An AI image-based ice thickness monitoring method

Abstract

The application discloses an ice thickness monitoring method based on AI images, which improves monitoring accuracy and saves labor cost. The application comprises an AI image analysis mechanism, a weighing method analysis mechanism, a temperature analysis mechanism, a comprehensive analysis mechanism and a fault alarm mechanism. The comprehensive analysis mechanism comprises an AI image module, a weighing module, an environment module and a fault verification module for receiving result information analyzed by the AI image analysis mechanism, the weighing method analysis mechanism and the temperature analysis mechanism respectively. The weighing method data is judged in priority, and then it is determined whether the calculation result of the weighing method analysis mechanism is used. Then, the application function of the AI image module is selected according to the result of the weighing module. The fault verification module is used for verifying the validity of the calculation result again, and the effectiveness of each module is judged from the perspective of fault information. When one of the modules is invalid, the analysis result of the module is vetoed. The application is applied to the technical field of power transmission automation.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power transmission automation, and particularly relates to an ice thickness monitoring method based on AI images. BACKGROUND

[0002] China has invested a lot in the research of icing devices, especially in cold regions such as the northeast and northwest. The research focuses on icing process simulation, ice load testing, ice-resistant design, and deicing technology. Domestic has developed a variety of simulated icing devices and applied them to the design of power transmission lines, material selection, structural optimization, and the development of deicing technology. Foreign countries, especially Canada, the United States, and some European countries, are at the international advanced level in the research of power transmission line simulation icing devices, with mature technology and wide application. Their research direction focuses more on interdisciplinary cooperation, such as the combination of meteorology, fluid mechanics, materials science, and electrical engineering. Through comprehensive methods, they improve the accuracy and practicality of icing simulation and make many innovations in deicing technology, material research, and monitoring technology. For example, they have developed more efficient electric heating deicing technology and new ice-resistant coatings.

[0003] With the development of artificial intelligence and automation technology, the demand for icing devices will be more intelligent and automated, and it will be able to automatically adjust experimental conditions, accurately simulate the icing process under different weather conditions, and integrate icing monitoring and early warning systems. If a method for monitoring the thickness of ice based on AI images can be developed, it will be able to monitor the icing conditions of power transmission lines in real time and provide early warnings to reduce the impact of ice damage and meet the demand for intelligentization. SUMMARY

[0004] The present application solves the defects or deficiencies of the prior art, and provides an ice thickness monitoring method based on AI images that improves monitoring accuracy and saves labor costs.

[0005] The technical solution adopted by the present application is: the monitoring system used by the method includes a camera assembly, a micro-meteorological sensor, a tension sensor, a simulated conductor, and an icing detection structure. The camera assembly is installed on the icing detection structure. The simulated conductor is located below the icing detection structure, and both ends of the simulated conductor are fixed to both ends of the icing detection structure through the tension sensor. The micro-meteorological sensor is installed on the upper end of the camera assembly. The method includes an AI image analysis mechanism, a weighing method analysis mechanism, a temperature analysis mechanism, a comprehensive analysis mechanism, and a fault alarm mechanism.

[0006] The AI image analysis mechanism comprises the following steps: a1. acquiring an analog conductor image through a camera assembly, and judging the validity of the analog conductor image data through a fault alarm mechanism; a2. if the analog conductor image is invalid, output an abnormal marker code; a3. if the analog conductor image is valid, obtain a mask picture of the analog conductor through a yolov8-seg segmentation algorithm, and calculate the height of the pixels in the mask picture to obtain the ice thickness of the analog conductor;

[0007] The weighing method analysis mechanism comprises the following steps: b1. acquiring the tension of the analog conductor through a tension sensor, acquiring the wind speed and direction through a micro-meteorological sensor, and judging the validity of the tension, wind speed and direction data through a fault alarm mechanism; b2. if the tension value is invalid, the weighing method analysis result is determined to be invalid, the current calculation is stopped, and a fault description and invalid state code are returned; b3. when the tension value is valid, the validity of the wind speed and direction data is first judged, when the wind speed and direction are invalid, the weighing method analysis mechanism will call a preset value to replace the calculation result of the wind speed and direction for subsequent calculation, and when the wind speed and direction data is valid, the wind pressure on the analog conductor is calculated, the force direction of the calculated wind pressure is in the direction of gravity, and the horizontal direction pressure is not affected by the vertical gravity center, so the ice weight of the analog conductor in the ice state can be obtained, and the ice thickness of the analog conductor can be analyzed through the ice weight;

[0008] The temperature analysis mechanism comprises the following steps: c1. acquiring temperature data in a set time interval through a micro-meteorological sensor, and judging the validity of the temperature data through a fault alarm mechanism; c2. if the temperature data is invalid, the temperature analysis result is determined to be invalid, the current calculation is stopped, and a fault description and invalid state code are returned; c3. when the temperature data is valid, the temperature value is calculated by using a dynamic average value, the initial condition is opened, the initial ice temperature condition threshold is set, if the calculated temperature value is less than the ice threshold, the temperature analysis mechanism judges that the ice condition is met, otherwise it is judged that the ice condition is not met;

[0009] The comprehensive analysis mechanism includes an AI image module, a weighing module, an environment module, and a fault verification module. The AI image module, the weighing module, and the environment module are used to receive the result information analyzed by the AI image analysis mechanism, the weighing method analysis mechanism, and the temperature analysis mechanism, respectively. The comprehensive analysis mechanism includes the following steps: d1. The information received by the three modules is analyzed and calculated comprehensively; d2. The effectiveness of the weighing method analysis mechanism is detected. When the weighing method analysis mechanism is effective, the ice thickness analyzed by the weighing method analysis mechanism is given to the comprehensive analysis of the ice thickness, and then the AI image module is called to determine whether the simulated conductor is covered with ice; d3. If there is no ice, the comprehensive analysis of the ice thickness is assigned a value of 0. Next, if the AI image analysis is covered with ice, it is determined whether the weighing module is effective. If not, the thickness value analyzed by the AI image analysis is assigned to the comprehensive analysis of the ice thickness; d4. The environment module is called to determine whether the temperature is effective. If the temperature analysis result is effective, the correctness of the ice state is verified through the analysis result of the temperature. Finally, the fault verification module is called to verify the effectiveness of the analysis result from the weighing method mechanism, the AI image analysis mechanism, and the temperature analysis mechanism in sequence, and the current ice thickness and state are output.

[0010] Further, the camera assembly includes a pan-tilt camera, a night light camera, and a daylight camera. The monitoring system further includes a battery system and a solar panel. The battery system provides power for the monitoring system and is externally connected to the solar panel. The tension sensor, the micro-weather sensor, the tension sensor, and the battery system are electrically connected to the camera assembly. The pan-tilt camera is provided with a controller and is connected to the back-end platform in data communication through a wireless transceiver module.

[0011] Further, the fault alarm mechanism in step a1 is used to determine whether a picture can be taken and whether the simulated conductor in the picture can be recognized. The ice thickness of the simulated conductor in step a3 is calculated by the ice calculation formula: k = (L-l)*d / 2l, where k is the ice thickness of the simulated conductor calculated by the AI image analysis mechanism, L is the average pixel height after icing, l is the average pixel height of the simulated conductor without icing, and d is the diameter of the simulated conductor.

[0012] Further, in step b3, when the wind speed and direction data are effective, the formula:

[0013] F_P = W_O * S * COS(θ), wherein: F_P: wind pressure force in the direction of gravity, W_O: wind pressure, S: cross-sectional area of the simulation conductor on which the wind pressure acts, and θ: angle between the wind pressure and the direction of gravity of the simulation conductor; the wind pressure acting on the simulation conductor is calculated, and the direction of the calculated wind pressure is in the direction of gravity; the pressure in the horizontal direction is less affected by the vertical gravity center, so the ice weight M_P of the simulation conductor in the icing state is F-F_P-F_N, wherein: M_P is the ice weight of the simulation conductor, F is the value of the tension sensor, F_P is the wind pressure, and F_N is the initial tension value of the tension sensor in the non-icing state; after the ice weight is calculated, the formula:

[0014]

[0015] wherein: d is the diameter of the simulation conductor, L_P is the horizontal span of the simulation conductor, M_P is the ice weight of the simulation conductor, B_P is the ice thickness analyzed by the weighing analysis mechanism, and f is the density of ice; the ice thickness value and the icing state analyzed by the weighing analysis are calculated, when the ice thickness = 0, the simulation conductor is not iced, and when the ice thickness > 0, the simulation conductor is in the icing state; after the weighing analysis mechanism is completed, the ice thickness analyzed by the weighing analysis and the icing state analyzed by the weighing analysis and the fault information state code are output.

[0016] Further, the logical steps of the fault verification module in step d4 are: judging whether the data of the two tension sensors is missing or out of range, if both are normal, calculating the ice thickness by the comprehensive analysis mechanism, if there is missing or out of range, further judging whether the image can be shot and whether the simulation conductor can be recognized from the shot image, if the image can be shot and the simulation conductor can be recognized, outputting the ice thickness calculated by the AI image segmentation algorithm; if the image cannot be shot or the simulation conductor cannot be recognized, outputting the value of the last ice thickness.

[0017] Further, the fault verification module in step d4 is further provided with a data mutation prevention logic, and the steps are: judging whether the difference between the current simulation conductor ice thickness and the last obtained simulation conductor ice thickness is greater than 2 mm, if greater than 2 mm, changing the current simulation conductor ice thickness to the value of the last obtained simulation conductor ice thickness plus 2 mm.

[0018] The beneficial effects of the present application are: 1. The AI image-based ice thickness monitoring method of the present application uses a camera assembly to collect multi-scene image information, combines micro-meteorological sensor data and tension sensor data for comprehensive analysis to obtain the ice state and ice results of the simulated conductor, solves the problems of false reporting, missing reporting, insufficient accuracy and inability to self-check and verify of the existing single sensor affected by the environment, and the monitoring data can not only be used for real-time early warning, but also provide theoretical data guidance for subsequent engineering planning and schemes; 2. Remote monitoring and remote operation can be realized, reducing the dependence on personnel, reducing the damage and maintenance frequency of the power transmission line caused by ice and snow weather, reducing maintenance costs, and providing important technical support for the design, maintenance and ice prevention of the power transmission line, which helps to improve the stability and reliability of the power system. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a schematic diagram of a monitoring system;

[0020] Figure 2 is a schematic diagram of the present application. DETAILED DESCRIPTION

[0021] As shown in Figure 1 and Figure 2 , in the present embodiment, an AI image-based ice thickness monitoring method, the monitoring system used by the method includes a camera assembly 1, a micro-meteorological sensor 2, a tension sensor 3, a simulated conductor 4 and an ice detection structure 5, the ice detection structure 5 is a test rack, the camera assembly 1 is installed on the middle part of the ice detection structure 5, the simulated conductor 4 is located below the ice detection structure 5, and both ends of the simulated conductor 4 are fixed to both ends of the ice detection structure 5 through the tension sensor 3, the tension sensor 3 collects the gravity data on the simulated conductor 4, the micro-meteorological sensor 2 is installed on the upper end of the camera assembly 1, and is used to collect real-time air temperature, relative humidity, wind speed, wind direction and atmospheric pressure data, the camera assembly 1 includes a pan-tilt camera, a night light camera and a daylight camera, the monitoring system further includes a battery system 10 and a solar panel 11, the battery system 10 provides power for the monitoring system, and is externally connected to the solar panel 11, both the tension sensor 3, the micro-meteorological sensor 2 and the battery system 10 are electrically connected to the camera assembly 1, a controller is arranged in the pan-tilt camera, and the pan-tilt camera is in data communication connection with a back-end platform 12 through a wireless transceiver module;

[0022] When the whole monitoring system is installed, the camera assembly 1 is powered on and the self-test starts successfully, and then the camera shooting point, image analysis processing parameters and original tension value of the tension sensor 3 are initialized. After the initialization is completed, the camera assembly 1 is automatically connected to the pre-set back-end platform 12. The camera assembly 1 periodically turns on the power supply of the micro-meteorological sensor 2. The micro-meteorological sensor 2 autonomously collects the ambient temperature, relative humidity, wind speed, wind direction and atmospheric pressure data of the surrounding environment of the sensor. After the micro-meteorological sensor 2 data collection is ready, the camera assembly 1 collects the micro-meteorological data through the RS485 communication mode. After the micro-meteorological data collection is completed, the power supply of the micro-meteorological sensor 2 is turned off. The power supply of the tension sensor 3 is turned on, and the tension sensor 3 autonomously collects the weight data of the analog conductor and transmits the data to the camera assembly 1 through the RS485 mode. After the camera assembly 1 collects the data, the power supply of the tension sensor 3 is turned off. After the sensor data collection is completed, the camera assembly 1 starts to collect image data according to the pre-set setting point. The image data includes analog conductor pictures and actual power transmission conductor pictures.

[0023] The method includes an AI image analysis mechanism 6, a weighing method analysis mechanism 7, a temperature analysis mechanism 8, a comprehensive analysis mechanism 9 and a fault alarm mechanism. After the data collection is completed, the camera assembly 1 transmits all the data into the algorithm library through the interface. The algorithm first calls the picture data to call the AI image analysis mechanism 6. The AI image analysis mechanism 6 includes the following steps: a1. obtaining the image of the analog conductor 4 through the camera assembly 1 and judging the validity of the image data of the analog conductor 4 through the fault alarm mechanism; a2. if the image of the analog conductor 4 is invalid, output an abnormal marker code; a3. if the image of the analog conductor 4 is valid, obtain the mask picture of the analog conductor 4 through the yolov8-seg segmentation algorithm, and calculate the height of the pixels in the mask picture to obtain the ice thickness of the analog conductor 4. The mask picture of the analog conductor 4 obtained through the yolov8-seg segmentation algorithm is sent to the neural network inference model for inference. The general process of inference is as follows: the convolutional neural network can extract features from the picture, classify and regress the extracted picture features, and identify the position of the analog conductor 4 in the picture and the distribution of the pixel points of the analog conductor 4 when it is covered with ice. According to the pixel point distribution, the pixel point height of the conductor when it is covered with ice can be calculated, that is, the ice thickness of the analog conductor 4 is calculated through the ice calculation formula: k = (L-l)*d / 2l, where k is the ice thickness of the analog conductor 4 calculated through the AI image analysis mechanism 6, L is the average pixel point height after icing, l is the average pixel point height of the analog conductor 4 without icing, and d is the diameter of the analog conductor 4.

[0024] The weighing method analysis mechanism 7 includes the following steps: b1. obtaining the tension of the analog conductor 4 by the tension sensor 3, obtaining the wind speed and direction by the microclimate sensor 2, and determining the validity of the tension, wind speed and direction data by the fault alarm mechanism; b2. if the tension value is invalid, determining that the weighing method analysis result is invalid, stopping the current calculation, and returning the fault description and invalid state code; b3. when the tension value is valid, first determine the validity of the wind speed and direction data, when the wind speed and direction are invalid, the weighing method analysis mechanism 7 will call a preset value to replace the calculation result of the wind speed and direction for subsequent calculation, and the preset value will be obtained by the long-term stable operation of the device wind speed and direction trend, when the wind speed and direction data is valid, calculate the wind pressure on the analog conductor 4, the force direction of the calculated wind pressure is in the direction of gravity, and the pressure in the horizontal direction will not be affected by the vertical gravity center, that is, the mechanical model is used to decompose the comprehensive force on the analog conductor 4 into vertical load in the gravity direction and horizontal load in the horizontal direction, then the force parallelogram rule is applied to remove the wind speed and direction data to calculate the wind load of the analog conductor 4, so the icing weight of the analog conductor 4 under the icing state can be obtained, and the icing thickness of the analog conductor 4 can be analyzed through the icing weight;

[0025] The temperature analysis mechanism 8 is applied to auxiliary verification, which is used to determine whether the current state meets the icing condition, and only when the condition is met, the icing calculation thickness value is allowed to be output, otherwise the previous calculation result is denied. The temperature analysis mechanism 8 includes the following steps: c1. obtaining the temperature data of the set time interval by the microclimate sensor 2, and determining the validity of the temperature data by the fault alarm mechanism; c2. if the temperature data is invalid, determining that the temperature analysis result is invalid, stopping the current calculation, and returning the fault description and invalid state code; c3. when the temperature data is valid, calculating the temperature value by using the dynamic average value, and opening the initial condition, the set initial icing temperature condition threshold, if the calculated temperature value is less than the icing threshold, the temperature analysis mechanism 8 determines that the icing condition is met, otherwise it is determined that the icing condition is not met;

[0026] The comprehensive analysis mechanism 9 includes an AI image module 91, a weighing module 92, an environment module 93, and a fault verification module 94. The AI image module 91, the weighing module 92, and the environment module 93 are respectively used to receive the result information analyzed by the AI image analysis mechanism 6, the weighing method analysis mechanism 7, and the temperature analysis mechanism 8. The validity of the weighing method data is determined first. Then, it is determined whether to use the calculation result of the weighing method analysis mechanism 7. Then, the application function of the AI image module 91 is selected according to the result of the weighing module 92. The comprehensive analysis mechanism 9 includes the following steps: d1. The information received by the three modules is started to be comprehensively analyzed and calculated; d2. The validity of the weighing method analysis mechanism 7 is detected. When the weighing method analysis mechanism 7 is valid, the ice thickness analyzed by the weighing method analysis mechanism 7 is given to the comprehensive analysis of the ice thickness, and then the AI image module 91 is called to determine whether the simulated conductor 4 exists ice; d3. If there is no ice, the comprehensive analysis of the ice thickness is assigned a value of 0. Then, if the AI image analysis exists ice, it is determined whether the weighing module 92 is valid. If not, the thickness value of the AI image analysis is assigned to the comprehensive analysis of the ice thickness; d4. The environment module 93 is called to determine whether the temperature is invalid. If the temperature analysis result is valid, the correctness of the ice state is verified through the analysis result of the temperature. Finally, the fault verification module 94 is called to verify the validity of the analysis result from the weighing method mechanism, the AI image analysis mechanism 6, and the temperature analysis mechanism 8 in turn, and the current ice thickness and state are output.

[0027] In this embodiment, the fault alarm mechanism in step a1 is used to determine whether a picture can be taken and whether the simulated conductor 4 in the picture can be recognized.

[0028] In this embodiment, when the wind speed and direction data are valid in step b3, the wind pressure on the simulated conductor 4 is calculated by the formula: F_P = W_O * S * COS(θ), where F_P is the wind pressure in the direction of gravity, W_O is the wind pressure, S is the cross-sectional area of the simulated conductor 4, and θ is the angle between the wind pressure and the direction of gravity of the simulated conductor 4. The direction of the calculated wind pressure is in the direction of gravity. The horizontal pressure will not be affected according to the vertical gravity center. Therefore, the ice weight M_P of the simulated conductor 4 in the ice state is F-F_P-F_N, where M_P is the ice weight of the simulated conductor 4, F is the value of the tension sensor 3, F_P is the wind pressure, and F_N is the initial tension value of the tension sensor 3 in the non-icing state. After the ice weight is calculated, the formula:

[0029]

[0030] Wherein: d: the diameter of the simulation conductor 4, L_P: the horizontal span of the simulation conductor 4, M_P: the ice weight of the simulation conductor 4, B_P: the ice thickness analyzed by the weighing analysis mechanism 7, f: the density of ice, the ice thickness value and the ice state analyzed by the weighing analysis are calculated, when the ice thickness = 0, and the device is not covered with ice, the ice thickness > 0, the device is in the ice state, after the weighing analysis mechanism 7 is completed, the ice thickness analyzed by the weighing analysis and the ice state analyzed by the weighing analysis, the fault information state code are output.

[0031] In the embodiment, the fault verification module 94 in step d4 is used to verify the validity of the calculation result again, and the validity of each module is judged from the perspective of fault information, and the analysis result of the module is invalidated when the module is invalid; the logical steps of the fault verification module 94 are: judging whether the data of the two tension sensors 3 is missing or out of range, if both are normal, the ice thickness calculated by the comprehensive analysis mechanism 9 is passed, if the data of the tension sensor 3 is missing or out of range, it is further judged whether the image can be shot and whether the simulation conductor 4 can be recognized from the shot image, if the image can be shot and the simulation conductor 4 can be recognized, the ice thickness calculated by the AI image segmentation algorithm is output; if the image cannot be shot or the simulation conductor 4 cannot be recognized from the image, the value of the last ice thickness is output.

[0032] In the embodiment, the fault verification module 94 in step d4 is further provided with a data mutation prevention logic, and the steps are: judging whether the difference between the ice thickness of the simulation conductor 4 this time and the ice thickness of the simulation conductor 4 obtained last time is greater than 2mm, if greater than 2mm, the ice thickness of the simulation conductor 4 this time is changed to the value of the ice thickness of the simulation conductor 4 obtained last time plus 2mm.

[0033] Although the embodiments of the present application are described in actual schemes, but do not constitute a limitation on the meaning of the present application, for those skilled in the art, according to the modification of the embodiments and the combination with other schemes according to the present application are obvious.

Claims

1. An AI image-based ice thickness monitoring method, a monitoring system used in the method comprising a camera assembly (1), a micro-meteorological sensor (2), a tension sensor (3), a simulation conductor (4) and an ice detection structure (5), the camera assembly (1) being installed on the ice detection structure (5), the simulation conductor (4) being located below the ice detection structure (5), and both ends of the simulation conductor (4) being fixed to both ends of the ice detection structure (5) through the tension sensor (3), the micro-meteorological sensor (2) being installed on the upper end of the camera assembly (1); characterized in that, The method comprises an AI image analysis mechanism (6), a weighing method analysis mechanism (7), a temperature analysis mechanism (8), a comprehensive analysis mechanism (9) and a fault alarm mechanism; The AI image analysis mechanism (6) comprises the following steps: a1. acquiring an analog conductor (4) image through a camera assembly (1), and judging the validity of the analog conductor (4) image data through the fault alarm mechanism; a2. if the analog conductor (4) image is invalid, output an abnormal marker code; a3. if the analog conductor (4) image is valid, obtain a mask picture of the analog conductor (4) through a yolov8-seg segmentation algorithm, and calculate the height of the pixels in the mask picture to obtain the ice thickness of the analog conductor (4); The weighing method analysis mechanism (7) comprises the following steps: b1. acquiring the tension of the analog conductor (4) through a tension sensor (3), acquiring the wind speed and direction through a micro-meteorological sensor (2), and judging the validity of the tension, wind speed and direction data through the fault alarm mechanism; b2. if the tension value is invalid, determining that the weighing method analysis result is invalid this time, stopping the current calculation, and returning a fault description and an invalid state code; b3. when the tension value is valid, first judging the validity of the wind speed and direction data, when the wind speed and direction are invalid, the weighing method analysis mechanism (7) will call a preset value to replace the calculation result of the wind speed and direction for subsequent calculation, and when the wind speed and direction data are valid, calculating the wind pressure on the analog conductor (4), the force direction of the calculated wind pressure being in the direction of gravity, and the horizontal direction pressure being not affected by the vertical gravity center, so that the ice weight of the analog conductor (4) in the ice state can be obtained, and the ice thickness of the analog conductor (4) can be analyzed through the ice weight; The temperature analysis mechanism (8) comprises the following steps: c1. acquiring temperature data in a set time interval through a micro-meteorological sensor (2), and judging the validity of the temperature data through the fault alarm mechanism; c2. if the temperature data is invalid, determining that the temperature analysis result is invalid this time, stopping the current calculation, and returning a fault description and an invalid state code; c3. when the temperature data is valid, calculating the temperature value in a dynamic average manner, and opening the set initial ice temperature condition threshold under the initial condition, if the calculated temperature value is less than the ice threshold, the temperature analysis mechanism (8) judges that the ice condition is met, otherwise it is judged that the ice condition is not met. The comprehensive analysis mechanism (9) comprises an AI image module (91), a weighing module (92), an environment module (93), and a fault verification module (94). The AI image module (91), the weighing module (92), and the environment module (93) are respectively used for receiving the result information analyzed by the AI image analysis mechanism (6), the weighing method analysis mechanism (7), and the temperature analysis mechanism (8). The comprehensive analysis mechanism (9) comprises the following steps: d1. starting the comprehensive analysis calculation of the information received by the three modules; d2. detecting the effectiveness of the weighing method analysis mechanism (7). When the weighing method analysis mechanism (7) is effective, the ice thickness analyzed by the weighing method analysis mechanism (7) is given to the comprehensive analysis ice thickness, and then the AI image module (91) is called to judge whether the simulated conductor (4) exists ice; d3. if there is no ice, the comprehensive analysis ice thickness is assigned a value of 0. Next, if the AI image analysis exists ice, it is judged again whether the weighing module (92) is effective. If not, the thickness value of the AI image analysis is assigned to the comprehensive analysis ice thickness; d4. calling the environment module (93) for judgment. If the temperature is invalid, the judgment of the environment module (93) is skipped. If the temperature analysis result is valid, the correctness of the ice state is verified through the analysis result of the temperature. Finally, the fault verification module (94) is called to verify the effectiveness of the analysis result from the weighing method mechanism, the AI image analysis mechanism (6), and the temperature analysis mechanism (8) in turn, and the current ice thickness and state are output.

2. The AI image-based ice thickness monitoring method of claim 1, wherein: The camera assembly (1) comprises a pan-tilt camera, a night light camera, and a daylight camera. The monitoring system further comprises a battery system (10) and a solar panel (11). The battery system (10) provides power for the monitoring system and is externally connected to the solar panel (11). The tension sensor (3), the micro-weather sensor (2), the tension sensor (3), and the battery system (10) are electrically connected to the camera assembly (1). The pan-tilt camera is provided with a controller and is in data communication connection with the back-end platform (12) through a wireless transceiver module.

3. The AI image-based ice thickness monitoring method of claim 2, wherein: The fault alarm mechanism in step a1 is used to judge whether the picture can be taken and whether the simulated conductor (4) in the picture can be recognized; the ice thickness of the simulated conductor (4) in step a3 is calculated by the ice calculation formula: k = (L-l)*d / 2l, wherein k is the ice thickness of the simulated conductor (4) calculated by the AI image analysis mechanism (6), L is the average pixel height after icing, l is the average pixel height of the simulated conductor (4) without icing, and d is the diameter of the simulated conductor (4).

4. The AI image-based ice thickness monitoring method of claim 3, wherein: The wind speed and direction data in step b3 is valid, and the wind pressure on the simulated conductor (4) is calculated by the formula: F_P = W_O * S * COS (θ), wherein: F_P: wind pressure in the direction of gravity, W_O: wind pressure, S: cross-sectional area of the simulated conductor (4) under wind pressure, θ: angle between wind pressure and the direction of gravity of the simulated conductor (4); the force direction of the calculated wind pressure is in the direction of gravity, and the horizontal pressure will not be affected according to the vertical gravity center, so the ice weight M_P of the conductor under icing condition is F-F_P-F_N, wherein: M_P is the ice weight of the simulated conductor (4), F is the value of the tension sensor (3), F_P is the wind pressure, and F_N is the initial tension value of the tension sensor (3) under the non-icing state; after calculating the ice weight, the formula: wherein: d: diameter of the simulated conductor (4), L_P: horizontal span of the simulated conductor (4), M_P: ice weight of the simulated conductor (4), B_P: ice thickness analyzed by the weighing analysis mechanism (7), f: density of ice, the ice thickness value and the icing state analyzed by the weighing analysis are calculated, when the ice thickness = 0, the simulated conductor (4) is not iced, and when the ice thickness > 0, the simulated conductor (4) is in the icing state; after the weighing analysis mechanism (7) is completed, the ice thickness analyzed by the weighing analysis and the icing state analyzed by the weighing analysis, and the fault information state code are output.

5. The AI image-based ice thickness monitoring method of claim 4, wherein: The logic steps of the fault verification module (94) in step d4 are: judging whether the data of the two tension sensors (3) is missing or out of range, if both are normal, the ice thickness calculated by the comprehensive analysis mechanism (9) is output, if there is missing or out of range, further judging whether the image can be shot and whether the simulated conductor (4) can be recognized in the shot image, if the image can be shot and the simulated conductor (4) can be recognized, the ice thickness calculated by the AI image segmentation algorithm is output; if the image cannot be shot or the simulated conductor (4) cannot be recognized in the image, the last ice thickness value is output.

6. The AI image-based ice thickness monitoring method of claim 5, wherein: After the fault verification module (94) in step d4, a data mutation prevention logic is further set, and the steps are: judging whether the difference between the current ice thickness of the simulated conductor (4) and the last obtained ice thickness of the simulated conductor (4) is greater than 2mm, if greater than 2mm, the current ice thickness of the simulated conductor (4) is changed to the last obtained ice thickness of the simulated conductor (4) plus 2mm.

Images