When the detector (2-5) scans within the plane where the goniometer (2-4) circle is located, the disk collection unit (2-3) and the detector (2-5) are linked at a speed of 1:2, and the diffraction lines generated at the angular position that meets the Bragg condition are recorded in sequence by the detector (2-5) and converted into electrical pulse signals, and the diffraction spectrum is obtained through the electronic processing unit (2-6).

4. The method for online monitoring of insulator contamination based on light diffraction is characterized by: The online monitoring system according to claim 3 is adopted, specifically as follows: X-rays generated by the X-ray tube (2-2) are irradiated onto the contamination of the disk collection unit (2-3); when the angle between the contamination surface and the incident X-ray line is the Bragg angle of a certain crystal plane, the detector (2-5) receives the diffraction line of the crystal plane and records and converts it into an electrical pulse signal, and a diffraction spectrum is obtained through the electronic processing module (2-6); the main control unit (1) packages the diffraction spectrum and the data collected by the wind speed and direction sensor (3), the rainfall sensor (4), and the temperature and humidity sensor (5), and sends them to the monitoring center (7) through the communication module (6); the monitoring center (7) analyzes and calculates the received data to obtain the ESDD and NSDD of the insulator.

5. The method for online monitoring of insulator contamination based on light diffraction according to claim 4, characterized in that: The diffraction angle of the diffraction line received by the detector (2-5) on the crystal plane is measured by the goniometer (2-4). When the detector (2-5) scans within the plane where the goniometer (2-4) circle is located, the disk collection unit (2-3) and the detector (2-5) are linked at a speed of 1:

2. The diffraction lines generated at the angular position that meets the Bragg condition are sequentially recorded by the detector (2-5) and converted into electrical pulse signals. The monitoring center (7) obtains the pollution components, the proportion of each component, and the thickness of the pollution layer on the disk collection unit based on the diffraction spectrum; The monitoring center (7) has a built-in disk collection unit and an equivalent model of the thickness of the pollution layer on the surface of the insulator, wherein the equivalent model of the disk collection unit and the thickness of the pollution layer on the surface of the insulator equates the thickness of the pollution layer on the surface of the disk collection unit (2-3) to the thickness of the pollution layer on the surface of the insulator; The monitoring center (7) has a built-in trained insulator surface contamination detection model. The insulator surface contamination detection model takes the thickness of the insulator surface contamination layer, the contamination components and the proportion of each component as inputs to correspond to the contamination degree, that is, a neural network model built and trained for outputting the insulator ESDD and NSDD values.

6. The method for online monitoring of insulator contamination based on light diffraction according to claim 5 or claim 4, characterized in that: The monitoring center (7) obtains the pollution components according to the diffraction spectrum as follows: Step 1.1, use MDI JADE software to smooth the diffraction spectrum, subtract the background, and subtract deal with; Step 1.2, according to whether the mathematical "second-order derivative" is 0, the diffraction spectrum processed in step 1.1 is searched for peaks, and the 2nd order derivative of the diffraction peak is obtained. , and then calculate the interplanar spacing d : （1） in, d is the interplanar spacing; is the X-ray wavelength; It is Bragg Corner; Step 1.3, according to the interplanar spacing d Calculate relative diffraction intensity I : （2） in, I is the relative diffraction intensity; F is the structure factor; P It is a multifactorial one; Step 1.4, the relative diffraction intensity of the highest peak I max Set to 100, use 100 I / I max Each diffraction peak in the diffraction spectrum is normalized to obtain the corresponding normalized relative intensity 100 I / I max ; Step 1.5, normalize the relative intensity to 100 I / I max and the corresponding interplanar spacing d Compare with the PDF card to obtain the contamination components.

7. The method for online monitoring of insulator contamination based on light diffraction according to claim 6 or claim 4, characterized in that: The monitoring center (7) obtains the specific proportion of the pollutant components based on the diffraction spectrum as follows: For n The filth composed of filthy ingredients, j The relationship between the proportion of the polluting components and the relative diffraction intensity is: （3） in, I j It is j The relative diffraction intensity of the pollutant components; C is a physical constant; K j is the intensity factor; is the density of filth; V j and W j They are j Volume fraction and weight fraction of the polluting components; is the mass absorption coefficient; The monitoring center (7) obtains the thickness of the contamination layer on the disk collection unit based on the diffraction spectrum as follows: The crystal structure of the contamination layer is determined based on the diffraction pattern, and then the thickness of the contamination layer on the disk collection unit is calculated from the lattice constant: （4） In the above formula, t is the thickness of the dirt layer; is the X-ray wavelength; h is the lattice spacing; θ is the Bragg angle corresponding to the acquisition; is the angle of incidence, that is, the angle between the X-ray and the surface normal of the disk collection unit.

8. The method for online monitoring of insulator contamination based on light diffraction according to claim 7, characterized in that: The equivalent model of the disk collection unit and the thickness of the pollution layer on the insulator surface is obtained specifically according to the following steps: Take several disk collection units on the tower, use the goniometer and detector to collect the diffraction intensity information at different diffraction angles, and record the generated diffraction spectrum, and obtain the temperature collected by the wind speed and direction sensor (3), rainfall sensor (4), and temperature and humidity sensor (5) when collecting the corresponding diffraction spectrum. T , relative humidity H , wind speed v , Windward Angle ,rainfall R , with temperature T , relative humidity H , wind speed v , Windward Angle ,rainfall R As the input of the neural network, under the same environment, the collision conditions of particles on the disk collection unit and the insulator surface are obtained respectively. According to the definition of local collision coefficient, the local collision coefficient of the disk collection unit is obtained. and the local collision coefficient of the insulator surface , to calculate the ratio of the local collision coefficients at the corresponding positions of the insulator surface and the disk collection unit surface under the same external conditions As output, through at least 500 iterations, establish the external environment and The relationship between particles, disk collection units and insulators under different environments is obtained. Under the same environment, the equivalent model of the thickness of dirt accumulation on the disc collection unit and the insulator surface is obtained: According to formula (4), the thickness of the contamination layer on the surface of the disk collection unit is obtained, and then combined with the ratio of the local collision coefficient , which is equivalent to the insulator surface, and then the optical detection results of the disk collection unit surface are equivalent to the insulator surface, specifically: The thickness of the dirt layer on the surface of the disc collection unit t 0=[ t 01 t 02 ... t 0i ... t 0m ] is equivalent to the thickness of the dirt layer on the surface of the insulator t I =[ t I1 t I2 ... t Ii ... t Im ],Right now （5） in, t Ii Is a disc collection unit i The thickness of the pollution layer is equivalent to the thickness of the pollution layer on the surface of the insulator; t 0i Is a disc collection unit i Thickness of the pollution layer at yes i The local collision coefficient of the insulator surface corresponds to Is a disc collection unit i The local collision coefficient at m The disc collection unit and the insulator surface are divided into corresponding m local area, and then according to the local collision coefficient, the disk collection unit m The pollution thickness of each area is converted into the corresponding m The thickness of the region; The local collision coefficient between the insulator surface and the disk collection unit surface is calculated according to the following formula: （6） in, dS 0 is the area of the triangle formed by three adjacent particles at an infinite distance from the insulator or disk collection unit; dS It is the micro area enclosed by three particles that are eventually captured by the insulator or disk collection unit; v 0 is the initial velocity of the particle; v 1. v 2. v 3 are the velocities of the three captured particles; According to formula (6), the local collision coefficient of the insulator surface at the corresponding position of the insulator and the disk collection unit surface is calculated: and the local collision coefficient of the disk collection unit surface .

9. The method for online monitoring of insulator contamination based on light diffraction according to claim 8, characterized in that: The speed of the three captured particles is obtained as follows: Use RNG The model analyzes the airflow field outside the insulator and the disk collection unit, and obtains the discrete phase Lagrangian equation of the spherical particle motion: （7） in, F D ( v - v p ) is the drag force per unit mass of the pollution particle; v p is the speed of movement of the dirt particles; t p It is the movement time of the dirt particles; v is the fluid velocity; G It is the gravity exerted on the dirt particles themselves; in, F D Expressed as: （8） In the above formula, is the fluid density; is the density of dirt particles; d p is the diameter of the dirt particle; C D is the resistance rate of the dirt particles; Re is the Reynolds number; is the fluid dynamic viscosity; 、 、 is the empirical coefficient; Then, by performing integration operations on formula (7) at discrete time steps, the velocity of the pollution particle at each position on the pollution particle's trajectory can be obtained.

10. The method for online monitoring of insulator contamination based on light diffraction according to claim 8, characterized in that: The training process of the insulator surface contamination detection model is as follows: Step 2.1, obtaining a variety of pollution samples from the area to be measured; Step 2.2: According to the solid pollution layer method, the pollution sample is evenly coated on the surface of the insulator and the thickness of the pollution layer is measured; Step 2.3, cleaning the insulator surface contaminated in step 2.2 with deionized water, measuring the conductivity of the filtrate and calculating the ESDD value of the insulator by referring to the surface area of the insulator; Step 2.4: Dry and weigh the insoluble matter in step 2.3, and calculate the NSDD value of the insulator based on the surface area; Step 2.5: Perform at least five experiments on the same contaminated sample in steps 2.2-2.4 to obtain the corresponding contamination layer thickness, insulator ESDD, and insulator NSDD values; In step 2.6, the data obtained in step 2.5 is used as a training set. Each set of training data includes the pollution components, proportions, pollution layer thickness, and corresponding pollution degrees, that is, the ESDD value and NSDD value of the insulator. The pollution components, proportions, and pollution layer thickness are used as inputs, and the corresponding pollution degrees are used as outputs to train a neural network to obtain an insulator surface pollution degree detection model, that is, to obtain the conversion relationship between the pollution components, proportions, pollution layer thickness, and ESDD and NSDD.

Images