Resonant deicing method based on inherent frequency of power transmission line

Abstract

The invention relates to the technical field of distribution network overhead line operation and maintenance, and particularly discloses a power transmission line inherent frequency-based resonant deicing method, which comprises the following steps that: firstly, an ice observation module performs icing detection on a power transmission line by using an AI image recognition technology; and then, the background management system starts the vibration sensor according to a detection result to obtain inherent frequency information of the line. Then, a vibration module starting signal is achieved under the system, the vibration module generates PWM waves to control the vibration frequency, and amplitude data is fed back in real time through an accelerometer so as to optimize the vibration effect. And finally, the vibration module continuously monitors the icing condition in the working process until the ice layer is completely removed. According to the resonance type deicing method, external excitation frequency is applied on the basis of the inherent frequency of the power transmission line, so that the power transmission line generates a resonance phenomenon, the vibration amplitude of the distribution network overhead line is enhanced, the power transmission line is waved, and anti-icing protection of the distribution network overhead line is carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network overhead line operation and maintenance, and in particular to a resonance deicing method based on the natural frequency of a transmission line. Background Art

[0002] Winter is a low-temperature, low-rain climate. When there is an inversion layer in the lower atmosphere and the surface temperature is between -5 and -1°C, the raindrops on it will freeze into rime and quickly cover the transmission lines, which can easily form line icing. When it snows, snow is also likely to accumulate on the transmission lines. When the ice load generated by ice and snow exceeds a certain limit, it will cause the tower to collapse and the power supply to be interrupted, endangering the safety of power transmission. Icing of transmission lines has caused great harm to countries around the world.

[0003] At present, in the case of thickly covered transmission lines, the only way to remove ice is to use manual deicing. However, manual deicing is not only less safe, but also less efficient. Manual deicing methods are sometimes limited by terrain, landform, climate, safety and other conditions, and are difficult to meet the needs of large-scale deicing tasks in terms of work efficiency. When the ice on the transmission line is thick and has been frozen for a long time, manual deicing often fails to completely remove the ice, is prone to secondary freezing, and even cannot be effectively deiced.

[0004] The icing problem of transmission lines has always been a significant challenge faced by some regions in winter. It not only threatens the stable operation of the power system, but also increases the safety risks of workers when performing de-icing operations. Summary of the invention

[0005] In view of the current problems, the present invention provides a resonant de-icing method based on the natural frequency of a transmission line. The resonant de-icing method applies an external excitation frequency based on the natural frequency of the transmission line to cause the transmission line to resonate, thereby enhancing the vibration amplitude of the distribution network overhead line and causing the transmission line to dance, thereby performing anti-icing protection for the distribution network overhead line.

[0006] The technical solution adopted by the present invention is: A resonant deicing method based on the natural frequency of a power transmission line, the resonant deicing method is based on installing an ice observation module and a vibration module at intervals on an overhead line of a distribution network, and comprises the following steps: Step 1: The ice observation module uses AI image recognition to detect ice coverage on the distribution network overhead lines; Step 2: The background management system starts the vibration sensor according to the icing detection result of the distribution network overhead line and collects the natural frequency data of the distribution network overhead line through the vibration sensor when it is confirmed that there is icing. Step 3: The background management system sends a start signal to the vibration module based on the natural frequency data of the distribution network overhead line collected by the vibration sensor; when the vibration module is started, the system parameters are fully initialized and enter a continuous operation state; Step 4, when the vibration module is working, its single chip microcomputer generates a PWM wave to control the vibration frequency of the vibration generating device, and at the same time, the accelerometer of the vibration module collects acceleration data and calculates it into amplitude data, which is used as feedback to control the vibration generating device to always operate at the maximum amplitude point; Step 5: The vibration module continues to work, and uses AI image recognition through the ice observation module to detect ice coverage on the distribution network overhead lines until the background management system recognizes that the ice coverage has been cleared through the ice observation module, and then controls the vibration module to stop; The resonant deicing method applies an external excitation frequency through a vibration module based on the natural frequency of the transmission line to cause the transmission line to resonate, and enhances the vibration amplitude of the distribution network overhead line by adjusting the vibration frequency, so that the transmission line dances to perform anti-icing protection for the distribution network overhead line.

[0007] Furthermore, ice detection is performed in step 1, including: collecting distribution network overhead line images, external environment temperature and humidity data and distribution network overhead line temperature data, and transmitting the distribution network overhead line images, external environment temperature and humidity data and distribution network overhead line temperature data to the background management system.

[0008] Further, in step 2, the background management system performs analysis based on the distribution network overhead line image, the external environment temperature and humidity data, and the distribution network overhead line temperature data to obtain the distribution network overhead line icing detection result; the analysis process includes the following steps: Step 2.1, the background management system performs image enhancement based on the distribution network overhead line image through grayscale correction to obtain a pre-processed image; Step 2.2, the background management system uses wavelet transform or morphological method to detect the edge of the image; Step 2.3, the background management system performs boundary tracking based on the angle bisector fitting method to obtain the image edge extraction result; uses pixel-level fusion to fuse the image; reads the bare wire image data and compares it with the image fusion data, and outputs the distribution network overhead line ice coverage determination result.

[0009] Furthermore, in step 2.1, the grayscale correction adopts histogram equalization, which makes the grayscale distribution of the whole image more uniform by adjusting the grayscale value of the pixel based on the principle of probability theory; If the number of gray levels of a grayscale image is n, assuming that the probability of a pixel with gray level i appearing is , the entropy corresponding to the i-level grayscale in the image is: ; In the formula, is the entropy corresponding to the i-level grayscale in the image; is the probability of occurrence of a pixel with gray level i.

[0010] Further, in step 4, based on the synergistic effect of the three algorithms of Kalman filtering, Fourier transform and time domain integration, the acceleration data collected by the accelerometer is deeply processed to calculate the amplitude data, which specifically includes the following steps: Step 4.1, the ice and snow on the overhead line of the overhead distribution network generate forced vibration under the action of the uninterrupted periodic external force of the vibration generating device, and the periodic external force generated by the rotation of the vibration generating device is the driving force; based on the driving force, the motion equation of the forced vibration of the overhead line of the distribution network is obtained; based on the equal amplitude motion, the position of the overhead line of the distribution network is calculated; according to the theoretical calculation, the amplitude data is obtained; Step 4.2, based on the amplitude data obtained in step 4.1, perform Kalman filter calculation. The Kalman filter will generate an estimate of the unknown variable based on the values ​​measured at different times, considering the joint distribution at each time; Step 4.3, convert the time domain signal into frequency domain representation through Fourier transform; Step 4.4, convert the acceleration vibration signal into velocity or displacement signal through time domain integration.

[0011] Furthermore, in step 4.1, the driving force F is set to: ; In the formula, is the driving force amplitude, is the angular frequency of the driving force; t is the time; Under the driving force, the equation of motion is: ; make , , we get: ; In the formula, is the mass of the object; represents the position of the object; t is the time; is the elastic constant; is the magnitude of the driving force; is the angular frequency of the driving force; is the natural frequency; β is the damping coefficient; Based on the constant amplitude motion, we get: ; In the formula, Represents the location of an object; is the amplitude; is the angular frequency of the driving force; t is the time; is the initial phase, indicating the position phase at the beginning of the movement; According to theoretical calculation, the amplitude is: ; In the formula, is the magnitude of the driving force; is the mass of the object; is the natural frequency; is the angular frequency of the driving force; β is the damping coefficient.

[0012] Further, step 4.2 includes the following steps: Predicted state estimation: ; In the formula, is the predicted estimate of the state at time k; for The optimal estimate of the state when is the state transfer matrix; is the control matrix; It is the external control input; Forecast error covariance: ; In the formula, is the prediction error covariance matrix of k; represents the state transfer matrix, which is used to describe the change of state from time k-1 to time k; The optimal estimated covariance matrix for the state error when is k-1; is the covariance matrix of process noise; is the state transfer matrix The transpose of represents the impact on error propagation; Update the Kalman gain: ; In the formula, represents the Kalman gain; is the prediction error covariance matrix of k; is the observation matrix; is the observation matrix The transpose of is the covariance matrix of the observation noise; Update the state estimate: ; In the formula, represents the updated state estimate at time k; is the predicted estimate of the state at time k; represents the Kalman gain; is the observed value at time k; is the observation matrix; Update the error covariance: ; In the formula, is the state error covariance matrix after update at time k; is the identity matrix; represents the Kalman gain; is the observation matrix; is the prediction error covariance matrix of k; The optimal estimation of the system state is performed by predicting the state estimate, predicting the error covariance, updating the Kalman gain, updating the state estimate and updating the error covariance.

[0013] Furthermore, in step 4.3, the N-point discrete Fourier transform of the N-point sequence x(r) can be expressed as: ; In the formula, is the frequency domain signal after discrete Fourier transform DFT, which represents the complex amplitude and phase information corresponding to frequency k; represents the time domain signal, which represents the sample value at time r; represents a complex exponential function, which is used to convert the time domain signal into the frequency domain to form a basis function of frequency k; j is an imaginary unit; N is the total number of sampling points of the signal; k is the index of the frequency domain signal, ranging from 0 to N-1; r is the index of the time domain signal, ranging from 0 to N-1.

[0014] Further, in step 4.4, the time domain integration algorithm is as follows: ; In the formula, The integration result at time k represents the integral value calculated from the time domain signal x; is the time step, which indicates the time interval between two sampling points; is the sampling value of the time domain signal, the signal value at time i; i is the index of the integration process, representing each sampling point from 1 to k; k represents the current moment index, integrating from 1 to N.

[0015] Furthermore, the power supply module installed on the overhead line of the distribution network is also installed to supply power to the ice observation module and the vibration module; the power supply module includes a current transformer, an energy storage element connected to the current transformer, and a rectifier circuit connected in parallel with the energy storage element; the rectifier circuit includes a clamping circuit, a rectifier filter circuit, and a DC / DC module located on the connection line between the current transformer and the energy storage element; the energy storage element adopts a lithium iron phosphate battery and a super capacitor; When the primary transmission current is small, the energy storage element serves as supplementary energy storage to supply power to the ice viewing module and the vibration module; when the primary current is large and the secondary output power is large, the current transformer directly supplies power to the ice viewing module and the vibration module, and stores the excess electric energy in the energy storage element; According to electromagnetic theory, assuming the peak value of the output voltage is U, we get: ; In the formula, is the frequency of the coil excitation current; N2 is the number of turns of the special coil; is the magnetic flux amplitude of the coil cross section; Bm is the magnetic induction intensity amplitude; S is the core cross-sectional area; λ is the core lamination coefficient; According to Ampere's circuit law: ; In the formula, is the magnetic field intensity amplitude; is the number of turns of the primary winding; L is the average magnetic path length; I is the current in the primary winding; The relationship between the magnetic induction intensity amplitude Bm and the magnetic field intensity amplitude Hm is: ; In the formula, is the vacuum permeability; is the relative magnetic permeability; The number of turns of the corresponding energy harvesting coil can be calculated, and a current transformer suitable for the de-icing device can be selected to inductively harvest power from the transmission line to supply power to the ice observation module and the vibration module.

[0016] Beneficial effects: The beneficial effects of the resonant deicing method based on the natural frequency of the transmission line are mainly reflected in the following aspects: 1. Improve de-icing efficiency: By utilizing the natural frequency of the transmission line, this method can produce a resonance effect at a specific frequency, which can effectively vibrate and remove ice and snow. This resonance enhances the vibration amplitude, making the de-icing process faster and more effective, reducing the time and labor costs required for de-icing operations.

[0017] 2. Improved safety: Traditional manual deicing operations have high safety risks, especially in severe winter weather conditions. Based on this resonant deicing method, the automated vibration module and AI technology are fully utilized to reduce the reliance on manual operations, thereby significantly reducing the safety risks of workers during deicing operations.

[0018] 3. Accurate real-time monitoring: AI image recognition technology and environmental data monitoring system are used to track and monitor the icing of power transmission lines in real time. By analyzing road conditions and climate change, the backend management system can make timely decisions to ensure the best deicing time.

[0019] 4. Strong adaptability: This method can be adjusted according to different environmental conditions and line parameters, greatly enhancing the adaptability of the system. For example, the vibration frequency can be set according to the external temperature and humidity and the natural frequency of the line in different distribution network overhead line sections to ensure the best deicing effect.

[0020] 5. Reduce maintenance costs The need for manual intervention and equipment maintenance is reduced, reducing operation and maintenance costs. In addition, since it can effectively prevent the accumulation of ice, it can reduce the failure rate and maintenance frequency of equipment in the long run, thereby achieving lower long-term maintenance costs.

[0021] 6. Friendly environmental impact: By reducing the reliance on chemical de-icing agents, this method is particularly suitable for areas with high environmental protection requirements. De-icing is achieved through physical vibration, reducing potential environmental pollution.

[0022] 7. Enhanced power supply reliability: It eliminates the negative impact of icing on transmission lines, improves the stability and reliability of the power system, thereby ensuring continuous power supply in severe winter weather and reducing power outages caused by icing.

[0023] In summary, the resonant de-icing method based on the natural frequency of the transmission line shows significant advantages in improving efficiency, safety and economy, and provides an innovative solution for the maintenance and operation of overhead lines in modern power system distribution networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0025] Figure 1 This is the wire vibration waveform during a single tap; Figure 2 This is the vibration waveform of the wire when the exciting force is applied; Figure 3 This is the vibration waveform of the wire when the exciting force is applied; Figure 4 This is the vibration waveform of the wire when the exciting force is applied; Figure 5 This is the vibration waveform of the ice-covered conductor when struck once; Figure 6 It is the vibration waveform of ice-covered conductor when low-frequency exciting force is applied; Figure 7 It is the vibration waveform of ice-covered conductor when high-frequency exciting force is applied; Figure 8 It is the vibration waveform of ice-covered conductor when the exciting force is applied to reach resonance; Fig. 9 It is a schematic diagram of installing the ice viewing module and the vibration module of the present invention; Fig.10 This is a flow chart of ice detection according to the present invention; Fig.11 The pre-processed image of the power transmission line obtained by the image enhancement of the present invention; wherein, Figure (a) is the original grayscale image, and Figure (b) is the enhanced image; Fig.12 For the present invention Fig.11 Histogram of the original grayscale image in ; Fig.13 For the present invention Fig.11 Histogram of the image after mid-gray enhancement; Fig.14 This is a flow chart of the background management system of the present invention regulating the vibration module based on the vibration sensor; Fig.15 It is a flow chart of the steps of deep processing of control data by the background management system of the present invention; Fig.16 This is a power supply principle diagram of the power supply module of the present invention; In the figure, 1 is an optical camera, 2 is a 4G antenna, 3 is an air temperature and humidity sensor, 4 is a wire temperature sensor, 5 is a vibration sensor, 6 is an ice viewing module, 7 is a vibration module, 8 is a power supply module, 9 is a single chip microcomputer, and 10 is a vibration generating device. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings of the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0027] In the power transmission system, both a single conductor and the complete transmission system have their natural frequencies, but the natural frequency characteristics of the two are different. For a single distribution network overhead line, its natural frequency mainly depends on the elastic modulus of the material of the distribution network overhead line, which is a basic mechanical parameter of the distribution network overhead line. In the entire transmission system, the characteristics of the natural frequency are not only affected by the material properties, but also significantly affected by gravity. In the overall transmission system, if the equipment matches the natural frequency of the system, resonance may occur, resulting in widespread dancing of the transmission line, which needs to be actively prevented. However, for a single distribution network overhead line section, the resonance phenomenon can be used to increase the amplitude of the distribution network overhead line, thereby enhancing the transmission effect of signals or energy. It is worth noting that due to the difference in vibration amplitude of the distribution network overhead line in different sections, this may cause slight distortion of the distribution network overhead line, which is beneficial for deicing operations because it can improve deicing efficiency. By adjusting the tension and support points of the distribution network overhead lines, these vibration characteristics can be controlled and utilized to a certain extent to achieve anti-icing protection of the distribution network overhead lines while avoiding the galloping problem caused by large-scale resonance.

[0028] Overhead lines of power transmission and distribution networks are usually made of steel-core aluminum stranded wires, which have a certain elasticity and can resist elastic deformation when subjected to force. Therefore, when the overhead lines of the distribution network are struck, they will deform to a certain extent. Under the application of a certain excitation force, the frequency is gradually adjusted to observe the response of the vibration system. It is found that when the excitation frequency approaches the natural frequency, the vibration amplitude of the system will increase significantly. This is because when the external excitation frequency matches the natural frequency of the system, resonance occurs, causing the amplitude to reach a peak. This phenomenon reveals the importance of resonance to the response of the vibration system.

[0029] Therefore, the test is carried out; when the distribution network overhead line is struck once, the amplitude and distance waveforms generated are as follows Figure 1 When a continuous excitation force is added to the distribution network overhead line, the amplitude and distance waveforms generated by the distribution network overhead line are as follows: Figure 2 When another exciting force of different frequency is applied to the distribution network overhead line, the amplitude and distance waveforms generated by the distribution network overhead line are as follows: Figure 3 When multiple exciting forces of different frequencies are applied to the distribution network overhead line, when the frequency of the exciting force applied is the same as the natural frequency of the distribution network overhead line, the amplitude and distance waveforms of the distribution network overhead line will be amplified, but the frequency will not change. The amplified waveform is as follows: Figure 4 shown.

[0030] When exciting forces of different frequencies are applied to the distribution network overhead lines, especially when the frequency of the exciting force matches the natural frequency of the distribution network overhead lines, the distribution network overhead lines will cause resonance. This resonance phenomenon will significantly enhance the vibration amplitude of the distribution network overhead lines, allowing the vibration signal to propagate farther. It can be clearly observed from the previous simulation graphics that the frequency of the exciting force has a significant effect on the vibration transmission effect of the distribution network overhead lines. In particular, when the frequency of the exciting force coincides with the natural frequency of the distribution network overhead lines, it will significantly enhance the vibration signal, thereby extending the transmission distance of the vibration. This phenomenon reveals the important role of frequency matching in enhancing vibration transmission.

[0031] Similarly, for ice-covered distribution network overhead lines, the excitation forces of different frequencies also have different effects on the propagation of distribution network overhead line vibration. The effects of a single strike and different frequency excitation forces on the vibration propagation distance of ice-covered distribution network overhead lines are as follows: Figures 5 to 8 shown.

[0032] In summary, whether it is an un-iced distribution network overhead line or an iced distribution network overhead line, when the frequency of the exciting force applied to the distribution network overhead line is the same as the natural frequency of the distribution network overhead line, a resonance phenomenon will occur to amplify the vibration amplitude of the distribution network overhead line, so that the vibration signal of the distribution network overhead line is transmitted farther.

[0033] Based on the above principles, this embodiment proposes a resonant deicing method based on the natural frequency of the power transmission line. The resonant deicing method is to install an ice-viewing module 6 and a vibration module 7 at intervals on the overhead line of the distribution network. The ice-viewing module 6 and the vibration module 7 are as follows: Fig. 9 As shown, the resonant deicing method applies an external excitation frequency based on the natural frequency of the transmission line to make the transmission line resonate, enhance the vibration amplitude of the distribution network overhead line, and make the transmission line dance, thereby performing anti-icing protection for the distribution network overhead line. Specifically, it includes the following steps: Step 1, the ice observation module 6 uses AI image recognition to detect ice coating on the distribution network overhead lines; like Fig. 9 As shown, the hardware of the ice viewing module 6 includes a 5-megapixel optical camera 1 equipped with a heating component, a 4G antenna 2, an air temperature and humidity sensor 3, and a conductor temperature sensor 4; a through hole is provided on the axial side of the shell of the ice viewing module 6 for the distribution network overhead line to pass through, so as to facilitate installation on the distribution network overhead line.

[0034] Icing detection includes: collecting images of overhead lines of the distribution network through a 5-megapixel optical camera 1, collecting external environment temperature and humidity data through an air temperature and humidity sensor 3, and collecting temperature data of overhead lines of the distribution network through a conductor temperature sensor 4; the ice observation module 6 transmits the images of overhead lines of the distribution network, external environment temperature and humidity data, and temperature data of overhead lines of the distribution network to the background management system through a 4G antenna 2.

[0035] Step 2: The background management system starts the vibration sensor 5 according to the icing detection result of the distribution network overhead line and collects the natural frequency data of the distribution network overhead line through the vibration sensor 5 when it is confirmed that there is icing.

[0036] The background management system analyzes the distribution network overhead line image, external environment temperature and humidity data, and distribution network overhead line temperature data to obtain the distribution network overhead line ice detection results; Fig.10 As shown, the analysis process includes the following steps: Step 2.1, the background management system performs image enhancement based on the distribution network overhead line image through grayscale correction to obtain a preprocessed image. In practical applications, the collected images often contain various noises, and the clarity and resolution may not be satisfactory. Therefore, the station management system must first perform a preprocessing step on the original image, which includes reducing the impact of noise and enhancing the useful information in the image.

[0037] Image enhancement is one of the effective means to improve image quality. Especially in complex geographical environments, due to factors such as weather changes, differences in lighting conditions, and mechanical vibration or jitter, the collected images may appear brighter or darker overall, resulting in blurred images and low recognition. This uneven distribution of gray values ​​will increase the difficulty of subsequent edge extraction. In order to optimize the effect of edge detection, appropriate image enhancement methods need to be used to improve the overall clarity and feature expression of the image, thereby improving image quality and enriching the information content.

[0038] Grayscale correction uses histogram equalization. Based on the principle of probability theory, the grayscale value of the pixel is adjusted to make the grayscale distribution of the entire image more uniform. After histogram equalization, the contrast of the image is improved, and the details are more clearly visible, which helps to improve the subsequent edge detection and analysis. This method can significantly improve the quality of ice monitoring images and provide strong support for accurate identification and analysis of ice conditions.

[0039] Probability theory principle: If the number of gray levels of a grayscale image is n, and the probability of a pixel with gray level i appearing is , the entropy corresponding to the i-level grayscale in the image is: ; In the formula, is the entropy corresponding to the i-level grayscale in the image; is the probability of occurrence of a pixel with gray level i.

[0040] like Fig.11As shown in the figure, the contrast of the transmission line preprocessed image obtained after image enhancement is significantly enhanced compared with the original image, the overall level is clearer, and the line edge is more obvious. The grayscale histogram of the original grayscale image and the enhanced image is shown in Fig.12 and Fig.13 As shown in the grayscale histogram, the horizontal axis and the vertical axis represent the grayscale level and the frequency of occurrence of the grayscale level respectively. By comparison, it can be seen that in the image after grayscale enhancement, the maximum number of occurrences of a single grayscale level has dropped from 1613 times to 1271 times, and the distribution of each grayscale level in the entire grayscale space is more uniform. The information contained in the image is enriched, which is conducive to the next step of extracting the edge of the ice-covered transmission line.

[0041] Step 2.2: The background management system uses wavelet transform or morphological method to perform edge detection on the image.

[0042] The core of the ice detection method for power transmission lines based on image processing is to compare the edge radial width changes of the un-iced conductor image and the iced conductor image, that is, the conductor diameter becomes larger after ice coating, and its diameter is equal to the sum of the conductor diameter when not ice coated and the ice thickness of the transmission line. By accurately identifying the edge contour of the ice-covered conductor, the edge detection of the transmission line is realized.

[0043] Different graphics have different grayscales, but there is always an edge at the boundary of each graphic, and the edge of the graphic can be used to separate the image. The boundary of an object is not the same as the edge of a graphic. The boundary of an object refers to the boundary between an object and other objects in reality, while the edge of a graphic refers to the location where the pixel value suddenly changes in the image. The pixel value sudden change point may not be the boundary, but may also be a part of the object. If the object is similar in color to its surrounding environment, the pixel points at its boundary may not have obvious sudden changes.

[0044] Therefore, this embodiment provides two edge detection methods: Method 1: Morphological method: The morphological method is based on existing technology. The existing edge extraction algorithm has been improved to achieve better extraction effect. The principle of the classic edge detection algorithm is to detect the edge of the image based on the spatial differential operator and image convolution calculation. According to different differential operators, it is divided into the following methods, including Roberts operator, Sobel operator, Prewitt operator, LoG operator, Canny operator, etc. These methods are divided into first-order and second-order. The first-order algorithm refers to the use of the first-order derivative information of the image to detect the edge of the image. This method is often difficult to effectively find the boundary of an image with a relatively uniform grayscale value change. At this time, it is necessary to extract the second-order derivative information of the image as a reference, but the second-order derivative information of the image is often mixed with a lot of noise. Therefore, before edge detection of the image, the noise information in the image should be eliminated as much as possible to facilitate subsequent processing and recognition.

[0045] Method 2 uses wavelet transform: Existing technologies include edge detection methods based on wavelet transform, edge detection methods based on mathematical morphology, edge detection methods based on genetic algorithms, and edge detection methods based on neural networks. When wavelet transform is applied to the field of image processing, it has the following significant advantages: First, using wavelets to decompose the target signal can make it cover the entire frequency domain, which provides a complete mathematical description in theory; second, by selecting appropriate filters, wavelet transform can greatly reduce the correlation between different features, making its results relatively independent; third, wavelet transform has a unique "zoom" feature, that is, when processing high-frequency signals, a narrow analysis window can be used, that is, low frequency resolution and high time resolution can be selected, and when processing low-frequency signals, a wide analysis window can be used, that is, high frequency resolution and low time resolution can be selected, so that the local characteristics of the signal can be better displayed according to needs; fourth, wavelet transform has a fast algorithm when implemented.

[0046] Step 2.3, the background management system performs boundary tracking based on the angle bisector fitting method to obtain the image edge extraction result; uses pixel-level fusion to fuse the image; reads the bare wire image data and compares it with the image fusion data, and outputs the distribution network overhead line ice coverage determination result.

[0047] When the background management system confirms that there is ice covering, it starts the vibration sensor 5 and collects the natural frequency data of the distribution network overhead line through the vibration sensor 5, such as Fig.14 As shown, perform the following steps: Step 3, the background management system sends a start signal to the vibration module 7 based on the natural frequency data of the distribution network overhead line collected by the vibration sensor 5; when the vibration module 7 is started, the system parameters are fully initialized first, and then enters a continuous operation working state.

[0048] like Fig. 9 As shown, the vibration sensor 5 in this embodiment is installed in the ice viewing module 6. Since the ice viewing module 6 is equipped with a 4G antenna 2, installing it in the ice viewing module 6 facilitates data transmission between the vibration sensor 5 and the background management system.

[0049] like Fig. 9As shown, the vibration module 7 in this embodiment includes a single-chip microcomputer 9 installed on a circuit board in the second housing, a vibration generating device 10 connected to the single-chip microcomputer 9 and installed on the overhead line of the distribution network, and an accelerometer. The single-chip microcomputer 9 adopts the STM32L476RET6 single-chip microcomputer, and the single-chip microcomputer 9 includes a central processing unit, a power supply, a clock generation circuit and a reset mechanism; in addition, in order to facilitate the simulation software used during modulation, a program debugging interface circuit and a running simulation loop are also set. The accelerometer adopts the BOSCH BMI270 six-axis gyroscope IC. The accelerometer can obtain accurate acceleration data and obtain the current system amplitude through the algorithm.

[0050] Step 4: When the vibration module 7 is working, its single-chip microcomputer 9 generates a PWM wave to control the vibration frequency of the vibration generating device 10. At the same time, the accelerometer of the vibration module collects acceleration data and calculates it as amplitude data, which is used as feedback to control the vibration generating device 10 to always operate at the maximum amplitude point. The output starts from 0, and the PWM wave generated by the single-chip microcomputer 9 increases continuously until the accelerometer detects the maximum amplitude and records the frequency at that moment; finally, the frequency PWM wave output is locked until the end.

[0051] Specifically, based on the synergy of three algorithms, Kalman filtering, Fourier transform and time domain integration, the acceleration data collected by the accelerometer is deeply processed to calculate the amplitude data, such as Fig.15 As shown, the specific steps include: Step 4.1: The ice and snow on the overhead wire generate forced vibrations under the action of the uninterrupted periodic external force of the vibration generating device 10. The periodic external force generated by the rotation of the vibration generating device 10 is the driving force. The driving force F is set as: ; In the formula, is the driving force amplitude, is the angular frequency of the driving force; t is the time; Under the driving force, the equation of motion is: ; make , , we get: ; In the formula, is the mass of the object; represents the position of the object; t is the time; is the elastic constant; is the magnitude of the driving force; is the angular frequency of the driving force; is the natural frequency; β is the damping coefficient.

[0052] Based on the constant amplitude motion, we get: ; In the formula, Represents the location of an object; is the amplitude; is the angular frequency of the driving force; t is the time; is the initial phase, which indicates the position phase when the motion starts.

[0053] According to theoretical calculation, the amplitude is: ; In the formula, is the magnitude of the driving force; is the mass of the object; is the natural frequency; is the angular frequency of the driving force; β is the damping coefficient.

[0054] Step 4.2, based on the amplitude data obtained in step 4.1, Kalman filter calculation is performed. Kalman filter will generate an estimate of the unknown variable based on the values ​​measured at different times, considering the joint distribution at each time.

[0055] Specifically, the predicted state estimate is: ; In the formula, is the predicted estimate of the state at time k; for The optimal estimate of the state when is the state transfer matrix; is the control matrix; For external control input.

[0056] Forecast error covariance: ; In the formula, is the prediction error covariance matrix of k; represents the state transfer matrix, which is used to describe the change of state from time k-1 to time k; The optimal estimated covariance matrix for the state error when is k-1; is the covariance matrix of process noise; is the state transfer matrix The transpose of , represents the impact on error propagation.

[0057] Update the Kalman gain: ; In the formula, represents the Kalman gain; is the prediction error covariance matrix of k; is the observation matrix; is the observation matrix The transpose of is the covariance matrix of the observation noise.

[0058] Update the state estimate: ; In the formula, represents the updated state estimate at time k; is the predicted estimate of the state at time k; represents the Kalman gain; is the observed value at time k; is the observation matrix.

[0059] Update the error covariance: ; In the formula, is the state error covariance matrix after update at time k; is the identity matrix; represents the Kalman gain; is the observation matrix; is the prediction error covariance matrix for k.

[0060] The optimal estimation of the system state is performed by predicting the state estimate, predicting the error covariance, updating the Kalman gain, updating the state estimate and updating the error covariance.

[0061] In step 4.2, the Kalman filter method is used to estimate the state of the acquired amplitude data, so that noise and uncertainty are taken into account during the measurement process. Improving the system's ability to resist interference with signals can more accurately estimate the true state, especially in the presence of uncertainty and measurement errors. Through the prediction and update mechanism of the dynamic system, real-time monitoring and control become more accurate and reliable, and the credibility of the data is enhanced.

[0062] Step 4.3, the signal is transformed from the time domain description to the frequency domain description is called time-frequency domain transformation: The N-point discrete Fourier transform of the N-point sequence x(r) can be expressed as: ; In the formula, is the frequency domain signal after discrete Fourier transform DFT, which represents the complex amplitude and phase information corresponding to frequency k; represents the time domain signal, which represents the sample value at time r; represents a complex exponential function, which is used to convert the time domain signal into the frequency domain to form a basis function of frequency k; j is an imaginary unit; N is the total number of sampling points of the signal; k is the index of the frequency domain signal, ranging from 0 to N-1; r is the index of the time domain signal, ranging from 0 to N-1.

[0063] In step 4.3, the time domain signal is converted into a frequency domain representation for analyzing the frequency characteristics and spectrum features of the signal. Frequency domain analysis can reveal the frequency components and energy distribution in the signal.

[0064] Step 4.4, convert the acceleration vibration signal into velocity or displacement signal by time domain integration: The time domain integration algorithm is as follows: ; In the formula, The integration result at time k represents the integral value calculated from the time domain signal x; is the time step, which indicates the time interval between two sampling points; is the sampling value of the time domain signal, the signal value at time i; i is the index of the integration process, representing each sampling point from 1 to k; k represents the current moment index, integrating from 1 to N.

[0065] In step 4.4, the acceleration signal is converted into a velocity or displacement signal by time domain integration. This provides a complete description of the vibration signal, and the velocity or displacement signal is more easily associated with the actual motion state of the equipment. In vibration analysis, the obtained velocity and displacement information can be used to more effectively evaluate dynamic characteristics and safety.

[0066] Through the linkage of step 4.1 to step 4.4, the acceleration data collected by the accelerometer can be processed and analyzed in depth and comprehensively, and more accurate amplitude data can be obtained. This is of great significance for the monitoring and maintenance of overhead lines in distribution networks, ensuring the safety and stability of power supply.

[0067] Based on step 3 and step 4, the background management system collects the natural frequency data of the distribution network overhead line through the vibration sensor 5, and adjusts the vibration module 7 to apply an exciting force to the distribution network overhead line. After one or more adjustment processes, the external excitation frequency is matched with the natural frequency, a resonance phenomenon is generated, and the vibration amplitude of the conductor is enhanced.

[0068] In step 5, the vibration module 7 continues to work, and uses AI image recognition through the ice observation module 6 to detect ice coverage on the distribution network overhead lines until the background management system recognizes through the ice observation module 6 that the ice coverage has been cleared, and then controls the vibration module 7 to stop.

[0069] In summary, the resonant de-icing method applies an external excitation frequency through the vibration module 7 based on the natural frequency of the transmission line to make the transmission line resonate, and enhances the vibration amplitude of the distribution network overhead line by adjusting the vibration frequency, so that the transmission line dances to perform anti-icing protection for the distribution network overhead line.

[0070] In addition, considering the power supply problem of the ice viewing module 6 and the vibration module 7, as shown in FIG. Fig. 9 As shown, in this embodiment, a power supply module 8 is installed on the overhead line of the distribution network with the help of the ice viewing module 6, and the ice viewing module 6 and the vibration module 7 are powered by the power supply module 8.

[0071] The power supply module 8 includes a current transformer, an energy storage element connected to the current transformer, and a rectifier circuit connected in parallel with the energy storage element; the current transformer is different from a general current transformer for measurement or protection. It is for reliable and stable transmission of secondary power during the primary current change time, and maximizes the secondary output power as much as possible. In order to prevent the primary current of the transmission line from fluctuating greatly and affecting the output power of the electrical equipment, the rectifier is connected in parallel with the energy storage element structure. When the primary transmission current is small, the energy storage element supplies power to the load. When the primary current is large and the secondary output power is large, the current transformer can supply power to the load and charge the energy storage element. In order to prevent the primary current from being too large, causing the secondary voltage U to be too high and damage the energy storage element and the load, a rectifier circuit is used to stabilize the output voltage within a reasonable range to ensure that the output voltage does not exceed the maximum operating voltage of the energy storage element and the load.

[0072] The working principle of current transformer is as follows Fig.16 As shown, according to electromagnetic induction, the current transformer is close to short circuit when working; according to electromagnetic theory, assuming the peak value of the output voltage is U, we can get: ; In the formula, is the frequency of the coil excitation current; N2 is the number of turns of the special coil; is the magnetic flux amplitude of the coil cross section; Bm is the magnetic induction intensity amplitude; S is the core cross-sectional area; λ is the core lamination coefficient.

[0073] According to Ampere's circuit law: ; In the formula, is the magnetic field intensity amplitude; is the number of turns of the primary winding; L is the average magnetic path length; I is the current in the primary winding; The relationship between the magnetic induction intensity amplitude Bm and the magnetic field intensity amplitude Hm is: ; In the formula, is the vacuum permeability; is the relative magnetic permeability.

[0074] The corresponding number of turns of the energy harvesting coil can be calculated, and a current transformer suitable for the de-icing device can be selected to perform induction power harvesting from the transmission wire to supply power to the ice viewing module 6 and the vibration module 7.

[0075] Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and describes these embodiments in detail in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A resonant deicing method based on the natural frequency of a power transmission line, characterized in that: The resonant deicing method is based on installing an ice observation module and a vibration module at intervals on the overhead line of the distribution network, and includes the following steps: Step 1: The ice observation module uses AI image recognition to detect ice coverage on the distribution network overhead lines; Step 2: The background management system starts the vibration sensor according to the icing detection result of the distribution network overhead line and collects the natural frequency data of the distribution network overhead line through the vibration sensor when it is confirmed that there is icing. Step 3: The background management system sends a start signal to the vibration module based on the natural frequency data of the distribution network overhead line collected by the vibration sensor; when the vibration module is started, the system parameters are fully initialized and enter a continuous operation state; Step 4, when the vibration module is working, its single chip microcomputer generates a PWM wave to control the vibration frequency of the vibration generating device, and at the same time, the accelerometer of the vibration module collects acceleration data and calculates it into amplitude data, which is used as feedback to control the vibration generating device to always operate at the maximum amplitude point; Step 5: The vibration module continues to work, and uses AI image recognition through the ice observation module to detect ice coverage on the distribution network overhead lines until the background management system recognizes that the ice coverage has been cleared through the ice observation module, and then controls the vibration module to stop; The resonant deicing method applies an external excitation frequency through a vibration module based on the natural frequency of the transmission line to cause the transmission line to resonate, and enhances the vibration amplitude of the distribution network overhead line by adjusting the vibration frequency, so that the transmission line dances to perform anti-icing protection for the distribution network overhead line.

2. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 1, characterized in that: In step 1, ice detection is performed, including: collecting distribution network overhead line images, external environment temperature and humidity data, and distribution network overhead line temperature data, and transmitting the distribution network overhead line images, external environment temperature and humidity data, and distribution network overhead line temperature data to the background management system.

3. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 2, characterized in that: In step 2, the background management system analyzes the distribution network overhead line image, the external environment temperature and humidity data, and the distribution network overhead line temperature data to obtain the distribution network overhead line icing detection result; The analysis process includes the following steps: Step 2.1, the background management system performs image enhancement based on the distribution network overhead line image through grayscale correction to obtain a pre-processed image; Step 2.2, the background management system uses wavelet transform or morphological method to detect the edge of the image; Step 2.3, the background management system performs boundary tracking based on the angle bisector fitting method to obtain the image edge extraction result; uses pixel-level fusion to fuse the image; reads the bare wire image data and compares it with the image fusion data, and outputs the distribution network overhead line ice coverage determination result.

4. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 3, characterized in that: In step 2.1, grayscale correction uses histogram equalization, which adjusts the grayscale value of the pixel based on the principle of probability theory to make the grayscale distribution of the entire image more uniform; If the number of gray levels of a grayscale image is n, assuming that the probability of a pixel with gray level i appearing is , the entropy corresponding to the i-level grayscale in the image is: ； In the formula, is the entropy corresponding to the i-level grayscale in the image; is the probability of occurrence of a pixel with gray level i.

5. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 1, characterized in that: In step 4, based on the synergistic effect of the three algorithms of Kalman filtering, Fourier transform and time domain integration, the acceleration data collected by the accelerometer is deeply processed to calculate the amplitude data, which specifically includes the following steps: Step 4.1, the ice and snow on the overhead line of the overhead distribution network generate forced vibration under the action of the uninterrupted periodic external force of the vibration generating device, and the periodic external force generated by the rotation of the vibration generating device is the driving force; based on the driving force, the motion equation of the forced vibration of the overhead line of the distribution network is obtained; based on the equal amplitude motion, the position of the overhead line of the distribution network is calculated; according to the theoretical calculation, the amplitude data is obtained; Step 4.2, based on the amplitude data obtained in step 4.1, perform Kalman filter calculation. The Kalman filter will generate an estimate of the unknown variable based on the values ​​measured at different times, considering the joint distribution at each time; Step 4.3, convert the time domain signal into frequency domain representation through Fourier transform; Step 4.4, convert the acceleration vibration signal into velocity or displacement signal through time domain integration.

6. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 5, characterized in that: In step 4.1, the driving force F is set to: ； In the formula, is the driving force amplitude, is the angular frequency of the driving force; t is the time; Under the driving force, the equation of motion is: ； make , , we get: ； In the formula, is the mass of the object; represents the position of the object; t is the time; is the elastic constant; is the magnitude of the driving force; is the angular frequency of the driving force; is the natural frequency; β is the damping coefficient; Based on the constant amplitude motion, we get: ； In the formula, Represents the location of an object; is the amplitude; is the angular frequency of the driving force; t is the time; is the initial phase, indicating the position phase at the beginning of the movement; According to theoretical calculation, the amplitude is: ； In the formula, is the magnitude of the driving force; is the mass of the object; is the natural frequency; is the angular frequency of the driving force; β is the damping coefficient.

7. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 5, characterized in that: In step 4.2, The following steps are involved: Predicted state estimation: ； In the formula, is the predicted estimate of the state at time k; for The optimal estimate of the state when is the state transfer matrix; is the control matrix; It is the external control input; Forecast error covariance: ； In the formula, is the prediction error covariance matrix of k; represents the state transfer matrix, which is used to describe the change of state from time k-1 to time k; The optimal estimated covariance matrix for the state error when is k-1; is the covariance matrix of process noise; is the state transfer matrix The transpose of represents the impact on error propagation; Update the Kalman gain: ； In the formula, represents the Kalman gain; is the prediction error covariance matrix of k; is the observation matrix; is the observation matrix The transpose of is the covariance matrix of the observation noise; Update the state estimate: ； In the formula, represents the updated state estimate at time k; is the predicted estimate of the state at time k; represents the Kalman gain; is the observed value at time k; is the observation matrix; Update the error covariance: ； In the formula, is the state error covariance matrix after update at time k; is the identity matrix; represents the Kalman gain; is the observation matrix; is the prediction error covariance matrix of k; The optimal estimation of the system state is performed by predicting the state estimate, predicting the error covariance, updating the Kalman gain, updating the state estimate and updating the error covariance.

8. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 5, characterized in that: In step 4.3, the N-point discrete Fourier transform of the N-point sequence x(r) can be expressed as: ； In the formula, is the frequency domain signal after discrete Fourier transform DFT, which represents the complex amplitude and phase information corresponding to frequency k; represents the time domain signal, which represents the sample value at time r; represents a complex exponential function, which is used to convert the time domain signal into the frequency domain to form a basis function of frequency k; j is an imaginary unit; N is the total number of sampling points of the signal; k is the index of the frequency domain signal, ranging from 0 to N-1; r is the index of the time domain signal, ranging from 0 to N-1.

9. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 5, characterized in that: In step 4.4, the time domain integration algorithm is as follows: ； In the formula, The integration result at time k represents the integral value calculated from the time domain signal x; is the time step, which indicates the time interval between two sampling points; is the sampling value of the time domain signal, the signal value at time i; i is the index of the integration process, representing each sampling point from 1 to k; k represents the current moment index, integrating from 1 to N.

10. The method for resonant deicing based on the natural frequency of a power transmission line according to claim 1, characterized in that: The power supply module installed on the overhead line of the distribution network is also installed to supply power to the ice observation module and the vibration module; the power supply module includes a current transformer, an energy storage element connected to the current transformer, and a rectifier circuit connected in parallel with the energy storage element; the rectifier circuit includes a clamping circuit, a rectifier filter circuit, and a DC / DC module located on the connection line between the current transformer and the energy storage element; the energy storage element adopts a lithium iron phosphate battery and a super capacitor; When the primary transmission current is small, the energy storage element serves as supplementary energy storage to supply power to the ice viewing module and the vibration module; when the primary current is large and the secondary output power is large, the current transformer directly supplies power to the ice viewing module and the vibration module, and stores the excess electric energy in the energy storage element; According to electromagnetic theory, assuming the peak value of the output voltage is U, we get: ； In the formula, is the frequency of the coil excitation current; N2 is the number of turns of the special coil; is the magnetic flux amplitude of the coil cross section; Bm is the magnetic induction intensity amplitude; S is the core cross-sectional area; λ is the core lamination coefficient; According to Ampere's circuit law: ； In the formula, is the magnitude of magnetic field intensity; is the number of turns of the primary winding; L is the average magnetic path length; I is the current in the primary winding; The relationship between the magnetic induction intensity amplitude Bm and the magnetic field intensity amplitude Hm is: ； In the formula, is the vacuum permeability; is the relative magnetic permeability; The number of turns of the corresponding energy harvesting coil can be calculated, and a current transformer suitable for the de-icing device can be selected to inductively draw power from the transmission line to supply power to the ice observation module and the vibration module.

Images