A method and system for identifying icing types of a mountainous power transmission line

Abstract

The application discloses a mountainous power transmission line icing type identification method and system, and aims to solve the problems of limited deployment quantity of existing icing online monitoring devices and difficulty in obtaining effective data under extreme conditions. The application screens a mountainous tower point with an icing image as a sample tower point, constructs a first characteristic index (air temperature) and a second characteristic index (cloud particle) by interpolating meteorological reanalysis data to the sample tower point and icing observation data of the tower point, interpolates the meteorological reanalysis data to a target tower point, screens icing times according to the first characteristic index, identifies an icing type according to the second characteristic index, and forms an hourly icing type sequence of the target tower point. The application realizes efficient calculation by using decoupled meteorological indexes, breaks through the limitation of mountainous terrain, and provides a reliable reference for power grid icing disaster prevention.

Description

Technical Field

[0001] This invention relates to the field of power transmission line icing identification, and more specifically, to a method and system for identifying icing types of power transmission lines in mountainous areas. Background Technology

[0002] Online icing monitoring data provides comprehensive meteorological information and icing conditions for power pole locations, offering crucial support for power grid icing disaster prevention. However, complete icing data is still unavailable for many pole locations. This indicates that meteorological and icing monitoring during severe icing periods remains a significant challenge. Furthermore, existing monitoring data reveals that even closely spaced poles located on different mountainsides may exhibit completely different icing patterns—rime and hoarfrost—demonstrating that icing observation data from a single pole cannot be easily extrapolated to neighboring pole locations, making it impossible to grasp the overall icing situation in complex mountainous areas from a limited number of observation points.

[0003] Transmission line icing is essentially formed by the convergence of warm and cold air currents, resulting in precipitation-like water products (freezing rain and wet snow) and non-precipitation-like water products (cloud fog) on ​​transmission lines at temperatures below 0°C. Many scholars have used numerical weather models to reproduce the distribution and transformation processes of primary meteorological features and microscopic particles during icing, effectively simulating the growth process of line icing caused by freezing rain, wet snow, and cloud fog. However, numerical weather model simulations require significant computational resources, and currently, simulations of single icing events in a specific region are more common. Global meteorological reanalysis data assimilates data from all radiosonde stations and conventional meteorological stations in China, including temperature, specific humidity, relative humidity, geopotential height, and microphysical quantities such as cloud water and cloud snow water across nearly 40 pressure levels. Although the horizontal resolution is slightly coarse (around 28 miles), it has certain applicability under winter weather systems dominated by stratiform cloud precipitation. Evaluation shows that the ERA5 reanalysis data has high reliability in densely populated areas of eastern China and is suitable for identifying icing types in the hilly regions of eastern China. This invention uses reanalysis data to analyze the patterns of atmospheric macro-meteorological characteristics corresponding to different icing types on mountain routes, summarizes the causal mechanisms of different icing types from a meteorological perspective, and combines it with ground mountain topographic data to identify whether there is icing and the type of icing at any location within the study area at a certain time. Summary of the Invention

[0004] This invention overcomes the shortcomings of existing online icing monitoring devices, which cannot cover all tower points and may even struggle to obtain usable data in extreme cases, making it difficult to obtain icing information for each tower point. It provides a method and system for identifying icing types of transmission lines in mountainous areas. Based on meteorological reanalysis data, it can analyze and identify whether mountain lines are iced and the type of icing, thereby providing a reference for power grid maintenance agencies regarding icing disasters.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for identifying icing types on power transmission lines in mountainous areas includes: Meteorological reanalysis data and icing observation data of pole towers in mountainous areas are obtained. Pole towers with icing images are selected as sample pole towers. Meteorological reanalysis data are constructed for the sample pole towers using interpolation methods. Based on the icing observation data, corresponding meteorological characteristic indicators are constructed. The meteorological characteristic indicators include the first characteristic indicator affecting icing and the second characteristic indicator affecting icing type. Select any pole point in the mountainous area as the target pole point, interpolate the target pole point based on meteorological reanalysis data, and perform hourly calculations using the first characteristic index to screen out all times when the target pole point is covered with ice. Based on the second characteristic index, the icing type is identified for all occurrences of icing, forming an hourly icing type sequence for the target tower point during an icing process.

[0006] As a preferred option, the process of constructing meteorological characteristic indicators is as follows: The meteorological characteristics closely related to icing at the sample tower points were analyzed. The first characteristic indicator included temperature characteristics, and the second characteristic indicator included cloud particle characteristics. Based on the actual icing types of the sample tower points, the first and second characteristic indices of various icing types of transmission lines in mountainous areas are calculated.

[0007] As a preferred method, the icing observation data is obtained as follows: Select an icing process with a large icing area and a long duration. Use the image recognition method of the power grid icing online monitoring system to screen the icing tower points. The icing types of the icing tower points should include at least rime ice and mixed rime ice, and the number of icing tower points should be greater than or equal to 10.

[0008] As a preferred method, the process of interpolating the sample tower points based on meteorological reanalysis data is as follows: Based on the region and time interval of icing occurrence at the power grid icing monitoring system, the corresponding time and range of global climate reanalysis grid data are downloaded, retaining meteorological element data at all levels below 300 hPa, and then interpolating to the sample power grid points using interpolation methods, including bilinear interpolation.

[0009] As a preferred option, the process for the first characteristic index is as follows: Based on the typical distribution characteristics of air temperature and specific humidity during icing, we analyze whether the vertical distribution of air temperature at the sample tower points meets the "cold-warm-cold" stratification characteristics and the air temperature at the center of the middle warm layer is greater than the threshold T1. When the threshold condition is met, we record the lowest altitude of the bottom cold layer as H1, thus forming the air temperature index for judging icing. Based on the actual icing conditions of the sample tower points, the value of T1 is determined as the first feature index for identifying the icing type.

[0010] As a preferred method, the process for calculating the second characteristic index is as follows: The icing type ICE1 formed by freezing rain corresponds to cloud and rainwater exceeding the threshold R1 at the target tower height; The icing type ICE2 formed by fog in the clouds corresponds to the presence of cloud liquid water at the target tower height, and the threshold of cloud liquid water is greater than L1; The icing type ICE3 formed by wet snow corresponds to the presence of snowmelt and rainwater at the target tower height. The threshold for rainwater is greater than S1 and the threshold for rainwater is in the range [R2, R1). Based on the actual icing type of the sample, the values ​​of each threshold are determined as the second feature index.

[0011] As a preferred method, the process of interpolating the target pole point based on meteorological reanalysis data is as follows: based on the location of the target pole point and the expected time interval, download its global climate reanalysis grid data, the altitude of the target pole point is H2, and use the interpolation method to assign meteorological characteristic indicators to the target pole point.

[0012] As a preferred method, the process of selecting all times during the period when icing occurs at the target tower point is as follows: Based on the first characteristic index, calculate whether the target tower point meets the first characteristic index of temperature and specific humidity on an hourly basis. For times that meet the index, calculate H2-H1. If it is greater than 0, then the time is assigned as a possible time of icing and recorded as 1; otherwise, it is recorded as 0.

[0013] As a preferred method, the process of forming an hourly icing type sequence for a target tower point during an icing process is as follows: based on the second characteristic index, only the time when the target tower point is marked as a possible icing time is judged, and the time when icing occurs is represented by a specific icing type, while the time when no icing occurs is represented by ICE0, and an hourly icing type sequence is formed according to the time.

[0014] A system for identifying icing types on power transmission lines in mountainous areas includes: The indicator construction module is used to acquire meteorological reanalysis data and icing observation data of pole towers in mountainous areas, select pole towers with icing images as sample pole towers, construct meteorological reanalysis data for sample pole towers through interpolation, and construct corresponding meteorological characteristic indicators based on icing observation data. The meteorological characteristic indicators include the first characteristic indicator affecting icing and the second characteristic indicator affecting icing type. The icing time filtering module is used to select any pole point in the mountainous area as the target pole point, interpolate the target pole point based on meteorological reanalysis data, and perform hourly calculations using the first feature index to filter out all times when the target pole point is iced. The icing type identification module identifies the icing type for all occurrences of icing based on the second feature index, forming an hourly icing type sequence for the target tower point during an icing process.

[0015] Compared with the prior art, the beneficial effects of the present invention are: (1) Use the first characteristic indicator to determine whether icing has occurred, and use the second characteristic indicator to identify the icing type at all times when icing occurs. Through decoupled meteorological indicators and by setting simple and efficient logical judgments, efficient calculation can be achieved. (2) By interpolating the weather reanalysis data, meteorological reanalysis data of each tower point were obtained, which provided basic data for subsequent icing type identification, thus breaking through the problem that two towers on different small mountains in the mountains may have different weather even if they are close to each other. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the present invention; Figure 2 This is an hourly sequence diagram of icing types at towers 1 to 5 during an icing event, according to the present invention. Figure 3 This is the hourly icing type sequence calculated for tower point No. 4 in December 2023 according to the present invention; Figure 4 This is the hourly icing type sequence calculated for tower point No. 4 in February 2024 according to the present invention; Figure 5 This is a curve showing the standard ice thickness variation based on the tension inversion at tower point No. 4 in December 2024. Figure 6 This is a system schematic diagram of the present invention. Detailed Implementation

[0017] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0018] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0019] Example: To achieve the identification of icing types on power transmission lines in hilly and mountainous areas, such as Figure 1 As shown, a method for identifying icing types on power transmission lines in mountainous areas is provided, including: Meteorological reanalysis data and icing observation data of pole towers in mountainous areas are obtained. Pole towers with icing images are selected as sample pole towers. Weather data is constructed for the sample pole towers using interpolation methods. Based on the icing observation data, corresponding meteorological characteristic indicators are constructed. The meteorological characteristic indicators include the first characteristic indicator affecting icing and the second characteristic indicator affecting icing type. Select any pole point in the mountainous area as the target pole point, interpolate the target pole point based on meteorological reanalysis data, and perform hourly calculations using the first characteristic index to screen out all times when the target pole point is covered with ice. Based on the second characteristic index, the icing type is identified for all occurrences of icing, forming an hourly icing type sequence for the target tower point during an icing process.

[0020] To obtain the first and second characteristic indicators, a hilly area with a large number of mountain pole points needs to be selected, where at least a few mountain pole points are equipped with online icing monitoring devices. This ensures that a sufficient number of pole points collect long-term icing image data during a given icing event, thereby increasing the sample size. Each online icing monitoring device communicates with a server host, collects these icing image data, and identifies the icing type. The online icing monitoring devices and the server host constitute an online icing monitoring system. In this embodiment, an area with 100 mountain pole points equipped with online icing monitoring devices is selected. The online icing monitoring devices take pictures of the pole points hourly, obtaining image records and recording the altitude and latitude / longitude of the pole points, ensuring that a sufficient number of pole points collect long-term image data during a given icing event for analysis. In some embodiments, when images are missing, conductor tension load inversion can be used to indirectly obtain the ice weight, thereby obtaining the icing weight. During this process, these images are collected and analyzed to obtain information on towers with icing and identify their icing types, as well as information on towers without icing. A single icing event with a large icing area and long duration is selected. Based on the power grid icing online monitoring system, icing towers are screened using image recognition methods. The icing types among the icing towers must include at least two types: rime ice and mixed frost ice, and the number of icing towers must be greater than or equal to 10. Sampling is conducted during this icing event to obtain the corresponding first and second characteristic indicators.

[0021] The meteorological reanalysis data includes global atmospheric reanalysis data and the geographical locations of the mountainous power poles (including their altitude, height, and specific grid locations). Global atmospheric reanalysis data with rich vertical gradations was selected to meet the requirements of this application. Based on the regional range and time interval of icing at mountainous power poles obtained from the power grid icing online monitoring system, global reanalysis grid data for the corresponding time and range were downloaded. Relevant element data at all levels from 300 hPa to 1000 hPa were retained, and then interpolation methods were used to interpolate to the power poles to obtain the meteorological reanalysis data for the corresponding transmission lines.

[0022] Types of icing include high-density icing formed from freezing rain, fog in clouds, and wet snow. The type of icing that forms varies depending on the weather conditions.

[0023] Based on the aforementioned icing image data, icing types are identified, including at least typical icing types such as rime ice and mixed rime ice, thereby constructing corresponding first and second characteristic indicators. Through analysis of existing samples, the correspondence between the first and second characteristic indicators and icing types is established, allowing for the inference of future icing types based on meteorological reanalysis data.

[0024] The process of constructing the first and second characteristic indicators of icing type for power transmission lines in mountainous areas is as follows: Interpolation of sample tower points is performed based on meteorological reanalysis data; The analysis focused on the first and second characteristics closely related to icing above the sample tower points. The first characteristic included temperature indicators, and the second characteristic included cloud particle indicators. Based on the actual icing types of icing poles, we construct the first and second characteristic indicators for various icing types of transmission lines in hilly and mountainous areas.

[0025] Among them, interpolation of sample tower points is performed based on meteorological reanalysis data: Based on the region and time interval of icing occurrence at sample tower points obtained from the power grid icing online monitoring system, global climate reanalysis grid data for the corresponding time and range are downloaded, meteorological element data at all levels below at least 300 hPa are retained, and then interpolation methods are used to interpolate to the icing tower points. The interpolation methods include bilinear interpolation.

[0026] The process of analyzing the first characteristic closely related to icing above the tower point is as follows: Based on the typical distribution characteristics of air temperature and specific humidity during icing, we analyze whether the vertical distribution of air temperature at the tower point meets the "cold-warm-cold" stratification characteristics and the air temperature at the center of the middle warm layer is greater than the threshold T1. When the above conditions and thresholds are met, we record the lowest altitude of the bottom cold layer as H1 to form the air temperature index for judging icing. Based on the actual icing types at multiple pole points on the ground, the above threshold values ​​were determined as the first feature indicator for identifying icing types.

[0027] The process of analyzing the second characteristic closely related to icing above the tower point is as follows: The icing formed by freezing rain ICE1 corresponds to cloud and rain water R1 exceeding a certain threshold at the height of that tower point; The icing ICE2 formed by fog in the clouds corresponds to the height of the tower point where there is cloud liquid water, and the threshold of cloud liquid water is greater than L1; The icing ICE3 formed by wet snow corresponds to the presence of snowmelt and rainwater at the target tower height. The threshold of rainwater is greater than S1 and the threshold of rainwater is in [R2, R1). By combining multiple pole points on the ground and the actual icing type, the above threshold values ​​are determined one by one as the second feature indicator for identifying the icing type.

[0028] For cases involving multiple sample tower points, in some embodiments, the above process is performed on each of the multiple sample tower points to obtain a first characteristic index and a second characteristic index. Statistical analysis is then used to determine the first and second characteristic indices. In some embodiments, a linear averaging method is used to average the thresholds of the multiple sample tower points to obtain the first and second characteristic indices.

[0029] After obtaining the first and second characteristic indicators, the future icing type can be inferred based on meteorological reanalysis data. For sample tower points without online icing monitoring devices, the first and second characteristics are obtained through interpolation based on meteorological reanalysis data.

[0030] Based on the first characteristic, calculate whether the target tower point meets the temperature index hourly. For the time when the index is met, calculate H2-H1. If it is greater than 0, then the time is assigned as the possible time of icing and recorded as 1; otherwise, it is recorded as 0.

[0031] Based on the second feature index, the icing type identification is completed for each icing time at the target tower point, forming an hourly icing type sequence for one icing process at the target tower point. Specifically, based on the second feature index, only the time marked as a possible icing time at the target tower point is judged, the icing type identification is completed for each icing time at the target tower point, and the icing time is represented by a specific icing type (ICE1, ICE2 or ICE3), and the time without icing is represented by ICE0, forming an hourly icing type sequence according to time.

[0032] To verify the actual effectiveness of the method described in this invention, this embodiment selects a transmission line icing process that occurred in a mountainous area from November 30 to December 2, 2022, to determine the first and second meteorological characteristic indicators, and selects No. 4 with multiple icing records to identify the main icing types in the icing processes in December 2023 and February 2024, thereby verifying the reliability of the invention.

[0033] Step 1: From the data collected by the local power grid icing online monitoring system, select the tower points where transmission lines experienced icing between November 30 and December 2, 2022, and have hourly image records (no images were taken at night due to poor image quality). Identify the icing type and duration of the tower points based on the hourly images, and record the latitude, longitude, and altitude H2 of the tower points. At the same time, collect images of icing accidents where lines fell due to icing, and record the latitude, longitude, and altitude H2 of the accident points (real-time monitoring data is often missing at icing accident points due to severe icing).

[0034] Step 2: Download the global reanalysis data ERA5, including meteorological elements such as temperature (t(K)) and geopotential height (z(m)). 2 / s 2 The data includes relative humidity (r(%)), cloud and rain water (crwc(kg / kg)), and cloud and snow water (cswc(kg / kg)). The downloaded reanalysis data is then interpolated to the icing tower points and icing accident points using bilinear interpolation for each element.

[0035] Step 3: From the perspective of icing mechanism, all icing points have a warm layer at an altitude of 2000-4000m. However, pure snow icing is significantly different from rime and hoarfrost icing. In pure snow icing, the temperature at the center of the warm layer is consistently below 0℃. Based on temperature characteristics, pure snow icing and periods without icing can be eliminated. Furthermore, the altitude of the icing pole point must be within the lower cold layer. If the altitude is too low and the ground temperature remains above 0℃, no type of icing will occur. Based on the above analysis, the following specific judgment method is determined, based on the interpolated temperature (t(K)) and geopotential height (z(m)) of the icing pole point. 2 / s 2 Data was used to calculate the "cold-warm-cold" stratification distribution during icing at pole points. Specifically, for each iced pole point, the distribution was calculated vertically upwards from the bottom layer to determine if a cold-warm-cold stratification was satisfied, and certain thresholds were set for the thickness of the intermediate warm layer and the bottom cold layer. Through comparative analysis with the icing types of all iced pole points on the ground, the minimum threshold for the bottom cold layer thickness was set at 200m, and the maximum at 3000m. The threshold for the intermediate warm layer thickness was also set at greater than 250m, and the maximum at 3600m, with the temperature at the center of the warm layer >0℃. For times meeting these threshold conditions, the lowest elevation of the bottom cold layer was recorded as H1. The elevation of the icing pole point was then calculated as H2-H1; if the difference was greater than 0, the time point was recorded as a potential time point for icing.

[0036] Step 4: Based on the potential icing times calculated in Step 3, further identification indicators are constructed according to the second feature. Based on relevant knowledge in this field, the vertical distribution range of low clouds at the tower point (the range between the low cloud base and the low cloud top) is calculated using air temperature and relative humidity in ERA5. If the tower point height is within the low cloud vertical distribution range, it can be considered that cloud-fog icing has occurred at that moment. Since the ice thickness caused by freezing rain icing is greater than that caused by cloud-fog icing, the icing type identification order is: freezing rain icing first, cloud-fog second, and wet snow icing third. Freezing rain icing requires the participation of rainwater. There is a certain conversion relationship between cloud rainwater and ground freezing rain. Based on the actual icing type at the tower point, this invention, through experiments, determines the threshold values ​​for cloud rainwater and cloud snowwater as follows: R1 = 0.02 g / kg, R2 = 0.01 g / kg, L1 = 0.09 g / kg, and S1 = 0.03 g / kg.

[0037] Step 5: Based on the icing type judgment order and threshold in Step 4, select representative tower points in this icing event, such as... Figure 2 As shown, tower point 1 experienced brief periods of freezing rain and rime icing during the rain-snow transition, after which the four icing types specifically mentioned in this invention were largely absent, consistent with the monitoring data from the power grid icing online monitoring system. Tower point 2 experienced frequent periods of rime icing, interspersed with brief periods of freezing rain icing, also consistent with actual conditions. Tower point 3 failed to reflect the freezing rain process on the night of November 30th, but its depiction of the second phase of freezing rain icing was relatively accurate. Tower point 4 experienced numerous periods of icing, with the three icing types appearing intermittently, ultimately resulting in a relatively severe mixed rime icing, which was also expected. Tower point 5 experienced five periods of pure rime icing, which also matched well with the transparent icing observed on the line, indicating that the method determined in this invention can effectively identify the icing types during the icing process.

[0038] Step 6: Using tower No. 4, which experienced icing, as the target tower point for verification, reanalysis grid data was collected during two icing events (excluding December 2022) and interpolated to target tower No. 4. Specifically, hourly reanalysis data for a certain area within a certain time range in 2023 and 2024 were collected and interpolated to tower No. 4 using bilinear interpolation. Figure 3 As shown, after freezing rain and icing occurred on December 16, 2023, the icing did not increase further on the 17th and 18th. However, due to the low temperature, the rime ice sublimated, and the freezing rain and icing still existed. On December 19, icing in the clouds occurred again. It can be seen that the identification method proposed in this invention effectively identified the icing time and icing type of this process. Figure 4 This was a severe icing event in late February 2024. The ice that fell on the fourth pole, after measurement and conversion to standard ice thickness, exceeded 36mm. Icing began at the tower location at 01:00 on the 22nd when the temperature dropped to -1℃ and continued until the ice fell on the 25th, meeting all the conditions for icing. During the icing process on the fourth pole, rime and hoarfrost icing occurred successively, with virtually no wet snow icing. The icing trend continued to increase throughout the process. Figure 5 The ice thickness growth curve derived from the tensile inversion also shows that the growth rate of rime ice is the fastest, and the growth of cloud fog ice is also considerable. It can be seen that the severe icing at the tower point was caused by both rime ice and cloud fog ice.

[0039] Step 7: Based on the second feature index in Step 4, determine the time periods marked as icing times at the target tower points, complete the icing type identification for each icing time period at the target tower point, and represent the icing times with specific icing types (ICE1, ICE2, and ICE3), and the non-icing times with ICE0. Based on the hourly icing type sequence, refer to the four icing type time series for 2023 and 2024. Figure 5This is basically consistent with the actual icing situation.

[0040] Example 2: like Figure 6 As shown, this embodiment provides a system for identifying the icing type of transmission lines in hilly and mountainous areas based on meteorological macro second features, used to implement the method for identifying the icing type of transmission lines in hilly and mountainous areas based on meteorological macro second features described in Embodiment 1, including: The indicator construction module is used to acquire meteorological reanalysis data and icing observation data of pole towers in mountainous areas, select pole towers with icing images as sample pole towers, construct weather data for sample pole towers through interpolation, and construct corresponding first and second feature indicators based on the icing observation data. The icing time screening module is used to screen pole points that do not have icing images as target pole points. It interpolates the target pole points based on meteorological reanalysis data and performs hourly calculations using the first feature index to screen all times when icing occurs at the target pole points. The icing type identification module identifies the icing type for all occurrences of icing based on the second feature index, forming an hourly icing type sequence for the target tower point during an icing process.

[0041] It should be noted that each unit in the above-mentioned method and system for identifying icing types of transmission lines in hilly and mountainous areas based on the second feature of meteorological macros can be implemented entirely or partially through software, hardware, or a combination thereof. Each of these units can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each unit. For specific limitations on the method and system for identifying icing types of transmission lines in hilly and mountainous areas based on the second feature of meteorological macros, please refer to the limitations of the method for identifying icing types of transmission lines in hilly and mountainous areas based on the second feature of meteorological macros (i.e., Example 1) above; both have the same function and role, and will not be repeated here.

[0042] It will be apparent to those skilled in the art that various modifications can be made to the above embodiments, and the general principles described herein can be applied to other embodiments without inventive effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made to the present invention by those skilled in the art based on the disclosure thereof should be within the scope of protection of the present invention.

Claims

1. A method for identifying icing types on power transmission lines in mountainous areas, characterized by comprising: Meteorological reanalysis data and icing observation data of power poles in mountainous areas were acquired. Power poles with icing images were selected as sample power poles. Meteorological reanalysis data of the corresponding transmission lines were constructed by interpolation. Based on the icing observation data of the power poles, corresponding meteorological characteristic indicators were constructed. The meteorological characteristic indicators include a first characteristic indicator affecting icing and a second characteristic indicator affecting icing type. The first characteristic indicator is air temperature, and the second characteristic indicator is cloud particle type. The process of constructing the first feature index is as follows: Based on the typical distribution characteristics of air temperature during icing, we analyze whether the vertical distribution of air temperature at the sample tower points meets the "cold-warm-cold" stratification characteristics and whether the central temperature of the middle warm layer is greater than the threshold T1. When the threshold condition is met, we record the lowest altitude of the bottom cold layer as H1, thus forming the temperature index for judging icing. Based on the actual icing conditions of the sample tower points, the value of T1 is determined as the first feature indicator for identifying the icing type. The process of constructing the second feature index is as follows: The icing formed by freezing rain ICE1 corresponds to cloud and rain water R1 exceeding a certain threshold at the height of that tower point; The icing ICE2 formed by fog in the clouds corresponds to the height of the tower point where there is cloud liquid water, and the threshold of cloud liquid water is greater than L1; The icing ICE3 formed by wet snow corresponds to the presence of snowmelt and rainwater at the target tower height. The threshold of rainwater is greater than S1 and the threshold of rainwater is in [R2, R1). By combining multiple pole points on the ground and the actual icing type, the above threshold values ​​are determined one by one as the second feature indicator for icing type identification; Select any pole point in the mountainous area as the target pole point, interpolate the target pole point based on meteorological reanalysis data, and calculate whether the target pole point meets the temperature index hourly using the first characteristic index. If the temperature index is met and the altitude is higher than the lowest altitude H1 of the bottom cold layer, then the corresponding time of the target pole point is the possible time of icing. Filter out all the time times when the target pole point is icing. Based on the second characteristic index, the icing type of all occurrences of icing is identified to form an hourly icing type sequence for the target tower point during an icing process. The identification process is as follows: based on the second characteristic index, only the times marked as possible icing times for the target tower point are judged to complete the icing type identification for each icing time at the target tower point, and the icing times are represented by specific icing types, including ICE1, ICE2 and ICE3. The times without icing are represented by ICE0, and an hourly icing type sequence is formed according to time.

2. The method for identifying icing types on power transmission lines in mountainous areas according to claim 1, characterized in that, The method for obtaining icing observation data is as follows: Select the icing process of a typical icing event, and screen the icing towers using image recognition methods according to the power grid icing online monitoring system. The icing types of the icing towers should include at least rime ice and mixed rime ice, and the number of icing towers should be greater than or equal to 10.

3. The method for identifying icing types on power transmission lines in mountainous areas according to claim 1, characterized in that, The process of interpolating sample tower points based on meteorological reanalysis data is as follows: Based on the region and time interval of icing occurrence at the power grid icing monitoring system, the corresponding time and range of global climate reanalysis grid data are downloaded, retaining meteorological element data at all levels below 300 hPa, and then interpolating to the sample power grid points using interpolation methods, including bilinear interpolation.

4. The method for identifying icing types on power transmission lines in mountainous areas according to claim 1, characterized in that, The process of interpolating the target pole point based on meteorological reanalysis data is as follows: Based on the location of the target pole point and the expected time interval, download its global climate reanalysis grid data. The altitude of the target pole point is H2. Then, use the interpolation method to assign meteorological characteristic indicators to the target pole point.

5. A system for identifying icing types on power transmission lines in mountainous areas, characterized in that, include: The indicator construction module is used to acquire meteorological reanalysis data and icing observation data of power poles in mountainous areas, select power poles with icing images as sample power poles, construct meteorological reanalysis data of the corresponding transmission lines through interpolation methods, and construct corresponding meteorological characteristic indicators based on the icing observation data of the power poles. The meteorological characteristic indicators include a first characteristic indicator affecting icing and a second characteristic indicator affecting icing type; the first characteristic indicator is air temperature, and the second characteristic indicator is cloud particle type. The process of constructing the first feature index is as follows: Based on the typical distribution characteristics of air temperature during icing, we analyze whether the vertical distribution of air temperature at the sample tower points meets the "cold-warm-cold" stratification characteristics and whether the central temperature of the middle warm layer is greater than the threshold T1. When the threshold condition is met, we record the lowest altitude of the bottom cold layer as H1, thus forming the temperature index for judging icing. Based on the actual icing conditions of the sample tower points, the value of T1 is determined as the first feature indicator for identifying the icing type. The process of constructing the second feature index is as follows: The icing formed by freezing rain ICE1 corresponds to cloud and rain water R1 exceeding a certain threshold at the height of that tower point; The icing ICE2 formed by fog in the clouds corresponds to the height of the tower point where there is cloud liquid water, and the threshold of cloud liquid water is greater than L1; The icing ICE3 formed by wet snow corresponds to the presence of snowmelt and rainwater at the target tower height. The threshold of rainwater is greater than S1 and the threshold of rainwater is in [R2, R1). By combining multiple pole points on the ground and the actual icing type, the above threshold values ​​are determined one by one as the second feature indicator for icing type identification; The icing time filtering module is used to select any pole point in the mountainous area as the target pole point. Based on meteorological reanalysis data, the target pole point is interpolated and the first feature index is used to calculate whether the target pole point meets the temperature index hourly. If the temperature index is met and the altitude is higher than the lowest altitude H1 of the bottom cold layer, then the corresponding time of the target pole point is the possible icing time. All times when the target pole point is icing are filtered out. The icing type identification module is used to identify the icing type of all occurrences of icing based on a second feature index, forming an hourly icing type sequence for the target tower point during an icing process. The identification process is as follows: based on the second feature index, only the times marked as possible icing times for the target tower point are judged, completing the icing type identification for each icing time at the target tower point, and representing the icing times with specific icing types, including ICE1, ICE2, and ICE3, while the times without icing formation are represented by ICE0, forming an hourly icing type sequence according to time.

Images