A de-icing control method and system for power transmission lines in cold regions

Abstract

The application provides a deicing control method and system for power transmission lines in cold regions, and relates to the technical field of power system transmission line operation and maintenance. The method comprises the following steps: collecting icing data of the power transmission line, the icing data being multi-source data, including ice layer thickness, ice type category and distribution characteristics, combining environmental temperature, humidity and geographical location information, dynamically selecting a plurality of spatial sampling positions on the target line to construct a spatial correlation region for feature analysis; obtaining a correction parameter based on the morphological characteristics of the spatial correlation region, performing spatial superposition and feature fusion analysis on the multi-source data according to the correction parameter, and extracting an icing feature set that reflects the overall icing state of the line after dynamic correction. The application can realize differentiated ice removal in different climate regions, and realize intelligent deicing control through precise identification, mode adaptation and closed-loop adjustment combined with multi-source monitoring data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system transmission line operation and maintenance, in particular to an ice removal control method and system for a transmission line in a cold region. BACKGROUND

[0002] Transmission line icing is a major threat to the safe operation of power grids in winter regions, especially in cold regions. Icing includes various types, among which thick mixed rime is easily formed in the high-cold northern region, which has high density and strong toughness, and thin and brittle glaze icing is formed in the southern region due to freezing rain. Icing can cause sudden increase of conductor load, tower deformation or even collapse, and may also cause line galloping to cause short circuit faults, which not only causes economic losses, but also seriously affects people's life rhythm.

[0003] Existing ice removal technologies mainly include thermal ice melting and mechanical ice removal. Thermal ice melting requires the application of kilo-ampere short-circuit current, which has high energy consumption and needs to be operated with power off, and is difficult to adapt to remote mountain lines. Mechanical ice removal relies on resonance effect to break the ice layer, but has limited effect on thick ice with high toughness in the northern region. Although the shovel scraping type mechanical ice removal can handle hard ice, it is easy to damage the conductor surface covered with thin ice in the southern region. More importantly, the existing technology has many limitations: on the one hand, the ice condition monitoring lacks spatial correlation correction, for example, the collected data is easily disturbed by temperature gradient, and the mechanical properties of icing in the northern and southern regions cannot be accurately distinguished; on the other hand, the ice removal mode is fixed, for example, the operation mode is not matched for the characteristics of thick ice with high toughness in the northern region and thin ice with high toughness in the southern region, resulting in the dilemma of thick ice not being broken and thin ice being damaged by shoveling; there is also a lack of closed-loop regulation mechanism, which cannot dynamically optimize parameters or switch modes according to the ice removal effect, and the problems of incomplete ice removal or over-removal often occur; in addition, there are technical problems such as spatial deviation and feature misjudgment of ice condition monitoring, and lack of adaptability of ice removal mode to the characteristics of ice regions. SUMMARY

[0004] The technical problem to be solved by the present application is to provide an ice removal control method and system for a transmission line in a cold region, which can realize differentiated ice removal in different climate regions, and realize intelligent ice removal control through precise identification, mode adaptation and closed-loop regulation combined with multi-source monitoring data.

[0005] To solve the above technical problems, the technical scheme of the present application is as follows:

[0006] In a first aspect, an ice removal control method for a transmission line in a cold region, the method comprising:

[0007] Collecting icing data of the transmission line, the icing data being multi-source data including ice layer thickness, ice type category and distribution characteristics, combining with environmental temperature, humidity and geographical location information, dynamically selecting a plurality of spatial sampling positions on the target line to construct a spatial correlation region for feature analysis;

[0008] The correction parameters are obtained based on the morphological features of the spatial correlation region, and the multi-source data are subjected to spatial superposition and feature fusion analysis according to the correction parameters, so as to extract an icing feature set reflecting the overall icing state of the line after dynamic correction;

[0009] According to the icing feature set, it is identified whether the current region belongs to a northern cold heavy icing area or a southern light icing area, wherein the northern cold heavy icing area is mainly thick icing and ductile ice type, and the southern light icing area is mainly thin icing and brittle ice type;

[0010] If it is identified as the northern cold heavy icing area, the shovel impact deicing mode is started, the deicing device is controlled to move and position along the line, the ice layer structure and adhesion state are identified in real time in combination with the visual perception information, and high-intensity impact deicing is implemented through the shovel impact mechanism; if it is identified as the southern light icing area, the vibration deicing mode is started, the vibration parameters are dynamically adjusted according to the visual feedback, and the deicing device is controlled to break and remove the icing in a high-frequency vibration mode;

[0011] In the deicing mode process, deicing effect information is continuously obtained;

[0012] Based on the deicing effect information, the deicing action parameters are dynamically feedback adjusted; when the deicing effect does not reach the set standard, the execution parameters of the current deicing mode are adaptively optimized or another deicing mode is entered, until the icing is completely removed, so as to realize closed-loop adaptive deicing.

[0013] Further, icing data of the power transmission line are collected, the icing data are multi-source data including ice layer thickness, ice type category and distribution characteristics, in combination with environmental temperature, humidity and geographical location information, a plurality of spatial sampling positions are dynamically selected on the target line to construct a spatial correlation region for feature analysis, including:

[0014] Through the sensor device deployed along the power transmission line, the ice layer thickness, ice type category and distribution information along the line of the line surface are obtained, the environmental temperature, humidity and geographical location parameters are synchronously collected, and a multi-source monitoring data set is obtained;

[0015] Based on the icing thickness variation trend, ice type distribution characteristics, and environmental temperature gradient and humidity variation characteristics contained in the multi-source monitoring data set, a plurality of representative spatial sampling positions are dynamically determined on the target line, and each sampling position is adaptively selected according to the continuity of the icing distribution and the regional regularity of the environmental influence;

[0016] Based on the spatial relationship of the spatial sampling positions, a spatial correlation region for icing feature extraction is constructed;

[0017] The morphological features of the spatial correlation region are analyzed to obtain compensation parameters for data correction;

[0018] The multi-source monitoring data sets are subjected to spatial superposition and feature fusion processing based on the compensation parameters to obtain an icing feature set that is preliminarily dynamically corrected.

[0019] Further, the correction parameters are obtained based on the morphological features of the spatially related regions, and the multi-source data are subjected to spatial superposition and feature fusion analysis according to the correction parameters to extract an icing feature set that is dynamically corrected and reflects the overall icing state of the line, including:

[0020] The contour features and spatial distribution attributes of the spatially related regions are obtained, and the morphological change trend and structural stability are analyzed;

[0021] According to the morphological change trend and structural stability, the correction parameters for correcting the spatial deviation of the multi-source monitoring data are obtained;

[0022] Based on the correction parameters, the ice layer thickness, ice type category and distribution features in the multi-source monitoring data sets are subjected to spatial registration and weight adjustment;

[0023] The multi-source data subjected to spatial registration and weight adjustment are subjected to spatial superposition and feature fusion to extract a feature vector that can reflect the overall icing state of the line;

[0024] Based on the feature vector, an icing feature set that is deeply dynamically corrected is constructed.

[0025] Further, according to the icing feature set, it is identified whether the current region belongs to a northern high-cold heavy icing area or a southern light icing area, wherein the northern high-cold heavy icing area is mainly characterized by thick icing and ductile ice type, and the southern light icing area is mainly characterized by thin icing and brittle ice type, including:

[0026] The icing thickness distribution features and ice type mechanical property features are extracted from the icing feature set that is deeply dynamically corrected;

[0027] Based on the icing thickness distribution features, the overall icing load level is determined, and the ice layer anti-breaking capacity is analyzed in combination with the ice type mechanical property features;

[0028] According to the combined features of the icing load level and the ice layer anti-breaking capacity, a regional type discrimination rule is established;

[0029] According to the regional type discrimination rule, the current region is identified as a northern high-cold heavy icing area characterized by high load and high ductility, or a southern light icing area characterized by low load and brittleness.

[0030] Further, if identified as a heavy ice area in the north, a shovel striking deicing mode is started, the deicing device is controlled to move and position along the line, the ice layer structure and adhesion state are identified in real time in combination with visual perception information, and high-strength impact deicing is implemented through the shovel striking mechanism; if identified as a light ice area in the south, a vibration deicing mode is started, vibration parameters are dynamically adjusted according to visual feedback, and the deicing device is controlled to break and remove ice in a high-frequency vibration mode, including:

[0031] The receiving area type identification result is received, if identified as a heavy ice area in the north, a shovel striking deicing mode start instruction is sent to the deicing device; if identified as a light ice area in the south, a vibration deicing mode start instruction is sent;

[0032] In the shovel striking deicing mode, the deicing device is controlled to move to a target position along the power transmission line, the ice layer surface image is obtained in real time through the visual sensor, and the ice layer thickness distribution and adhesion strength characteristics are identified;

[0033] According to the ice layer thickness distribution and adhesion strength characteristics, the impact angle and action strength parameters of the shovel striking mechanism are determined, and the shovel striking mechanism is driven to implement high-strength impact removal of ice;

[0034] In the vibration deicing mode, the deicing device is positioned to the ice-covered section, and the ice layer brittleness characteristics and breaking state are monitored through the visual sensor;

[0035] Based on the real-time feedback of the ice layer brittleness characteristics and breaking state, the vibration frequency and action amplitude are dynamically adjusted to effectively break and remove ice in real time.

[0036] Further, in the deicing mode, deicing effect information is continuously obtained, including:

[0037] In the process of shovel striking deicing or vibration deicing operation, the ice layer image sequence of the operation area is continuously collected through the visual sensor, and the remaining ice layer thickness data is obtained;

[0038] The ice layer image sequence is analyzed in real time, the ice layer coverage change trend and line surface bare state are identified, and the deicing effect evaluation index is obtained in combination with the remaining ice layer thickness data;

[0039] Based on the deicing effect evaluation index, it is judged whether the actual effect of the current deicing operation reaches the preset removal standard, and the deicing effect information is updated in real time.

[0040] Further, based on the deicing effect information, the deicing action parameters are dynamically feedback adjusted; when the deicing effect does not reach the set standard, the execution parameters of the current deicing mode are adaptively optimized or another deicing mode is entered, until the ice is completely removed, to realize closed-loop adaptive deicing, including:

[0041] receive deicing effect information, extract deicing effect evaluation index and real-time operation data therefrom;

[0042] based on the deicing effect evaluation index and real-time operation data, determine whether the current deicing operation efficiency meets the preset removal standard;

[0043] If the operation efficiency does not meet the set standard, analyze the deicing resistance source based on the current deicing process, and identify whether it is parameter mismatch or mode mismatch;

[0044] If it is parameter mismatch, dynamically adjust the execution parameters of the current deicing mode, including impact strength, vibration frequency or operation speed;

[0045] If it is mode mismatch, switch to another deicing mode, and reinitialize the operation parameters based on the current icing condition; repeat the above steps to obtain a closed-loop adjustment mechanism based on deicing effect feedback until the ice is completely removed.

[0046] In a second aspect, a deicing control system for power transmission lines in cold regions includes:

[0047] The acquisition module is configured to collect icing data of the power transmission line, the icing data being multi-source data including ice layer thickness, ice type category and distribution characteristics, and to dynamically select a plurality of spatial sampling positions on the target line to construct a spatial correlation region for feature analysis in combination with environmental temperature, humidity and geographic location information;

[0048] The analysis module is configured to obtain correction parameters based on the morphological characteristics of the spatial correlation region, to perform spatial superposition and feature fusion analysis on the multi-source data according to the correction parameters, and to extract an icing feature set that reflects the overall icing state of the line after dynamic correction;

[0049] The identification module is configured to identify whether the current region belongs to a northern cold and heavy ice region or a southern light ice region according to the icing feature set, wherein the northern cold and heavy ice region is mainly composed of thick ice and ductile ice type, and the southern light ice region is mainly composed of thin ice and brittle ice type;

[0050] The execution module is configured to start a shovel-impact deicing mode if the region is identified as a northern cold and heavy ice region, to control the deicing device to move and position along the line, to combine visual perception information to identify the ice layer structure and adhesion state in real time, and to implement high-intensity impact deicing through a shovel-impact mechanism; if the region is identified as a southern light ice region, start a vibration deicing mode, dynamically adjust vibration parameters according to visual feedback, and control the deicing device to break and remove the ice in a high-frequency vibration mode;

[0051] An adjusting module is configured to continuously acquire deicing effect information during the deicing mode, dynamically feedback adjust the deicing action parameters based on the deicing effect information, and adaptively optimize the execution parameters of the current deicing mode or enter another deicing mode when the deicing effect does not reach the set standard until the ice cover is completely removed, so as to realize the closed-loop adaptive deicing.

[0052] The above scheme of the present application at least includes the following beneficial effects:

[0053] Because the technical means of fusing multi-source icing data and environmental information, dynamically constructing a spatial correlation region, and generating a correction parameter are adopted, the feature misjudgment problem caused by the lack of spatial correlation correction of ice condition monitoring and the interference of temperature gradient in the prior art is overcome, and the accuracy of icing state recognition is improved, and the mechanical properties of icing in the north and south ice regions can be effectively distinguished; because the technical means of identifying regional types based on an icing feature set and matching a differentiated deicing mode (north shovel strike and south vibration) are adopted, the thick ice vibration and thin ice shovel damage line dilemma caused by the solidification of the deicing mode and the lack of adaptability to the ice region characteristics are overcome, and the line damage risk is reduced, and the deicing efficiency is improved; because the closed-loop adaptive adjustment technical means of real-time monitoring of deicing effect and dynamic optimization of parameters or switching of modes are adopted, the incomplete deicing or overwork problem caused by the lack of closed-loop adjustment mechanism is overcome, and a complete control chain of monitoring, identification, operation, and feedback is formed, and the thoroughness and safety of ice removal are ensured. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 is a flowchart of a deicing control method for a power transmission line in a cold region provided by an embodiment of the present application.

[0055] Figure 2 is a schematic diagram of a deicing control system for a power transmission line in a cold region provided by an embodiment of the present application. DETAILED DESCRIPTION

[0056] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art.

[0057] As Figure 1 shown, an embodiment of the present application proposes a deicing control method for a power transmission line in a cold region, which comprises the following steps:

[0058] Step 1, collecting icing data of the power transmission line, the icing data being multi-source data, including ice layer thickness, ice type category and distribution characteristics, combining with environmental temperature, humidity and geographic location information, dynamically selecting a plurality of spatial sampling positions on the target line to construct a spatial correlation region for feature analysis;

[0059] Step 2, obtaining a correction parameter based on the morphological characteristics of the spatial correlation region, and performing spatial superposition and feature fusion analysis on the multi-source data according to the correction parameter to extract an icing feature set reflecting the overall icing state of the line after dynamic correction;

[0060] Step 3, identifying whether the current region belongs to a northern cold heavy icing area or a southern light icing area according to the icing feature set, wherein the northern cold heavy icing area is mainly composed of thick ice and ductile ice type, and the southern light icing area is mainly composed of thin ice and brittle ice type;

[0061] Step 4, if the northern cold heavy icing area is identified, starting a shovel impact deicing mode, controlling the deicing device to move and position along the line, combining with visual perception information to identify the ice layer structure and adhesion state in real time, and implementing high-intensity impact deicing through the shovel impact mechanism; if the southern light icing area is identified, starting a vibration deicing mode, dynamically adjusting the vibration parameters according to the visual feedback, and controlling the deicing device to break and remove the ice in a high-frequency vibration mode;

[0062] Step 5, continuously obtaining deicing effect information during the deicing mode;

[0063] Step 6, dynamically feedback adjusting the deicing action parameters based on the deicing effect information; when the deicing effect does not reach the set standard, adaptively optimizing the execution parameters of the current deicing mode or entering another deicing mode until the ice is completely removed, so as to realize closed-loop adaptive deicing.

[0064] In the embodiment of the present application, the spatial correlation region is constructed by combining multi-source icing data with environmental temperature, humidity and geographic location information, and the correction parameter is generated, which overcomes the problem of one-sidedness and spatial deviation of traditional monitoring data, and obtains a precise feature set reflecting the overall icing state of the line; then the ice characteristics are extracted to establish rules, and the northern cold heavy icing area and the southern light icing area are accurately identified, solving the problem of misjudgment of ice area type; then the shovel impact or vibration differential deicing mode is matched according to different ice areas, and the operation parameters are adjusted combined with visual feedback to avoid the problems of thick ice being difficult to remove or thin ice damaging the line caused by the solidification of the deicing mode; at the same time, the deicing effect is continuously monitored, and the parameters are dynamically optimized or the mode is switched through closed-loop adjustment, overcoming the defects of incomplete deicing or excessive operation, and finally realizing the precision, differentiation and intelligence of deicing of power transmission lines in cold regions, and ensuring the operation efficiency and line safety.

[0065] In a preferred embodiment of the present application, the above step 1 can include:

[0066] Step 1.1, obtain the ice thickness, ice type and their distribution along the line, and synchronously collect the environmental temperature, humidity and geographical location parameters, to obtain a multi-source monitoring data set, including: evenly deploying monitoring sensors on the conductor section, insulator string and tower connection of the transmission line to realize multi-dimensional data acquisition; for the line surface icing, dynamic detection is carried out by the monitoring device to record the quantitative data of the ice thickness of different monitoring points in real time, and at the same time, the observation and analysis of the icing surface texture, transparency and structure form are combined to distinguish different ice type categories such as mixed rime and glaze; set monitoring nodes along the line at reasonable intervals to continuously track the existence state and coverage range of the icing at each node to obtain the spatial record of the icing distribution along the line; on this basis, the environmental temperature, relative humidity data at each monitoring node are synchronously collected, and the longitude, latitude and altitude of each monitoring point and other geographical location information are recorded, and finally all the collected ice thickness, ice type, distribution information, and environmental temperature, humidity and geographical location parameters are uniformly summarized and regularized to obtain a multi-source monitoring data set containing multi-dimensional information.

[0067] Step 1.2, based on the ice thickness variation trend, ice type distribution characteristics, and environmental temperature gradient and humidity variation characteristics contained in the multi-source monitoring data set, dynamically determine a plurality of representative spatial sampling positions on the target line, each sampling position is adaptively selected according to the continuity of the icing distribution and the regional regularity of the environmental influence, including: first, pre-process the data in the multi-source monitoring data set, after eliminating abnormal fluctuation data, focus on analyzing the ice thickness variation trend, by comparing the ice thickness values of each monitoring point in the continuous time period, determine the line section with faster ice thickness growth or larger thickness value; analyze the ice type distribution characteristics, mark the mixed rime concentrated distribution section and the glaze concentrated distribution section according to the image recognition result; analyze the environmental temperature gradient, calculate the temperature difference between adjacent monitoring points to determine the area with significant temperature change; analyze the humidity variation characteristics, record the high humidity section with humidity value higher than 85% for a long time; then, combined with the above analysis results, in the line section where the icing distribution presents continuous state and no obvious fault appears, preferentially select the ice thickness variation trend turning point, different ice type boundary point and position with significant temperature gradient and humidity variation as the candidate spatial sampling position, at the same time, ensure that the selected sampling position can cover different areas affected by the environment, such as the windward section and leeward section of the mountain line, the open section and sheltered section of the plain line, etc., finally determine a plurality of representative spatial sampling positions.

[0068] Step 1.3, based on the spatial relationship of the spatial sampling positions, constructing a spatial correlation region for ice-coating feature extraction, specifically including: first obtaining the accurate geographic coordinates of each sampling position, then marking these geographic coordinates on the electronic map of the power transmission line to obtain the distribution atlas of the spatial sampling positions; then analyzing the continuity of ice-coating distribution between adjacent spatial sampling positions, if the difference in ice-coating thickness between two adjacent sampling positions is less than 5 mm and the ice type categories are consistent, and the environmental temperature difference between the regions where the two sampling positions are located is less than 2°C and the relative humidity difference is less than 10%, then it is determined that the two sampling positions belong to the same ice-coating feature correlation region; according to the above determination standard, grouping all spatial sampling positions, and dividing the sampling positions with similar ice-coating features and environmental features and adjacent geographic positions into the same region, each region contains at least 3 to 5 spatial sampling positions, and finally a plurality of spatial correlation regions for ice-coating feature extraction are constructed.

[0069] Step 1.4, analyzing the morphological features of the spatial correlation region to obtain compensation parameters for data correction, specifically including: for each spatial correlation region, first integrating the ice-coating thickness data, ice type category data and corresponding environmental data of all spatial sampling positions in the region, drawing the ice-coating morphological distribution map of the region, marking the maximum thickness value, minimum thickness value and average thickness value of the ice-coating in the region on the distribution map, outlining the contour shape of the ice-coating, and determining whether the ice-coating contour presents a regular cylindrical shape or an irregular block shape to analyze the morphological change trend of the ice-coating; at the same time, according to the 24-hour ice-coating thickness change data and environmental parameter change data in the region, determining whether the ice layer has local shedding, cracking or continuous thickening, to evaluate the stability of the ice-coating structure; then comparing the actual monitoring data of the region with the ideal ice-coating monitoring data under ideal conditions, the ideal ice-coating monitoring data refers to the ice-coating data under standard temperature, humidity conditions and without external interference, calculating the deviation value between the actual monitoring data and the ideal data, calculating the compensation coefficient according to the possible deviation amplitude of the thickness measurement value for every 1°C temperature drop caused by temperature gradient, calculating the compensation weight according to the change of ice type recognition accuracy for every 10% humidity increase caused by humidity change, and finally obtaining the compensation parameters for correcting the spatial deviation of multi-source monitoring data by comprehensively considering these compensation coefficients and compensation weights.

[0070] Step 1.5, based on the compensation parameter, the multi-source monitoring data set is spatially superimposed and feature fusion processing is performed to obtain an icing feature set that is preliminarily dynamically corrected, which specifically includes: first, according to the compensation parameter, the ice thickness data of each spatial sampling position in the multi-source monitoring data set is corrected, for example, the ice thickness value measured by the ultrasonic sensor at a sampling position is 2mm smaller than the actual value due to the excessively low ambient temperature, so 2mm is added to the original measured value according to the compensation parameter to ensure that the corrected thickness data is close to the actual ice thickness; the ice type category data is subjected to consistency checking, if the ice type identified at a sampling position in the same spatially associated region is inconsistent with the ice type identified at most of the sampling positions in the region, and the ice type identification at the sampling position is judged to be greatly disturbed by humidity according to the compensation parameter, then the ice type category at the sampling position is adjusted by referring to the ice type category at most of the sampling positions in the region and combining the compensation parameter; then the ice thickness and ice type category data of all the corrected spatial sampling positions are superimposed in space according to their geographic coordinates to form an icing data spatial distribution map covering the entire target line; then the superimposed data is subjected to feature fusion, in which the icing thickness data is given a higher weight, and the ice type category data and distribution feature data are given corresponding weights, the weight values are determined according to the importance of each data in reflecting the overall icing state of the line, and finally the feature information representing the icing thickness distribution, ice type category proportion and icing coverage range of the entire target line is extracted through fusion calculation, and the feature information is arranged to obtain the icing feature set that is preliminarily dynamically corrected.

[0071] In the embodiments of the present application, multi-source icing data and environmental parameters are acquired through sensors, overcoming the problem of the traditional single-point monitoring information being one-sided; the sampling positions are dynamically selected according to the icing trend and environmental characteristics, solving the problem of poor adaptability of fixed sampling; the spatially associated region is constructed based on the sampling positions, making up for the lack of spatial correlation of the data; the compensation parameter is analyzed based on the region shape, making clear the basis for data correction; finally, the dynamically corrected icing feature set is obtained through superposition and fusion, solving the problems of data fragmentation and large deviation, and providing accurate data support for subsequent ice region identification and differentiated deicing.

[0072] In a preferred embodiment of the present application, step 2 can include:

[0073] Step 2.1, obtain the contour features and spatial distribution attributes of the spatially associated region, analyze the morphological change trend and structural stability, specifically including: based on the constructed spatially associated region, first associate the icing monitoring data and geographic location information of each spatial sampling point according to the geographic coordinates, extract the icing existence state data within the effective monitoring radius centered on each spatial sampling position, after marking all the points with icing, obtain the initial boundary of icing coverage by connecting the peripheral edge points, remove the points on the boundary with icing thickness below the minimum effective monitoring threshold, refer to the actual orientation of the transmission line to optimize and adjust the overall contour shape of icing (continuous strip or local block); then, calculate the actual length of the icing coverage section per unit length of the line to obtain the coverage ratio, and analyze the relative position relationship between the icing and the line components such as towers and insulator strings, and determine the distribution characteristics of the icing around the key components, to obtain the contour features and spatial distribution attributes of the spatially associated region; compare the icing contour and thickness data of the region at different times, and judge the morphological change trend of the icing by the extension and reduction of the icing boundary, the increase and decrease of the thickness, and the change of the local ice-free points; combined with the environmental temperature and humidity fluctuation data in the region, if the temperature suddenly changes and the icing thickness decreases discontinuously, it is judged that the ice layer may crack, if the humidity decreases from high humidity to medium humidity and there is a gap between the icing and the conductor, it is judged that the adhesion strength decreases, and accordingly the state of the icing is determined to be continuous and stable, slowly changing or rapidly evolving.

[0074] Step 2.2, according to the morphological change trend and structural stability, obtain the correction parameters for correcting the spatial deviation of multi-source monitoring data, specifically including: according to the analysis results of the morphological change trend and the structural stability, sort out the possible types of spatial deviation of multi-source monitoring data; if the morphological change trend shows that the icing thickness monitoring value of a certain spatially associated region decreases abnormally along the line, it is determined that there is a thickness monitoring deviation in this region due to terrain shielding, by calculating the thickness difference between the abnormal decreasing section and the normal section, combined with the height and angle parameters of the shielding terrain, the correction coefficient of the ice layer thickness data in this region is derived; if the structural stability analysis finds that the ice type identification in a certain region is confused due to local cracking (such as a mixed rime caused by cracking is misjudged as a rain rime), then for the ice type category data of this region, according to the correlation between the cracking degree and the ice type identification error, set the ice type correction weight; comprehensive all the correction coefficients and correction weights corresponding to the deviation types, form a complete set of correction parameters for correcting the spatial deviation of multi-source monitoring data, each correction parameter corresponds to a specific data type (thickness, ice type or distribution characteristics) of a specific spatially associated region.

[0075] Step 2.3, based on the correction parameters, the ice layer thickness, ice type category and distribution characteristics in the multi-source monitoring data set are spatially registered and weighted adjusted, specifically including: according to the correction parameters, first, the ice layer thickness, ice type category and distribution characteristics in the multi-source monitoring data set are spatially registered and processed: for ice layer thickness data, according to the thickness correction coefficient in the correction parameters, the original thickness values of each sampling position in the spatially related region are differentially adjusted to ensure that the thickness data of different sampling positions in the same region can truly reflect the actual spatial distribution of ice cover; for ice type category data, according to the ice type correction weight in the correction parameters, the ice type identification results of the region with poor structural stability are re-verified, if the ice type identification weight of a sampling position is lower than the set threshold, the ice type category of most sampling positions in the region is referred to for correction; for ice cover distribution characteristic data, according to the spatial coordinate offset in the correction parameters, the distribution record position of each monitoring point is adjusted to ensure that the distribution data can accurately correspond to the actual section of the line; after spatial registration, the influence degree of data on subsequent ice area identification is combined, and the weight of each data type is adjusted based on the correction parameters, such as increasing the weight of ice layer thickness data and appropriately reducing the weight of ice type data in the region with greater environmental interference.

[0076] Step 2.4, spatially superimpose and fuse the multi-source data after spatial registration and weight adjustment to extract feature vectors that can reflect the overall icing state of the line, specifically including: superimpose the multi-source data after spatial registration and weight adjustment on the corresponding line section according to the geographical coordinates of each data, and form a comprehensive data layer covering the spatially related region; on this basis, the information in the comprehensive data layer is integrated: extract the core features of the ice cover in the region, including the average thickness, maximum thickness, proportion of different ice types, total length of ice cover and number of continuous covered sections, and at the same time, according to the weight adjustment results, each feature is quantitatively processed (such as multiplying the average thickness by the thickness data weight and multiplying the ice type proportion by the ice type data weight); arrange all the quantified core features in a predetermined order to form a set of numerical combinations that can fully reflect the icing state of the spatially related region, i.e. the feature vector reflecting the overall icing state of the line, and the dimension of each feature vector corresponds to a key icing feature.

[0077] Step 2.5, constructing the ice-coating feature set after deep dynamic correction based on the feature vectors, specifically comprising: collecting the feature vectors of all spatially associated regions, sorting these feature vectors to ensure that each feature vector can correspond to a specific section of the line; for each feature vector, supplementing the corresponding spatially associated region number, data correction time and correction parameter source, then structurally integrating all feature vectors, taking the average thickness, ice type proportion, coverage length and other feature dimensions in each feature vector as columns, and taking the feature vectors of the corresponding spatially associated regions as rows to form a tabular ice-coating feature set; at the same time, performing consistency checking on the data in the feature set, if there is a significant difference in the feature vectors of adjacent spatially associated regions (such as a thickness difference exceeding a reasonable range), then backtracking to step 2.3 to recheck the weight adjustment process to ensure that the data in the feature set can reflect the changes in the ice-coating state of the entire line coherently, and finally constructing the ice-coating feature set after deep dynamic correction.

[0078] In the embodiments of the present application, by analyzing the profile, distribution and morphological and structural characteristics of the spatially associated regions, the limitations of ice condition analysis relying on single-point data are overcome, laying a foundation for data correction; the spatial deviation correction parameters obtained therefrom solve the problem of lack of basis for deviation correction; based on the parameters, the ice layer thickness, ice type category and distribution characteristics are spatially registered and the weights are adjusted, improving the problems of uncoordinated data and unreasonable weights; and then the feature vectors reflecting the overall ice-coating state of the line are extracted through spatial superposition and fusion, solving the problem of data fragmentation; and finally the ice-coating feature set after dynamic correction is constructed, providing high-quality data support for subsequent ice region identification and differentiated de-icing mode matching, and overall improving the accuracy of ice condition monitoring and feature extraction.

[0079] In a preferred embodiment of the present application, the above-mentioned step 3 can comprise:

[0080] Step 3.1, extracting the ice-coating thickness distribution characteristics and ice type mechanical property characteristics from the ice-coating feature set after deep dynamic correction, specifically comprising: from the ice-coating feature set after deep dynamic correction, preferentially extracting the ice-coating thickness distribution characteristics, which include the average thickness, maximum thickness and minimum thickness of the ice-coating in each spatially associated region of the target line, as well as the proportion of the length of the section with a thickness exceeding 20 mm to the total length of the line, and the variation gradient of the thickness along the line direction, etc., which can intuitively reflect the spatial distribution differences of the ice-coating; and extracting the ice type mechanical property characteristics, combining the ice type categories (such as mixed rime and glaze) recorded in the ice-coating feature set with the corresponding mechanical parameters, including the tensile strength, impact strength (reflecting toughness) and fracture toughness (reflecting brittleness) of the ice layer, wherein the tensile strength and impact strength of mixed rime are higher, and the fracture toughness of glaze is lower, and by quantifying these parameters, the ice-coating thickness distribution characteristics and ice type mechanical property characteristics set are obtained.

[0081] Step 3.2, judging the overall icing load level based on the icing thickness distribution characteristics, and analyzing the ice layer anti-breaking capacity combining with the mechanical property characteristics of ice type, specifically including: judging the overall icing load level based on the extracted icing thickness distribution characteristics, first setting the relative grade standard of load evaluation, combining the load bearing limit of line design and the industry ice area load grading specification, if the average icing thickness of each spatially related region of the line is in the higher interval, and the proportion of thick icing section in the total length of the line is large, then the overall icing load level is determined as high load; if the average icing thickness is in the lower interval, and the proportion of thick icing section is extremely small, then it is determined as low load; if it is between the two, then further judge the load grade combining the actual load bearing capacity of the line tower; when analyzing the ice layer anti-breaking capacity, combining the relative level of tensile strength, impact strength and fracture toughness of the ice layer in the mechanical property characteristics of ice type, if the impact strength and tensile strength corresponding to the ice type are at a higher level, it means that the ice layer has strong anti-breaking capacity; if the fracture toughness corresponding to the ice type is at a lower level, it means that the ice layer has weak anti-breaking capacity, and reference is made to the icing structure stability data, if there are many micro cracks or easy to fall off characteristics in the ice layer, then the anti-breaking capacity evaluation grade is further reduced.

[0082] Step 3.3, establishing regional type discrimination rules according to the combined characteristics of icing load level and ice layer anti-breaking capacity, specifically including: establishing regional type discrimination rules according to the combined characteristics of icing load level and ice layer anti-breaking capacity; Rule one: when the overall icing load level is high and the ice layer anti-breaking capacity is strong, corresponding to the ice condition characteristics of northern cold heavy ice area, the combination of characteristics is defined as the discrimination mark of northern ice area; Rule two: when the overall icing load level is low and the ice layer anti-breaking capacity is weak, corresponding to the ice condition characteristics of southern light ice area, the combination of characteristics is defined as the discrimination mark of southern ice area; At the same time, special situation supplementary rules are set, if the combination of high load but weak anti-breaking capacity, or low load but strong anti-breaking capacity appears, further verification is needed combining with geographical location information (such as whether it is in the northern high latitude cold area or the southern humid freezing rain area), for example, when high load weak anti-breaking capacity ice condition appears in the northern area, the ice type identification result needs to be reviewed, and after excluding data errors, the regional type is determined again to ensure the comprehensiveness and accuracy of the discrimination rules.

[0083] Step 3.4, according to the regional type judgment rule, the current region is identified as the northern cold heavy ice region characterized by high load and high toughness, or as the southern light ice region characterized by low load and brittleness, specifically comprising: according to the regional type judgment rule, first, the ice load level evaluation result and the ice layer anti-breaking capacity analysis result of the current region are obtained, and the two are combined and matched; if the combined characteristics of the current region meet the identification of high load + strong anti-breaking capacity in rule one, and the geographical location information confirms that it is in the northern high latitude, low altitude or mountain cold zone, then the current region is formally identified as the northern cold heavy ice region, the ice of which is mainly mixed rime with the typical characteristics of thick and strong toughness; if the combined characteristics of the current region meet the identification of low load + weak anti-breaking capacity in rule two, and the geographical location is in the southern subtropical or temperate humid area, and the winter weather is prone to freezing rain, then the current region is identified as the southern light ice region, the ice of which is mainly rain rime with the typical characteristics of thin and high brittleness; after the identification is completed, a regional type identification file is obtained, recording the correspondence between load level, anti-breaking capacity, geographical location and regional type.

[0084] In the embodiment of the application, by extracting the ice thickness distribution and ice type mechanical property characteristics in the dynamically corrected ice coating feature set, the limitations of single feature analysis are overcome, providing key basis for ice region identification; the ice load level and ice layer anti-breaking capacity are judged by combining the two types of characteristics, solving the problem of one-sided evaluation and realizing accurate judgment of ice condition influence; according to this, the regional type judgment rule is established, eliminating the identification confusion caused by the lack of unified standard; finally, the northern cold heavy ice region and the southern light ice region are distinguished according to the rule, solving the problem of ice region type misjudgment.

[0085] In a preferred embodiment of the application, the above-mentioned step 4 can comprise:

[0086] Step 4.1, receiving the regional type identification result, if it is identified as the northern cold heavy ice region, sending the shovel impact ice removal mode start instruction to the ice removal device; if it is identified as the southern light ice region, sending the vibration ice removal mode start instruction, specifically comprising: receiving the regional type identification result, which contains the type identification (northern cold heavy ice region or southern light ice region) of the current region and the corresponding ice condition core characteristics (such as average ice thickness, ice type toughness, brittleness level, etc.); after receiving the identification result, first, the integrity and validity of the result are checked to confirm that there is no data missing or logical contradiction, and then the corresponding ice removal mode start instruction is generated according to the regional type: if it is the northern cold heavy ice region, the instruction contains the start signal of the shovel impact ice removal mode and the initial moving path planning of the ice removal device (based on the thick ice section distribution in the ice coating feature set); if it is the southern light ice region, the instruction contains the start signal of the vibration ice removal mode and the initial positioning coordinates of the ice removal device (based on the thin ice section distribution in the ice coating feature set).

[0087] Step 4.2, in the shovel impact ice removal mode, control the ice removal device to move to the target position along the power transmission line, and obtain the ice layer surface image in real time through the visual sensor, identify the ice layer thickness distribution and adhesion strength characteristics, specifically including: after the shovel impact ice removal mode is started, according to the initial moving path planning in the instruction, send a control signal to the walking mechanism of the ice removal device, drive the device to move at a constant speed along the power transmission line, receive the position information feedback by the device in real time during the moving process, and compare with the preset path to ensure that the device does not deviate from the line trajectory; when the device approaches the thick ice target section marked in the ice coverage feature set, control the visual sensor carried by the device to start working, continuously shoot high-definition images of the line surface ice, analyze the boundary between the ice layer and the conductor through image gray scale comparison, calculate the ice layer thickness value at different positions, and obtain the ice layer thickness distribution map; at the same time, judge the adhesion strength characteristics of the ice layer by analyzing the tightness of the ice layer surface and the conductor and whether the ice layer edge is raised.

[0088] Step 4.3, according to the ice layer thickness distribution and adhesion strength characteristics, determine the impact angle and action strength parameters of the shovel impact mechanism, and drive the shovel impact mechanism to implement high-intensity impact removal of the ice coverage, specifically including: receiving the ice layer thickness distribution and adhesion strength characteristic data, first classifying the data: for the ice coverage section with large thickness and high adhesion strength, determine that higher impact strength is needed to break through the ice layer adhesion; for the section with small thickness but tight adhesion, appropriately reduce the impact strength to avoid damaging the conductor; combine the arc parameters of the power transmission line conductor to determine the impact angle of the shovel impact mechanism, ensure that the shovel impact head and the conductor surface form a preset safe included angle, which not only ensures that the impact energy is effectively applied to the ice layer, but also avoids directly scratching the conductor surface; according to the above analysis results, obtain the impact angle and action strength parameter instructions of the shovel impact mechanism, drive the shovel impact mechanism to reciprocate according to the set parameters, and implement high-intensity impact on the ice coverage; continuously receive the ice layer shedding condition feedback by the visual sensor during the impact process, if it is found that part of the ice layer has not been removed, adjust the impact parameters in time until the ice coverage in the section is removed.

[0089] Step 4.4, in the vibration ice removal mode, control the ice removal device to be positioned to the ice coverage section, and monitor the ice layer brittleness characteristics and broken state through the visual sensor, specifically including: after the vibration ice removal mode is started, according to the initial positioning coordinates in the instruction, control the walking mechanism of the ice removal device to accurately move to the thin ice section marked in the ice coverage feature set, fix the device at the current line position after reaching the target position to ensure the stability of the vibration operation; then start the visual sensor, focus on the line ice area in the device operation range, shoot the ice layer surface state image in real time, judge the ice layer brittleness level by analyzing whether there are micro cracks on the ice layer surface, crack propagation speed and other characteristics, where more cracks and faster crack propagation indicate high brittleness of the ice layer; at the same time, identify the broken state of the ice layer by comparing the coverage area change of the ice layer in the continuous frame images, and calculate the proportion of the ice layer that has been removed in the total ice coverage area.

[0090] Step 4.5, based on the real-time feedback of the ice layer brittleness characteristics and the breaking state, dynamically adjusting the vibration frequency and the action amplitude to effectively break and remove the ice in real time, specifically including: based on the ice layer brittleness characteristics and the breaking state data, dynamically adjusting the vibration parameters: if the ice layer brittleness is high and the breaking state is good (the proportion of ice layer that has fallen off reaches a preset threshold), appropriately reducing the vibration frequency and the action amplitude to avoid excessive vibration energy causing conductor fatigue damage; if the ice layer brittleness is low and the breaking state is poor (the proportion of ice layer that has fallen off is lower than the preset threshold), slightly increasing the vibration frequency within a preset safe range, or slightly adjusting the action amplitude to enhance the vibration energy transmission efficiency and promote ice layer breaking; during the adjustment process, the visual sensor continuously monitors the ice layer state changes, and every interval of a preset time, the updated breaking state data is fed back, and the vibration parameters are optimized again according to the feedback data, forming a real-time closed loop of monitoring, adjusting and re-monitoring, until the ice in the working range is completely broken and falls off.

[0091] In the embodiment of the application, by sending corresponding deicing mode starting instructions according to the area type, the problem of deicing mode solidification is overcome; in the shovel impact mode, the control device is moved and the ice layer characteristics are determined by combining visual recognition and impact parameters are determined to remove thick ice, solving the problem of incomplete deicing of thick ice in the north with high toughness; in the vibration mode, the ice covered section is positioned and the vibration parameters are adjusted according to the real-time feedback to break thin ice, avoiding damage to the line covered by thin ice with brittleness in the south; finally, differentiated and accurate deicing is realized, and the deicing efficiency and line protection effect are improved.

[0092] In a preferred embodiment of the application, the above step 5 can include:

[0093] Step 5.1, during the shovel impact deicing or vibration deicing operation, continuously acquiring ice layer image sequences in the working area by the visual sensor, and acquiring residual ice layer thickness data, specifically including: starting the visual sensor carried by the deicing device when the shovel impact deicing or vibration deicing operation is carried out, continuously shooting images of the line ice in the working area at a collection frequency matching the deicing operation speed (such as collecting one frame every 10 seconds), forming continuous ice layer image sequences, ensuring that the images can completely cover the line section of the current operation without missing the local ice state; at the same time, real-time acquisition of the thickness data of the residual ice layer in the working area; in the shovel impact deicing mode, recording the residual thickness at the corresponding position after each shovel impact action; in the vibration deicing mode, dynamically acquiring the residual thickness along with the ice layer breaking process, and associating the acquired residual ice layer thickness data with the corresponding image frame.

[0094] Step 5.2 involves real-time analysis of the ice layer image sequence to identify the trend of ice layer coverage changes and the exposed state of the line surface. Combined with the remaining ice layer thickness data, an evaluation index for the de-icing effect is obtained. Specifically, this includes: transmitting and processing the ice layer image sequence; comparing the coverage area of ​​ice in consecutive frames (e.g., the area of ​​ice covering the conductor surface) to calculate the rate of change of the ice coverage area, thereby identifying the trend of ice layer coverage changes; if the coverage area continuously decreases and the rate of decrease is stable, it indicates that the de-icing effect is progressing steadily; if the coverage area decreases slowly or stagnates, it indicates that there may be obstacles in the de-icing process; simultaneously capturing the exposed features of the line surface (e.g., the metallic texture or color of the conductor itself), and statistically analyzing the area ratio of the exposed area on the line surface to determine the exposed state of the line surface; subsequently, combining the synchronously acquired remaining ice layer thickness data, calculating the average thickness and maximum remaining thickness of the remaining ice layer in the work area, and integrating the rate of change of ice layer coverage area, the proportion of the exposed area on the line surface, the average thickness of the remaining ice layer, and the maximum thickness of the remaining ice layer into a complete set of evaluation indexes for the de-icing effect.

[0095] Step 5.3: Based on the de-icing effect evaluation indicators, determine whether the actual effect of the current de-icing operation meets the preset removal standards, and update the de-icing effect information in real time. Specifically, this includes: first, setting preset removal standards according to the safe operation requirements of transmission line de-icing (e.g., the exposed area of ​​the line surface is not less than 95%, the average thickness of the remaining ice layer does not exceed 2mm, and the maximum thickness of the remaining ice layer does not exceed 5mm); then, comparing the de-icing effect evaluation indicators with the preset removal standards item by item: if all evaluation indicators meet or exceed the preset standards, the actual effect of the current de-icing operation is determined to be up to standard; if any indicator does not meet the preset standards (e.g., the exposed area is only 80%, or the average thickness of the remaining ice layer is 3mm), the de-icing effect is determined to be down to standard; after completing the effect judgment, update the de-icing effect information in real time, including the location of the current work section, details of the evaluation indicators that have met or failed to meet the standards, and the remaining de-icing workload.

[0096] In this embodiment of the invention, by continuously collecting ice layer image sequences and remaining thickness data during de-icing operations, the problem of lacking real-time data support in the de-icing process is overcome, providing a basis for effect evaluation; by analyzing the image sequences and combining them with thickness data, de-icing effect evaluation indicators are obtained, solving the problem of lacking quantitative evaluation criteria for de-icing effects; based on the indicators, it is determined whether the preset clearing standard has been met and the effect information is updated, avoiding the situation where the progress of the operation cannot be grasped in real time, and ultimately achieving dynamic and precise control of the de-icing effect.

[0097] In a preferred embodiment of the present invention, step 6 above may include:

[0098] Step 6.1, receiving deicing effect information, extracting deicing effect evaluation indexes and real-time operation data therefrom, specifically including: receiving deicing effect information, which contains two parts of deicing effect evaluation indexes and real-time operation data; wherein the deicing effect evaluation indexes cover the change rate of ice layer coverage area in the operation area, the proportion of exposed area on the line surface, the average thickness and the maximum thickness of the remaining ice layer; the real-time operation data cover the currently enabled deicing mode, the execution parameters (such as the impact strength of the impact mechanism, the vibration frequency of the vibration mechanism, and the operation speed of the deicing device) in the current mode, and the real-time ice condition feedback (such as the residual state after the ice layer is broken) of the operation area; performing data analysis on the received deicing effect information, classifying and extracting the deicing effect evaluation indexes and real-time operation data and storing them.

[0099] Step 6.2, judging whether the current deicing operation efficiency meets the preset removal standard based on the deicing effect evaluation indexes and real-time operation data, specifically including: calling the preset removal standard, which is set based on the safe operation requirements of the power transmission line, specifically including that the proportion of exposed area on the line surface is not less than a preset threshold, the average thickness of the remaining ice layer is not more than a preset thickness value, the maximum thickness of the remaining ice layer is not more than a preset upper limit value, and the execution parameters in the real-time operation data are within the safe operation range of the equipment (such as the impact strength is not more than the withstand limit of the conductor, and the vibration frequency does not cause conductor resonance); comparing the extracted deicing effect evaluation indexes with the preset removal standard one by one, if all evaluation indexes meet the standard requirements and the real-time operation parameters do not exceed the safe range, it is determined that the current deicing operation efficiency meets the standard; if any evaluation index does not meet the standard (such as the proportion of exposed area is lower than the threshold, or the thickness of the remaining ice layer exceeds the standard), or the real-time operation parameters are close to the safety critical value, it is determined that the current deicing operation efficiency does not meet the preset removal standard.

[0100] Step 6.3, if the work performance does not reach the set standard, the source of deicing resistance is analyzed according to the current deicing process state to identify whether it is parameter mismatch or mode mismatch, specifically including: when it is determined that the work performance does not meet the standard, the deicing process state record data (including ice layer change video during work, execution parameter adjustment record, ice condition feedback log) is called, the current deicing mode is combined with the work area ice condition characteristics to analyze the deicing resistance source; if the current deicing mode matches the region type (such as enabling the shovel strike deicing mode in the northern cold heavy ice region, and enabling the vibration deicing mode in the southern light ice region), but the ice layer removal efficiency is low (such as thick ice remaining after shovel strike, and ice layer not fully broken after vibration), and through parameter record it is found that the execution parameter is fixed for a long time without adjustment, then it is judged that the resistance source is parameter mismatch; if the current deicing mode does not match the region type (such as using the vibration deicing mode in the northern cold heavy ice region, resulting in thick ice unable to be broken, and using the shovel strike deicing mode in the southern light ice region, resulting in conductor skin damaged), or the mode matches but the ice layer characteristics temporarily change (such as thick ice appearing locally in the southern light ice region), resulting in the current mode being unable to effectively remove the ice, then it is judged that the resistance source is mode mismatch.

[0101] Step 6.4, if it is parameter mismatch, the execution parameter of the current deicing mode is dynamically adjusted, the execution parameter including impact strength, vibration frequency or work speed, specifically including: if it is determined that the resistance source is parameter mismatch, the parameter adjustment scheme is formulated according to the current deicing mode and the evaluation index that does not meet the standard; if it is the shovel strike deicing mode, and the reason for not meeting the standard is that the residual ice layer thickness is over-standard, and there is more ice layer remaining after impact, then the impact strength of the shovel strike mechanism is appropriately increased within the safety range of the equipment, and the work speed of the deicing device is reduced, to ensure that the shovel strike mechanism has enough time to apply continuous impact to thick ice; if it is the vibration deicing mode, and the reason for not meeting the standard is that the ice layer is not fully broken, and the exposed area ratio is low, then the vibration frequency of the vibration mechanism is slightly increased, or the vibration amplitude is finely adjusted, to enhance the effect of vibration energy on the ice layer; during the parameter adjustment process, the change trend of the deicing effect evaluation index is associated in real time, after adjusting the parameter once, new deicing effect information is collected after waiting for a work cycle, to avoid work instability caused by frequent parameter adjustment.

[0102] Step 6.5, if the mode mismatch is determined, a mode switching instruction is sent to the deicing device, the execution mechanism of the current deicing mode is stopped, and the preparation process of another deicing mode is started; after the mode switching is completed, the working parameters are reinitialized based on the real-time icing conditions (such as the remaining ice layer thickness, ice layer toughness and brittleness characteristics) of the current working area: when switching to the shovel impact deicing mode, the initial impact strength is set according to the remaining ice layer thickness, and the impact angle is set in combination with the conductor arc; when switching to the vibration deicing mode, the initial vibration frequency is set according to the brittleness characteristics of the ice layer, and the vibration amplitude is set with reference to the broken state; after the initialization is completed, steps 6.1 to 6.4 are re-executed, that is, the new deicing effect information is received, the working efficiency is judged, the resistance source is analyzed, the parameters are adjusted or the mode is switched, a closed-loop adjustment mechanism based on deicing effect feedback is formed, and the cycle is continued until the ice on the working area is completely removed, and all deicing effect evaluation indexes meet the preset removal standard.

[0103] In the embodiment of the application, by receiving deicing effect information and extracting evaluation indexes and real-time working data, the problem of lack of data support in adjustment is overcome; according to the judgment of whether the working efficiency meets the preset removal standard, the problem of inability to accurately measure the working quality is solved; when the standard is not met, the deicing resistance source is analyzed to distinguish between parameter mismatch and mode mismatch, and blind adjustment is avoided; then the current mode parameters are adjusted or another mode is switched and the working parameters are reinitialized, and finally a closed-loop adjustment mechanism based on effect feedback is obtained, which overcomes the defects of incomplete deicing or excessive working, ensures that the ice is completely removed, and improves the reliability and efficiency of deicing.

[0104] As shown in Figure 2 , the embodiment of the application also provides a deicing control system for a power transmission line in a cold region, which comprises:

[0105] An acquisition module is configured to collect icing data of the power transmission line, the icing data being multi-source data including ice layer thickness, ice type category and distribution characteristics, and the spatial correlation region for feature analysis is constructed by dynamically selecting a plurality of spatial sampling positions on the target line in combination with environmental temperature, humidity and geographical location information;

[0106] An analysis module is configured to obtain correction parameters based on the morphological characteristics of the spatial correlation region, to perform spatial superposition and feature fusion analysis on the multi-source data according to the correction parameters, and to extract an icing feature set that reflects the overall icing state of the line after dynamic correction;

[0107] The identification module is configured to identify whether the current area belongs to a northern heavy-icing area or a southern light-icing area according to the icing feature set, wherein the northern heavy-icing area is mainly dominated by thick icing and ductile ice, and the southern light-icing area is mainly dominated by thin icing and brittle ice;

[0108] The execution module is configured to start a shovel-impact deicing mode if the current area is identified as the northern heavy-icing area, to control the deicing device to move along the line and be positioned, to identify the ice layer structure and adhesion state in real time in combination with the visual perception information, and to implement high-intensity impact deicing through the shovel-impact mechanism; and to start a vibration deicing mode if the current area is identified as the southern light-icing area, to dynamically adjust vibration parameters according to the visual feedback, and to control the deicing device to break and remove the icing in a high-frequency vibration manner.

[0109] The adjustment module is configured to continuously acquire deicing effect information during the deicing mode, to dynamically feedback and adjust the deicing action parameters based on the deicing effect information, and to adaptively optimize the execution parameters of the current deicing mode or enter another deicing mode when the deicing effect does not reach a set standard, until the icing is completely removed, so as to realize closed-loop adaptive deicing.

[0110] The above describes the preferred embodiments of the present application. It should be noted that those skilled in the art can make some improvements and refinements without departing from the principles of the present application, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method of de-icing control of a power transmission line in a cold region, characterized by, The method comprises: Collecting icing data of the power transmission line, the icing data being multi-source data including ice layer thickness, ice type category and distribution characteristics, combining with environmental temperature, humidity and geographical location information, dynamically selecting a plurality of spatial sampling positions on the target line to construct a spatial correlation region for feature analysis, comprising: acquiring the ice layer thickness, ice type category and their distribution information along the line through the sensor device deployed along the power transmission line, synchronously collecting the environmental temperature, humidity and geographical location parameters to obtain a multi-source monitoring data set; dynamically determining a plurality of representative spatial sampling positions on the target line based on the icing thickness variation trend, ice type distribution characteristics and environmental temperature gradient and humidity variation characteristics contained in the multi-source monitoring data set, each sampling position being adaptively selected according to the continuity of the icing distribution and the regional regularity of the environmental influence; constructing a spatial correlation region for icing feature extraction based on the spatial relationship of the spatial sampling positions; analyzing the morphological characteristics of the spatial correlation region to obtain compensation parameters for data correction; performing spatial superposition and feature fusion processing on the multi-source monitoring data set based on the compensation parameters to obtain an icing feature set that is preliminarily dynamically corrected; Based on the morphological characteristics of the spatial correlation region, the correction parameters are obtained, and the multi-source data is subjected to spatial superposition and feature fusion analysis according to the correction parameters, so as to extract an icing feature set that is dynamically corrected and reflects the overall icing state of the line; According to the icing feature set, it is identified whether the current region belongs to a northern cold and heavy icing area or a southern light icing area; If it is identified as a northern cold and heavy icing area, a shovel impact deicing mode is started, the deicing device is controlled to move and position along the line, the ice layer structure and adhesion state are identified in real time in combination with visual perception information, and high-intensity impact deicing is implemented through a shovel impact mechanism; if it is identified as a southern light icing area, a vibration deicing mode is started, vibration parameters are dynamically adjusted according to visual feedback, and the deicing device is controlled to break and remove the icing in a high-frequency vibration mode; In the deicing mode process, deicing effect information is continuously obtained; Based on the deicing effect information, the deicing action parameters are dynamically feedback adjusted; when the deicing effect does not reach the set standard, the execution parameters of the current deicing mode are adaptively optimized or another deicing mode is entered, until the icing is completely removed, so as to realize closed-loop adaptive deicing.

2. The ice-mitigation control method for a power transmission line in a cold region according to claim 1, characterized by, Based on the morphological characteristics of the spatial correlation region, the correction parameters are obtained, and the multi-source data is subjected to spatial superposition and feature fusion analysis according to the correction parameters, so as to extract an icing feature set that is dynamically corrected and reflects the overall icing state of the line, comprising: Obtaining the contour characteristics and spatial distribution attributes of the spatial correlation region, analyzing the morphological variation trend and structural stability thereof; According to the morphological variation trend and structural stability, correction parameters for correcting the spatial deviation of the multi-source monitoring data are obtained; Based on the correction parameters, the ice layer thickness, ice type category and distribution characteristics in the multi-source monitoring data set are subjected to spatial registration and weight adjustment; The multi-source data subjected to spatial registration and weight adjustment is subjected to spatial superposition and feature fusion to extract a feature vector that can reflect the overall icing state of the line; Based on the feature vector, an icing feature set that is deeply dynamically corrected is constructed.

3. The ice-mitigation control method for a power transmission line in a cold region according to claim 2, characterized by, According to the ice cover feature set, it is identified whether the current area belongs to the northern high-cold heavy ice area or the southern light ice area, including: Extracting ice thickness distribution features and ice type mechanical property features from the depth-dynamically corrected ice cover feature set; Judging the overall ice load level based on the ice thickness distribution features, and analyzing the ice layer anti-breaking capacity in combination with the ice type mechanical property features; Establishing regional type discrimination rules according to the combined features of the ice load level and the ice layer anti-breaking capacity; According to the regional type discrimination rules, the current area is identified as the northern high-cold heavy ice area characterized by high load and high toughness, or as the southern light ice area characterized by low load and brittleness.

4. The ice-mitigation control method for a power transmission line in a cold region according to claim 3, characterized by, If it is identified as the northern high-cold heavy ice area, the shovel-impact deicing mode is started, the deicing device is controlled to move and position along the line, the ice layer structure and adhesion state are identified in real time in combination with visual perception information, and high-strength impact deicing is implemented through the shovel-impact mechanism; if it is identified as the southern light ice area, the vibration deicing mode is started, the vibration parameters are dynamically adjusted according to the visual feedback, and the deicing device is controlled to break and remove the ice cover in a high-frequency vibration mode, including: Receiving the regional type identification result, if it is identified as the northern high-cold heavy ice area, a shovel-impact deicing mode start instruction is sent to the deicing device; if it is identified as the southern light ice area, a vibration deicing mode start instruction is sent; In the shovel-impact deicing mode, the deicing device is controlled to move to the target position along the power transmission line, the ice layer surface image is acquired in real time through the visual sensor, and the ice layer thickness distribution and adhesion strength features are identified; According to the ice layer thickness distribution and adhesion strength features, the impact angle and action strength parameters of the shovel-impact mechanism are determined, and the shovel-impact mechanism is driven to implement high-strength impact removal of the ice cover; In the vibration deicing mode, the deicing device is positioned to the ice-covered section, and the ice layer brittleness features and breaking state are monitored through the visual sensor; Based on the real-time feedback of the ice layer brittleness features and breaking state, the vibration frequency and action amplitude are dynamically adjusted to effectively break and remove the ice cover in real time.

5. The ice-mitigation control method for a power transmission line in a cold region according to claim 4, characterized by, In the deicing mode process, deicing effect information is continuously acquired, including: In the process of shovel-impact deicing or vibration deicing operation, ice layer image sequences of the operation area are continuously acquired through the visual sensor, and residual ice layer thickness data are also acquired; Real-time analysis is performed on the ice layer image sequences to identify the ice layer coverage change trend and the line surface bare state, and deicing effect evaluation indexes are obtained in combination with the residual ice layer thickness data; Based on the deicing effect evaluation indexes, it is judged whether the actual effect of the current deicing operation reaches the preset removal standard, and the deicing effect information is updated in real time.

6. The ice-mitigation control method for a power transmission line in a cold region according to claim 5, characterized by, Based on the deicing effect information, the deicing action parameters are dynamically feedback-adjusted; when the deicing effect does not reach the set standard, the execution parameters of the current deicing mode are adaptively optimized or another deicing mode is entered until the ice cover is completely removed, so as to realize closed-loop adaptive deicing, including: Step 6.1, receiving the deicing effect information, and extracting the deicing effect evaluation indexes and real-time operation data therefrom; Step 6.2, judging whether the current deicing operation efficiency reaches the preset removal standard based on the deicing effect evaluation indexes and the real-time operation data; Step 6.3, if the job performance does not meet the set standard, analyze the deicing resistance source according to the current deicing process state, identify whether it is parameter mismatch or mode mismatch; Step 6.4, if it is parameter mismatch, dynamically adjust the execution parameters of the current deicing mode, including impact strength, vibration frequency or operation speed; Step 6.5, if it is mode mismatch, switch to another deicing mode, and reinitialize the operation parameters based on the current ice condition; reexecute steps 6.1 to 6.4 to obtain a closed-loop adjustment mechanism based on deicing effect feedback until the ice is completely removed.

7. A de-icing control system for a power transmission line in a cold region, the system implementing the method of any one of claims 1 to 6, characterized in that, Comprise: An acquisition module is configured to collect icing data of a power transmission line, the icing data being multi-source data including ice layer thickness, ice type category and distribution characteristics, combined with environmental temperature, humidity and geographic location information, a plurality of spatial sampling positions are dynamically selected on the target line to construct a spatial correlation region for feature analysis; An analysis module is configured to obtain correction parameters based on the morphological characteristics of the spatial correlation region, and perform spatial superposition and feature fusion analysis on the multi-source data according to the correction parameters to extract an icing feature set reflecting the overall icing state of the line after dynamic correction; An identification module is configured to identify whether the current region belongs to a northern cold heavy ice area or a southern light ice area according to the icing feature set, wherein the northern cold heavy ice area is mainly composed of thick ice and ductile ice type, and the southern light ice area is mainly composed of thin ice and brittle ice type; An execution module is configured to start a shovel impact deicing mode if the region is identified as a northern cold heavy ice area, control the deicing device to move and position along the line, combine visual perception information to identify the ice layer structure and adhesion state in real time, and implement high-intensity impact deicing through a shovel impact mechanism; if the region is identified as a southern light ice area, a vibration deicing mode is started, vibration parameters are dynamically adjusted according to visual feedback, and the deicing device is controlled to break and remove the ice in a high-frequency vibration mode; An adjustment module is configured to continuously obtain deicing effect information during the deicing process; Based on the deicing effect information, the deicing action parameters are dynamically feedback adjusted; when the deicing effect does not meet the set standard, the execution parameters of the current deicing mode are adaptively optimized or another deicing mode is entered until the ice is completely removed, so as to realize closed-loop adaptive deicing.

Images