A high-voltage large-capacity mobile line ice melting dynamic control method and system

Abstract

The present application relates to the technical field of power system operation safety, and particularly relates to a line ice-melting dynamic control method and system based on high-voltage large-capacity movement, comprising real-time collection of icing state multi-dimensional feature vectors; determination of local icicle area of composite core conductor based on the icing state multi-dimensional feature vectors, and obtaining of local icicle morphology features according to temperature gradient deviation and impedance spectrum feature deviation of the local icicle area; integral calculation of the local icicle morphology features along the length direction of the composite core conductor to obtain icicle distribution parameters, and calculation of local conductor thermoelectric characteristic parameters according to the icicle distribution parameters; construction of a nonlinear local thermoelectric calibration model by using the least square method based on the local conductor thermoelectric characteristic parameters and actual thermoelectric measurement values; and obtaining of a target ice-melting current waveform by using the nonlinear local thermoelectric calibration model. The present application realizes rapid and safe ice-melting of local icing of high-voltage large-capacity composite core conductor through multi-dimensional data and target ice-melting current waveform optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system operation safety, and in particular to a line ice melting dynamic control method and system based on high-voltage large-capacity movement. BACKGROUND

[0002] In a high-voltage large-capacity power transmission system, the problem of conductor icing caused by extreme icing weather has become a core challenge to the safe and stable operation of the power grid. With the intensification of climate change, the local icing phenomenon of new transmission lines such as composite core conductors in complex terrain (such as narrow wind gaps) is becoming more and more serious, and the dynamic characteristics such as rapid growth of ice rib, sudden change of conductor sag, and nonlinear change of thermal-electric characteristics significantly increase the difficulty of ice melting control.

[0003] Traditional ice melting technology mainly relies on fixed DC ice melting devices or single sensor monitoring (such as only using distributed optical fiber temperature measurement or high-frequency impedance spectrum analysis), which is difficult to adapt to the dynamic icing characteristics of composite core conductors in high-voltage large-capacity mobile ice melting scenarios. First, the heterogeneous data streams (such as temperature gradient, impedance characteristics, or sag change) generated by the sensing systems such as distributed optical fiber temperature measurement, high-frequency impedance spectrum analysis, and laser sag measurement have space-time asynchronous problems. Especially when the local ice rib is formed, the temperature field presents nonlinear anomalies, and the impedance characteristic calculation complexity increases dramatically. However, the traditional data fusion method cannot dynamically balance the real-time and accuracy of multi-source information. Second, the thermal-electric characteristics of the conductor change dynamically with the conditions such as ice thickness, environmental temperature, and wind speed. The nonlinear thermal-electric characteristics of the composite core conductor exacerbate the difficulty of model calibration. However, the traditional fixed control model cannot capture the evolution law of the nonlinear thermal-electric characteristics of the local area (such as the wind gap) of the composite core conductor in real time, resulting in model parameter misalignment under local icing conditions. When the laser sensor detects a dramatic change in sag, the existing method cannot complete model recalibration in time due to limited computing resources, and can only use conservative ice melting current parameters, resulting in high energy consumption or incomplete ice melting. In addition, the existing mobile ice melting device mostly uses preset waveform templates, which cannot be optimized in real time according to the ice thickness distribution and dynamic temperature rise characteristics of the conductor. Especially in the local icing scenario of composite core conductors, the rapid evolution of ice rib morphology requires the control system to adjust the pulse amplitude, frequency, and duration synchronously. However, the traditional method lacks closed-loop feedback of multi-source sensing input and dynamic model, which easily leads to local overheating of the conductor or reduces the ice melting efficiency.

[0004] In summary, the existing ice melting technology has many defects, resulting in low ice melting efficiency and increased safety risk of the conductor, and is difficult to adapt to the dynamic changes of local icing. Therefore, how to realize efficient and safe local icing control of composite core conductors has become a key problem to be solved in this field. SUMMARY

[0005] To solve the above technical problems, the application provides a line ice melting dynamic control method and system based on high-voltage large-capacity movement.

[0006] In a first aspect, the application provides a line ice melting dynamic control method based on high-voltage large-capacity movement, which comprises the following steps:

[0007] Real-time collection of conductor surface temperature distribution data, conductor impedance frequency spectrum data and conductor sag data of the composite core conductor to form an icing state multi-dimensional feature vector;

[0008] Determination of a local ice ridge area on the composite core conductor based on the icing state multi-dimensional feature vector, and obtaining of local ice ridge morphological characteristics according to a temperature gradient deviation amount and an impedance frequency spectrum characteristic deviation amount of the local ice ridge area;

[0009] Integral calculation of the local ice ridge morphological characteristics along the length direction of the composite core conductor to obtain ice ridge distribution parameters, and calculation of local conductor thermoelectric characteristic parameters according to the ice ridge distribution parameters and the conductor sag data;

[0010] Real-time measurement of thermoelectric actual measurement values of the composite core conductor, and construction of a nonlinear local thermoelectric calibration model by using a least square method based on the local conductor thermoelectric characteristic parameters and the thermoelectric actual measurement values;

[0011] Obtaining of a target ice melting current waveform by using the nonlinear local thermoelectric calibration model according to a composite core conductor icing state evaluation parameter;

[0012] Control of real-time ice melting current pulses output by a mobile high-voltage ice melting device according to the target ice melting current waveform, and dynamic adjustment of pulse output timing and energy distribution ratio of the composite core conductor by the real-time ice melting current pulses.

[0013] In a further embodiment, the step of determining a local ice ridge area on the composite core conductor based on the icing state multi-dimensional feature vector comprises:

[0014] Preprocessing of the icing state multi-dimensional feature vector to obtain normalized feature data, and calculation of Euclidean distances between the normalized feature data and preset non-icing state reference data to generate icing state deviation quantization values of each monitoring area;

[0015] Comparison of the icing state deviation quantization values with a preset icing deviation threshold value, and marking of a corresponding monitoring area as a potential icing area when the icing state deviation quantization value exceeds the preset icing deviation threshold value;

[0016] The K-means clustering algorithm is used to cluster and group the potential icing area, determine the potential ice ridge boundary position coordinates, and extract the ice ridge temperature gradient change rate and ice ridge impedance frequency spectrum features at the potential ice ridge boundary position coordinates;

[0017] The ice ridge feature correlation coefficient between the ice ridge temperature gradient change rate and the ice ridge impedance frequency spectrum features is calculated, and if the ice ridge feature correlation coefficient exceeds a preset correlation coefficient threshold, the corresponding potential ice ridge boundary position coordinates are determined as ice ridge area feature correlation points;

[0018] The ice ridge area feature correlation points are spatially interpolated using a pre-constructed conductor physical model to locate a local ice ridge area.

[0019] In further embodiments, the step of preprocessing the icing state multi-dimensional feature vector to obtain normalized feature data includes:

[0020] The sliding window method is used to perform time domain average filtering on the conductor surface temperature distribution data to obtain smoothed temperature distribution data;

[0021] First-order difference calculation is performed on the smoothed temperature distribution data in the conductor length direction to obtain temperature gradient data between adjacent monitoring areas;

[0022] Wavelet transform is used to perform multi-scale decomposition on the conductor impedance frequency spectrum data to extract impedance frequency spectrum feature vectors of each monitoring area; the impedance frequency spectrum feature vectors include resonance frequencies corresponding to impedance amplitude local maxima and impedance modulus values;

[0023] The smoothed temperature distribution data, the temperature gradient data, and the impedance frequency spectrum feature vectors are normalized to obtain normalized feature data of each monitoring area.

[0024] In further embodiments, the impedance frequency spectrum feature deviation includes resonance frequency deviation and impedance modulus value deviation, and the step of obtaining local ice ridge morphological features according to the temperature gradient deviation and the impedance frequency spectrum feature deviation of the local ice ridge area includes:

[0025] The temperature gradient deviation, resonance frequency deviation, and impedance modulus value deviation corresponding to the local ice ridge area are weighted and fused to obtain multi-source fusion deviation of each monitoring area;

[0026] The multi-source fusion deviation of each monitoring area is input into a pre-constructed ice ridge thickness prediction model to obtain ice ridge thickness distribution data;

[0027] According to the ice ridge thickness distribution data, a continuous coverage area boundary coordinate of which the ice ridge thickness is greater than a preset thickness threshold is extracted, and a principal component analysis method is used to determine an ice ridge coverage main direction according to the continuous coverage area boundary coordinate; the ice ridge coverage main direction is a characteristic vector direction corresponding to a maximum eigenvalue of a covariance matrix of the continuous coverage area boundary coordinate;

[0028] An ice ridge coverage angle is obtained by calculating an included angle between the ice ridge coverage main direction and an axial direction of the composite core conductor, and a projection integral of the ice ridge thickness distribution data along the ice ridge coverage main direction is calculated to obtain an ice ridge extension length;

[0029] The ice ridge thickness distribution data, the ice ridge coverage angle and the ice ridge extension length are integrated to obtain a local ice ridge morphological feature.

[0030] In further embodiments, the step of integrating the local ice ridge morphological feature along the length direction of the composite core conductor to obtain an ice ridge distribution parameter comprises:

[0031] The composite core conductor is divided into equally spaced conductor sections along the length direction of the composite core conductor, and the ice ridge thickness distribution data in each conductor section is segmented and integrated along the axial direction of the composite core conductor to generate an ice ridge thickness integral value of each conductor section;

[0032] An ice ridge average thickness of each conductor section is obtained according to a ratio of the ice ridge thickness integral value to the length of the conductor section, and an average value of the ice ridge average thickness of all conductor sections is calculated to obtain an ice ridge basic thickness index;

[0033] An ice distribution uniformity index is obtained according to a relative deviation between the ice ridge average thickness of each conductor section and the ice ridge basic thickness index;

[0034] An ice ridge line density gradient of adjacent conductor sections is obtained according to a ratio between a difference value of the ice ridge average thickness of adjacent conductor sections and the length of the conductor section;

[0035] A modified ice ridge line density gradient is obtained by directionally modifying the ice ridge line density gradient according to the ice ridge coverage angle, and the ice distribution uniformity index and the modified ice ridge line density gradient are taken as the ice ridge distribution parameter.

[0036] In further embodiments, the step of calculating a local conductor thermoelectric characteristic parameter according to the ice ridge distribution parameter and the conductor sag data comprises:

[0037] The conductor sag data is filtered to obtain a conductor sag change amount, and the composite core conductor is divided into a plurality of equally spaced micro-segments along the axial direction;

[0038] According to the ice ridge distribution parameters and the composite core conductor material attributes, an ice mass-sag increment coupling equation of each micro-element section is established;

[0039] Based on the conductor sag variation, the ice mass-sag increment coupling equation is solved by an iterative method to obtain an equivalent temperature rise of each micro-element section;

[0040] According to the equivalent temperature rise, a linear interpolation method is used to calculate a reference dynamic resistivity of each micro-element section, and a skin effect correction factor is superimposed on the reference dynamic resistivity of the icing section to obtain a corrected resistivity;

[0041] Based on the equivalent temperature rise and environmental parameters, an equivalent thermal conductivity and an equivalent heat capacity of the composite core conductor are calculated, and the corrected resistivity, the equivalent thermal conductivity and the equivalent heat capacity are standardized to obtain local conductor thermoelectric characteristic parameters.

[0042] In further embodiments, the step of constructing a nonlinear local thermoelectric calibration model based on the local conductor thermoelectric characteristic parameters and the actual thermoelectric measurement values using a least squares method comprises:

[0043] According to the local conductor thermoelectric characteristic parameters and the actual thermoelectric measurement values, a conductor theoretical temperature distribution is calculated;

[0044] The conductor theoretical temperature distribution is spatiotemporally aligned with the measured temperature data to obtain temperature deviation distribution data;

[0045] With minimizing the root mean square error of the temperature deviation distribution data as the optimization goal, a least squares method is used to globally iteratively calibrate a heat exchange coefficient to obtain a calibrated heat exchange coefficient; the heat exchange coefficient includes an air convection heat exchange coefficient and a radiation heat exchange coefficient;

[0046] According to the temperature deviation distribution data, a nonlinear temperature effect of the composite core conductor material attributes is analyzed to obtain conductor material attribute correction factors in different temperature intervals;

[0047] According to the conductor material attribute correction factors, the local conductor thermoelectric characteristic parameters are corrected to obtain calibrated conductor thermoelectric characteristic parameters;

[0048] According to the calibrated conductor thermoelectric characteristic parameters and the calibrated heat exchange coefficient, a nonlinear local thermoelectric calibration model is constructed.

[0049] In further embodiments, the step of obtaining a target ice melting current waveform according to the composite core conductor icing state evaluation parameters using the nonlinear local thermoelectric calibration model comprises:

[0050] The icing state multi-dimensional feature vector is analyzed by using a principal component analysis method to obtain an icing state evaluation parameter of the composite core conductor;

[0051] The non-linear local thermal-electric calibration model is compensated on-line according to the icing state evaluation parameter of the composite core conductor to obtain a conductor ice-melting thermal-electric control model meeting the current icing condition;

[0052] According to the ice thickness distribution and the conductor material characteristics, the conductor ice-melting thermal-electric control model is used to calculate an initial ice-melting current waveform of the local ice ridge area of the composite core conductor;

[0053] The initial ice-melting current waveform is adjusted according to the real-time monitored environmental temperature data and ice layer state data to obtain a target ice-melting current waveform of the local ice ridge area of the composite core conductor.

[0054] In a further implementation, the target ice-melting current waveform includes a target ice-melting current frequency and a target ice-melting current amplitude.

[0055] In a second aspect, the application provides a line ice-melting dynamic control system based on high-voltage and large-capacity movement, which comprises:

[0056] A data acquisition module is configured to collect conductor surface temperature distribution data, conductor impedance spectrum data and conductor sag data of the composite core conductor in real time to form an icing state multi-dimensional feature vector;

[0057] An ice ridge analysis module is configured to determine a local ice ridge area on the composite core conductor based on the icing state multi-dimensional feature vector and to obtain a local ice ridge morphological feature according to a temperature gradient deviation amount and an impedance spectrum characteristic deviation amount of the local ice ridge area;

[0058] A thermal-electric analysis module is configured to integrate the local ice ridge morphological feature along a length direction of the composite core conductor to obtain an ice ridge distribution parameter and to calculate a local conductor thermal-electric characteristic parameter according to the ice ridge distribution parameter and the conductor sag data;

[0059] A model construction module is configured to measure a thermal-electric actual measurement value of the composite core conductor in real time and to construct a non-linear local thermal-electric calibration model by using a least square method based on the local conductor thermal-electric characteristic parameter and the thermal-electric actual measurement value;

[0060] A waveform generation module is configured to obtain a target ice-melting current waveform by using the non-linear local thermal-electric calibration model according to an icing state evaluation parameter of the composite core conductor;

[0061] The ice melting control module is used for controlling the mobile high-voltage ice melting device to output real-time ice melting current pulses according to the target ice melting current waveform, and dynamically adjusting the pulse output timing and energy distribution ratio of the composite core conductor through the real-time ice melting current pulses.

[0062] The application provides a high-voltage and large-capacity mobile line ice melting dynamic control method and system, which comprises the following steps: collecting the conductor surface temperature distribution data, the conductor impedance frequency spectrum data and the conductor sag data of the composite core conductor in real time to form an icing state multi-dimensional feature vector; determining the local ice ridge area on the composite core conductor based on the icing state multi-dimensional feature vector, and obtaining the local ice ridge morphological feature according to the temperature gradient deviation and the impedance frequency spectrum characteristic deviation of the local ice ridge area; integrating and calculating the local ice ridge morphological feature along the length direction of the composite core conductor to obtain the ice ridge distribution parameter, and calculating the local conductor thermoelectric characteristic parameter according to the ice ridge distribution parameter and the conductor sag data; constructing a nonlinear local thermoelectric calibration model by using the least square method based on the local conductor thermoelectric characteristic parameter and the actual thermoelectric measurement value; obtaining the target ice melting current waveform by using the nonlinear local thermoelectric calibration model according to the composite core conductor icing state evaluation parameter; controlling the mobile high-voltage ice melting device to output real-time ice melting current pulses according to the target ice melting current waveform, and dynamically adjusting the pulse output timing and energy distribution ratio of the composite core conductor through the real-time ice melting current pulses. Compared with the prior art, the method realizes the accurate positioning of the local ice ridge area of the high-voltage and large-capacity transmission line under complex working conditions and the adaptive ice melting current output control by using multi-dimensional data and target ice melting current waveform optimization, realizes the rapid and safe ice melting of the local icing of the high-voltage and large-capacity composite core conductor, and ensures the safe and stable operation of the power grid under extreme freezing weather. BRIEF DESCRIPTION OF DRAWINGS

[0063] Figure 1 It is the flowchart of the high-voltage and large-capacity mobile line ice melting dynamic control method provided by the embodiment of the application.

[0064] Figure 2 It is the block diagram of the high-voltage and large-capacity mobile line ice melting dynamic control system provided by the embodiment of the application.

[0065] Legend: 101, data acquisition module; 102, ice ridge analysis module; 103, thermoelectric analysis module; 104, model construction module; 105, waveform generation module; 106, ice melting control module. DETAILED DESCRIPTION

[0066] The embodiments of the present application are described below in detail with reference to the accompanying drawings, the embodiments are given only for illustrative purposes and cannot be understood as limiting the present application, the accompanying drawings are only for reference and illustration and do not constitute a limitation on the patent protection scope of the present application, because many changes can be made to the present application without departing from the spirit and scope of the present application.

[0067] Reference Figure 1 The embodiments of the present application provide a line ice melting dynamic control method based on high-voltage and large-capacity movement, as shown in the figure, the method comprises the following steps: Figure 1

[0068] S1. Real-time acquisition of conductor surface temperature distribution data, conductor impedance frequency spectrum data and conductor sag data of the composite core conductor, forming an icing state multi-dimensional feature vector.

[0069] In the present embodiment, distributed optical fiber temperature sensors, high-frequency impedance spectrum analyzers and laser sag measurement instruments are respectively installed in multiple monitoring areas of the composite core conductor, the distributed optical fiber temperature sensors are used to obtain conductor axial surface temperature distribution data at a sampling rate of 1 Hz, the high-frequency impedance analyzers are used to collect conductor impedance frequency spectrum data at a sampling rate of 10 Hz, and the frequency scanning range can be set to 10 kHz-1 MHz; and the laser range finder is used to measure conductor sag data at a sampling rate of 0.5 Hz, etc. In the present embodiment, the positions and intervals can be pre-set according to the conductor length and the severity of icing and other factors, the temperature distribution data, impedance frequency spectrum data and sag data collected in the present embodiment are processed in time and space synchronously under a unified time reference by using a cubic spline interpolation method, an icing state multi-dimensional feature vector aligned in time and space is generated, the icing state multi-dimensional feature vector is preprocessed to obtain normalized feature data, and the process of preprocessing the icing state multi-dimensional feature vector comprises:

[0070] The sliding window method is used to perform time domain average filtering on the conductor surface temperature distribution data to obtain smooth temperature distribution data;

[0071] First-order difference calculation is performed on the smooth temperature distribution data in the conductor length direction to obtain temperature gradient data between adjacent monitoring areas;

[0072] Wavelet transform is used to perform multi-scale decomposition on the conductor impedance frequency spectrum data to extract impedance frequency spectrum feature vectors of each monitoring area; the impedance frequency spectrum feature vectors include resonance frequencies corresponding to impedance amplitude local maximum values and impedance modulus values;

[0073] The smooth temperature distribution data, the temperature gradient data and the impedance frequency spectrum feature vectors are normalized to obtain normalized feature data of each monitoring area.

[0074] ​Specifically, for the collected conductor surface temperature distribution data of each monitoring area, the embodiment can set the sliding window length according to the fluctuation of the data and the actual demand. For example, the embodiment sets the sliding window size to 5, that is, the conductor surface temperature distribution data of 5 consecutive sampling points are taken each time for sliding average filtering to eliminate high-frequency noise. When the window starts to slide, the embodiment can first calculate the average value of the first 5 temperature data points, replace the temperature value of the data point at the starting position of the window with the average value of the first 5 temperature data points, as the smoothed temperature data point, and then slide the window by one data point (that is, the sliding step is 1 each time). The above operations of calculating the average value and replacing are repeated, and the entire temperature distribution time sequence is processed in turn, so that the smoothed temperature distribution data is finally obtained, and the noise interference in the original temperature data is effectively reduced. At the same time, the embodiment performs first-order difference calculation on the smoothed temperature distribution data of adjacent monitoring points along the axial direction (the length direction) of the conductor to generate the temperature gradient data in the axial direction of the conductor.

[0075] For the collected conductor impedance spectrum data, the embodiment uses a Daubechies wavelet basis function to perform multi-scale decomposition on the conductor impedance spectrum data. The scale number of decomposition can be determined according to the frequency range and resolution requirement of the data. On each scale layer, the embodiment extracts the resonance frequency corresponding to the local maximum value of the impedance amplitude and the impedance modulus value by analyzing the details and the approximate part of the signal. For example, in the decomposition result of a certain scale layer, the position where the local maximum value of the impedance amplitude appears is found, and the frequency corresponding to the position is the resonance frequency. At the same time, the impedance modulus value at this time is recorded. The two values together constitute a feature point in the impedance spectrum feature vector of the monitoring area. After all the extracted feature points are sorted, the complete impedance spectrum feature vector is obtained. Finally, the embodiment performs maximum-minimum normalization processing on the smoothed temperature distribution data, the temperature gradient data and the impedance spectrum feature vector according to the monitoring area, to form the normalized feature data of each monitoring area.

[0076] S2. Determine the local icicle region on the composite core conductor based on the icing state multi-dimensional feature vector, and obtain the local icicle morphology feature according to the temperature gradient deviation and the impedance spectrum feature deviation of the local icicle region.

[0077] In some embodiments, the step of determining the local icicle region on the composite core conductor based on the icing state multi-dimensional feature vector comprises:

[0078] The icing state multi-dimensional feature vector is preprocessed to obtain normalized feature data, and the Euclidean distance between the normalized feature data and the preset non-icing state reference data is calculated to generate the icing state deviation quantization value of each monitoring area.

[0079] The icing state deviation quantitative value is compared with a preset icing deviation threshold value, and when the icing state deviation quantitative value exceeds the preset icing deviation threshold value, the corresponding monitoring area is marked as a potential icing area;

[0080] The potential icing areas are clustered and grouped by using a K-means clustering algorithm, potential ice ridge boundary position coordinates are determined, and an ice ridge temperature gradient change rate and an ice ridge impedance frequency spectrum feature at the potential ice ridge boundary position coordinates are extracted;

[0081] An ice ridge feature correlation coefficient between the ice ridge temperature gradient change rate and the ice ridge impedance frequency spectrum feature is calculated, and if the ice ridge feature correlation coefficient exceeds a preset correlation coefficient threshold value, the corresponding potential ice ridge boundary position coordinates are determined as ice ridge area feature correlation points;

[0082] The ice ridge area feature correlation points are spatially interpolated by using a pre-constructed conductor physical model, and a local ice ridge area is located.

[0083] Specifically, in the normal operation of the composite core conductor and in the no icing state, the temperature distribution, the temperature gradient and the impedance frequency spectrum feature data of the composite core conductor are pre-collected and normalized to obtain no icing state reference data. The no icing state reference data can include the temperature field, the gradient distribution and the impedance frequency spectrum feature under the normal working condition of the conductor. The corresponding no icing state reference data is called by real-time matching of current environmental parameters (temperature and humidity, wind speed), and based on the preset no icing state reference data, the Euclidean distance between the normalized feature data of each monitoring area and the no icing state reference data is calculated to generate an icing state deviation quantitative value reflecting the deviation degree of the current state from the no icing state. According to the statistical distribution characteristics of the icing state deviation quantitative value, a mean-standard deviation dynamic threshold algorithm is used to set an icing deviation threshold value. The icing state deviation quantitative values of all monitoring areas are traversed in the embodiment. If the icing state deviation quantitative value of a certain monitoring point exceeds the icing deviation threshold value, it is considered that icing may occur in the area, and the corresponding monitoring area is determined as a potential icing area. These potential icing areas may be ice ridge formation areas. If the icing state deviation quantitative value of a certain monitoring point is lower than the icing deviation threshold value, the corresponding monitoring area is determined as a no icing area.

[0084] Next, the embodiment takes the position coordinates of all potential icing areas in the length direction of the conductor as input features, uses the K-means clustering algorithm to cluster the coordinates of the potential icing areas, automatically determines the optimal cluster number by the contour coefficient method, and divides the potential icing areas into different cluster groups according to the position coordinates. The areas in each group are relatively close and may belong to the same ice ridge area. The centroid coordinates and coverage radius of each cluster group are calculated, the boundary coordinates of each cluster group are extracted as the potential ice ridge boundary position coordinates, the temperature gradient change and impedance spectrum feature change amount at the ice ridge boundary point are extracted in the embodiment, the weighted grey correlation degree between the ice ridge temperature gradient change rate and the impedance spectrum feature (such as impedance modulus) is calculated, the ice ridge feature correlation coefficient is obtained, and when the ice ridge feature correlation coefficient exceeds the preset correlation coefficient threshold, it indicates that there is a strong correlation between the ice ridge temperature gradient change rate and the ice ridge impedance spectrum feature, which can verify each other that the position indeed exists the ice ridge feature. The corresponding potential ice ridge boundary position coordinates are determined as the effective ice ridge feature correlation points. The embodiment avoids the misjudgment caused by a single feature by using the correlation coefficient, ensures that only the areas that meet the temperature gradient and impedance spectrum feature correlation are recognized as ice ridge areas, better adapts to the complex situation of local icing of composite core conductors, and constructs a physical model of the composite core conductor. The physical model can simulate the state change of the conductor under different conditions. The position coordinates of the ice ridge area feature correlation points and the corresponding feature data are input into the physical model. According to the heat conduction equation, electric field distribution equation and other related equations in the physical model, the ice ridge area feature correlation points are kriging spatially interpolated by combining the feature data at the ice ridge area feature correlation points. The temperature gradient and impedance spectrum feature at the known ice ridge area feature correlation points are taken as boundary conditions to solve the temperature and impedance distribution at other positions of the conductor, so as to locate all possible local ice ridge areas on the conductor and realize accurate positioning of the local ice ridge area. The ice melting control can be more accurately operated on the actual existing ice ridge area, the ice melting efficiency is improved, and the energy consumption is reduced.

[0085] In some embodiments, the impedance spectrum feature deviation amount includes a resonance frequency deviation amount and an impedance modulus deviation amount, and the step of obtaining the local ice ridge morphological feature according to the temperature gradient deviation amount and the impedance spectrum feature deviation amount of the local ice ridge area includes:

[0086] The temperature gradient deviation amount, the resonance frequency deviation amount and the impedance modulus deviation amount corresponding to the local ice ridge area are weighted and fused to obtain a multi-source fusion deviation amount of each monitoring area;

[0087] The multi-source fusion deviation amount of each monitoring area is input into a pre-constructed ice ridge thickness prediction model to obtain ice ridge thickness distribution data;

[0088] According to the ice ridge thickness distribution data, the boundary coordinates of a continuous coverage area with an ice ridge thickness greater than a preset thickness threshold are extracted, and a principal component analysis method is used to determine an ice ridge coverage main direction according to the boundary coordinates of the continuous coverage area; the ice ridge coverage main direction is a direction of an eigenvector corresponding to a maximum eigenvalue of a covariance matrix of the boundary coordinates of the continuous coverage area;

[0089] An angle between the ice ridge coverage main direction and an axial direction of the composite core conductor is calculated to obtain an ice ridge coverage angle, and the ice ridge thickness distribution data is projected and integrated along the ice ridge coverage main direction to calculate an ice ridge extension length.

[0090] The ice ridge thickness distribution data, the ice ridge coverage angle, and the ice ridge extension length are integrated to obtain a local ice ridge morphological feature.

[0091] Specifically, the temperature gradient reference value, the resonance frequency reference value, and the impedance modulus reference value of the conductor section corresponding to the local ice ridge area are obtained from the no-ice coverage state reference data. For each local ice ridge area, the temperature gradient deviation, the resonance frequency deviation, and the impedance modulus deviation are first obtained, wherein the temperature gradient deviation can be obtained by the absolute difference between the temperature gradient data of the current local ice ridge area and the temperature gradient reference value under the normal no-ice coverage state; the resonance frequency deviation is the relative deviation between the resonance frequency of the current local ice ridge area and the resonance frequency reference value under the no-ice coverage state; and the impedance modulus deviation is the relative deviation between the impedance modulus of the current local ice ridge area and the impedance modulus reference value under the no-ice coverage state. Then, the feature weight is dynamically calculated by the entropy method, which automatically adjusts the weight according to the data dispersion degree to avoid subjective setting deviation. The temperature gradient deviation, the resonance frequency deviation, and the impedance modulus deviation are weighted and fused according to the set feature weight to obtain a multi-source fusion deviation.

[0092] Based on the pre-obtained training data, the regression model based on the radial basis function neural network is used to construct the ice ridge thickness prediction model. The training data can include the multi-source fusion deviation under different ice ridge thickness conditions and the corresponding ice ridge thickness actual measurement value. The training data is input into the regression model for training, so that the regression model can learn the mapping relationship between the multi-source fusion deviation and the ice ridge thickness. When the ice ridge thickness prediction is needed, the multi-source fusion deviation of each monitoring area is input into the pre-trained ice ridge thickness prediction model, and the corresponding ice ridge thickness value is output after the processing of the ice ridge thickness prediction model. The prediction is sequentially performed on all monitoring areas to obtain the ice ridge thickness distribution data of the entire conductor.

[0093] The embodiment traverses the ice ridge thickness distribution data to find all continuous coverage areas with ice ridge thickness greater than a preset thickness threshold. The continuous coverage areas are composed of multiple adjacent monitoring areas, and the ice ridge thickness of each monitoring area exceeds the preset thickness threshold. The preset thickness threshold can be determined according to actual ice melting control requirements and the degree of influence of the ice ridge on the conductor. For each continuous coverage area, the coordinates of the boundary monitoring area are extracted, and the extracted boundary coordinates of the continuous coverage area are combined to form a data set. The covariance matrix of the data set is calculated. The covariance matrix can reflect the distribution and correlation of the data in different directions. Then, the covariance matrix is subjected to eigenvalue decomposition to find the eigenvector direction corresponding to the largest eigenvalue. This direction is the main direction of ice ridge coverage. The embodiment compares the main direction of ice ridge coverage with the axis direction of the conductor to calculate the included angle therebetween, thereby obtaining the ice ridge coverage angle. Meanwhile, the embodiment establishes a projection line along the main direction of ice ridge coverage, and projects the ice ridge thickness distribution data onto the projection line. For example, for each monitoring area, the projection coordinates of the monitoring area in the main direction of ice ridge coverage are calculated according to the coordinates and the ice ridge thickness value of the monitoring area. The ice ridge thickness values of adjacent monitoring areas are integrated and accumulated. The integral method can be numerical integration, such as the trapezoidal rule. The ice ridge extension length is calculated. Assuming that the distribution of the ice ridge thickness value of each monitoring area on the projection line is known within the projection length range from the starting point to the ending point in the main direction of ice ridge coverage, the ice ridge extension length is 15 m, which is calculated by integration. The ice ridge thickness distribution data, the ice ridge coverage angle, and the ice ridge extension length are integrated to obtain the local ice ridge morphological characteristics.

[0094] S3. The local ice ridge morphological characteristics are integrated along the length direction of the composite core conductor to obtain an ice ridge distribution parameter, and a local conductor thermoelectric characteristic parameter is calculated according to the ice ridge distribution parameter and the conductor sag data.

[0095] In some embodiments, the step of integrating the local ice ridge morphological characteristics along the length direction of the composite core conductor to obtain the ice ridge distribution parameter comprises:

[0096] The composite core conductor is divided into equally spaced conductor sections along the length direction of the composite core conductor, and the ice ridge thickness distribution data in each conductor section is segmented and integrated along the axis direction of the composite core conductor to generate an ice ridge thickness integral value of each conductor section.

[0097] An ice ridge average thickness of each conductor section is obtained according to the ratio of the ice ridge thickness integral value to the length of the conductor section, and an average value of the ice ridge average thickness of all conductor sections is calculated to obtain an ice ridge base thickness index.

[0098] An ice distribution uniformity index is obtained according to the relative deviation between the ice ridge average thickness of each conductor section and the ice ridge base thickness index.

[0099] According to the ratio between the average ice ridge thickness difference of adjacent conductor sections and the conductor section length, the ice ridge along-line density gradient of adjacent conductor sections is obtained;

[0100] According to the ice ridge coverage angle, the ice ridge along-line density gradient is directionally corrected to obtain a corrected ice ridge along-line density gradient, and the ice ridge distribution uniformity index and the corrected ice ridge along-line density gradient are taken as the ice ridge distribution parameters.

[0101] Specifically, according to the ice ridge extension length, the composite core conductor is divided into equally spaced conductor sections along the length direction of the composite core conductor, for example, the total length of the conductor is 100 meters, which is divided into 100 sections, each section is 1 meter long, and the ice ridge thickness distribution data in each conductor section is segmented and integrated along the axial direction of the conductor. In the present embodiment, the conductor section can be divided into smaller micro units (for example, 0.1 meter for each micro unit), the product of the ice ridge thickness and the micro unit length in each micro unit is calculated, and the products of all micro units are accumulated to obtain the ice ridge thickness integral value of the conductor section. Then, the present embodiment calculates the ratio of the ice ridge thickness integral value to the conductor section length to obtain the average ice ridge thickness of the conductor section. The present embodiment calculates the arithmetic mean of the average ice ridge thickness of all conductor sections as the ice ridge basic thickness index. The ice ridge basic thickness index is equal to the sum of the average ice ridge thickness of all sections divided by the number of conductor sections.

[0102] Then, the embodiment calculates the relative deviation of the ice ridge average thickness of each conductor section and the ice ridge base thickness index, the relative deviation is equal to the absolute value of the difference between the ice ridge average thickness of the conductor section and the ice ridge base thickness index divided by the ice ridge base thickness index, and the average of the relative deviations of all conductor sections is obtained to obtain the ice distribution uniformity index, when the average relative deviation tends to 0, the uniformity index tends to 1, indicating that the ice distribution is highly uniform; when the average relative deviation increases, the uniformity index tends to 0, indicating that the ice distribution is less uniform; at the same time, the embodiment calculates the ice ridge average thickness difference of adjacent conductor sections, divides the ice ridge average thickness difference of adjacent conductor sections by the conductor section length to obtain the ice ridge line density gradient of adjacent conductor sections, and directionally corrects the ice ridge line density gradient according to the ice ridge coverage angle, which is the inclination angle of the ice ridge relative to the conductor axial direction. If the ice ridge coverage angle is less than a preset threshold (for example, 90 degrees), it is considered that the ice ridge distribution is locally covered, and the directional correction can be realized by multiplying the ice ridge line density gradient by the cosine value of the ice ridge coverage angle to obtain the directionally corrected density gradient, so as to consider the influence of the ice ridge coverage direction on the density gradient; if the ice ridge coverage angle is greater than or equal to the preset threshold, it is considered that the ice ridge distribution is fully covered, and the ice ridge line density gradient remains unchanged. The ice distribution uniformity index and the corrected ice ridge line density gradient are used as the ice ridge distribution parameters for subsequent ice melting dynamic control, wherein the ice distribution uniformity index is used to quantify the overall ice uniformity of the conductor, and the corrected ice ridge line density gradient is used to describe the spatial variation rate of the ice thickness along the conductor length direction and contains the influence weight of the ice ridge extension direction.

[0103] In some embodiments, the step of calculating the local conductor thermoelectric characteristic parameter according to the ice ridge distribution parameter and the conductor sag data comprises:

[0104] Filtering the conductor sag data to obtain a conductor sag change amount, and dividing the composite core conductor along the axial direction into a plurality of equal-interval micro-segments;

[0105] According to the ice ridge distribution parameter and the material properties of the composite core conductor, an ice mass-sag increment coupling equation of each micro-segment is established;

[0106] Based on the conductor sag change amount, the ice mass-sag increment coupling equation is solved by an iterative method to obtain an equivalent temperature rise amount of each micro-segment;

[0107] According to the equivalent temperature rise amount, a linear interpolation method is used to calculate the reference dynamic resistivity of each micro-segment, and a skin effect correction factor is superimposed on the reference dynamic resistivity of the ice-covered section to obtain a corrected resistivity;

[0108] The equivalent thermal conductivity and the equivalent heat capacity of the composite core conductor are calculated based on the equivalent temperature rise amount and the environmental parameters, and the modified resistivity, the equivalent thermal conductivity and the equivalent heat capacity are standardized to obtain the local conductor thermoelectric characteristic parameters.

[0109] Specifically, the Kalman filtering method is used to filter the conductor sag data in the embodiment, so as to eliminate noise interference and obtain the conductor sag change amount. The composite core conductor is divided into a plurality of equal-interval micro-element sections at an interval of 0.5 meters along the axial direction, the spatial resolution of the micro-element section and the ice ridge distribution parameter is aligned, the coupling relationship between the ice coating mass and the sag increment of the micro-element section is established by using the material mechanics catenary equation according to the ice ridge distribution parameter (the ice coating distribution uniformity index and the modified ice ridge density gradient) of the micro-element section and the material properties (such as the conductor outer diameter, the core wire diameter and the material density) of the composite core conductor. Specifically, the ice coating mass of each micro-element section is calculated according to the sag increment, the elastic modulus of the conductor and the cross-sectional area of the conductor, the ice coating distribution uniformity index is used as a mass distribution coefficient for correction, the modified ice ridge density gradient is used as a mass spatial variation rate, and the coupling equation of the ice coating mass-sag increment of each micro-element section is established by combining the relationship between the ice coating mass and the conductor sag change amount. The coupling equation of the ice coating mass-sag increment is used to describe the interaction relationship between the ice coating mass and the conductor sag increment, and the specific calculation formula of the coupling equation of the ice coating mass-sag increment is as follows:

[0110]

[0111] In the formula, m is the ice coating mass; E is the elastic modulus of the composite core conductor; A is the cross-sectional area of the conductor; Δs is the sag increment; g is the acceleration of gravity; and L is the span of the conductor. is the ice coating distribution uniformity, which is used to correct the unevenness of the mass distribution; is the gradient influence coefficient; is the modified ice ridge density gradient.

[0112] ​The initial equivalent temperature rise of each micro-element section is set to 0 (i.e. it is assumed that the initial temperature of the conductor is the same as the ambient temperature), the conductor sag change is substituted into the ice mass-sag increment coupling equation, the Newton iteration method is used to solve the ice mass-sag increment coupling equation, the residual error between the measured value of the sag change and the calculated value of the ice mass-sag increment coupling equation is minimized as the target, the temperature rise is iteratively updated, the real ice mass value is gradually approached through continuous iterative calculation, until the maximum iteration number is reached, thereby obtaining the equivalent temperature rise of each micro-element section, and the equivalent temperature rise reflects the influence of ice on the temperature of the conductor. In each iteration, the ice mass is calculated according to the current equivalent temperature rise (considering the influence of temperature on the ice ridge density), the sag increment is calculated according to the ice mass, and compared with the actual monitored sag change, the equivalent temperature rise is adjusted, so that the error between the calculated sag increment and the actual sag change is less than the preset sag change threshold. Specifically, the equivalent temperature rise of the kth iteration is:

[0113]

[0114] wherein,

[0115]

[0116]

[0117]

[0118] In the formula, is the equivalent temperature rise of the kth iteration; is the equivalent temperature rise of the (k-1)th iteration; is a residual error function, used to calculate the residual error between the measured value of the sag change and the calculated value of the ice mass-sag increment coupling equation; is a sensitivity function; is the current equivalent temperature rise substituted into the ice mass-sag increment coupling equation to calculate the sag increment; is the measured filtered sag increment; is the thermal expansion coefficient of the conductor.

[0119] The embodiment calculates the benchmark dynamic resistivity of each micro-element section by using linear interpolation method according to the equivalent temperature rise and the temperature coefficient of resistance of the conductor material. For the icing section, the embodiment considers the influence of the skin effect, superimposes a skin effect correction factor on the benchmark dynamic resistivity of the icing section to obtain a corrected resistivity. The corrected resistivity can more accurately reflect the actual resistive characteristics of the conductor under icing conditions. The skin effect correction factor is calculated according to the ratio between the conductor radius and the skin depth. The skin depth is obtained by multiplying the inverse of the square root of the material conductivity, the inverse of the square root of the ice-melting current frequency and a constant proportional factor. The corrected resistivity is equal to the product of the benchmark dynamic resistivity and the skin effect correction factor. The constant proportional factor can be set according to the specific implementation. The embodiment calculates the equivalent thermal conductivity and the equivalent heat capacity of the composite core conductor by using the heat conduction and heat convection models based on the equivalent temperature rise and the environmental parameters (such as the ambient temperature and the wind speed). The equivalent thermal conductivity and the equivalent heat capacity reflect the heat transfer and heat storage characteristics of the conductor under icing conditions. The equivalent thermal conductivity is equal to the weighted average of the thermal conductivity of the conductor material and the thermal conductivity of the icing layer. The equivalent heat capacity is equal to the weighted average of the heat capacity of the conductor material and the heat capacity of the icing layer. The corrected resistivity, the equivalent thermal conductivity and the equivalent heat capacity are standardized, and the standardized corrected resistivity, the equivalent thermal conductivity and the equivalent heat capacity are taken as the local conductor thermoelectric characteristic parameters. The local conductor thermoelectric characteristic parameters can comprehensively reflect the thermoelectric characteristics of the conductor under icing conditions, which are used for subsequent ice-melting dynamic control.

[0120] S4. Real-time measurement of the actual thermoelectric measurement value of the composite core conductor, and construction of a nonlinear local thermoelectric calibration model by using the least square method based on the local conductor thermoelectric characteristic parameters and the actual thermoelectric measurement value.

[0121] In some embodiments, the step of constructing a nonlinear local thermoelectric calibration model by using the least square method based on the local conductor thermoelectric characteristic parameters and the actual thermoelectric measurement value comprises:

[0122] Calculation of the conductor theoretical temperature distribution according to the local conductor thermoelectric characteristic parameters and the actual thermoelectric measurement value;

[0123] Temporal and spatial alignment of the conductor theoretical temperature distribution and the actual temperature data to obtain temperature deviation distribution data;

[0124] Global iterative calibration of the heat exchange coefficient by using the least square method with the optimization target of minimizing the root mean square error of the temperature deviation distribution data to obtain a calibrated heat exchange coefficient; the heat exchange coefficient includes the air convection heat exchange coefficient and the radiation heat exchange coefficient;

[0125] Analysis of the nonlinear temperature effect of the composite core conductor material properties according to the temperature deviation distribution data to obtain conductor material property correction factors in different temperature intervals;

[0126] correct the local conductor thermoelectric characteristic parameter according to the conductor material attribute correction factor, to obtain a calibrated conductor thermoelectric characteristic parameter;

[0127] construct a nonlinear local thermoelectric calibration model according to the calibrated conductor thermoelectric characteristic parameter and the calibrated heat exchange coefficient.

[0128] Specifically, the embodiment measures the actual thermoelectric measurement value of the composite core conductor in real time, which can include conductor surface temperature and ambient temperature data, etc. According to the local conductor thermoelectric characteristic parameter and the actual thermoelectric measurement value (such as conductor current and voltage), the theoretical temperature distribution of the conductor is calculated through the one-dimensional unsteady heat conduction control equation, which is the heat conduction equation. At the same time, the embodiment measures the temperature distribution along the conductor in real time through the distributed optical fiber temperature measurement system or the infrared temperature measuring instrument to obtain the measured temperature data. Through the time stamp and spatial position matching, the theoretical temperature distribution of the conductor is aligned with the measured temperature data in time and space, eliminating the measurement error and delay effect, calculating the difference between the theoretical temperature and the measured temperature, and obtaining the temperature deviation distribution data.

[0129] The embodiment takes minimizing the root mean square error of the temperature deviation distribution data as the optimization goal, wherein the root mean square error is equal to the square root of the average of the square of the temperature deviation at each position. The embodiment uses the least squares method to globally iteratively calibrate the heat exchange coefficient (air convection heat exchange coefficient and radiation heat exchange coefficient). The specific process is to initialize the air convection heat exchange coefficient and the radiation heat exchange coefficient. In each iteration, the theoretical temperature distribution is recalculated according to the current heat exchange coefficient, and the temperature deviation distribution data is updated. The heat exchange coefficient is adjusted so that the root mean square error gradually decreases until the root mean square error is less than the preset convergence error threshold. After multiple iterations, the calibrated heat exchange coefficient is obtained, which includes the calibrated air convection heat exchange coefficient and the radiation heat exchange coefficient. The embodiment analyzes the nonlinear effect of the material properties (such as resistivity, thermal conductivity, and heat capacity) of the composite core conductor with temperature change according to the temperature deviation distribution data, and divides the temperature interval into multiple subintervals according to the temperature deviation distribution data, to obtain the material property correction factor of each subinterval. The material property correction factor is equal to the average change rate of the material property in the current temperature interval. The embodiment corrects the local conductor thermoelectric characteristic parameter (corrected resistivity, equivalent thermal conductivity, and equivalent heat capacity) according to the conductor material attribute correction factor. Specifically, the local conductor thermoelectric characteristic parameter is multiplied by the conductor material attribute correction factor, and the calibrated local conductor thermoelectric characteristic parameter and the calibrated heat exchange coefficient are substituted into the heat conduction equation to construct a nonlinear local thermoelectric calibration model. The model can more accurately reflect the thermoelectric characteristics of the composite core conductor under actual operating conditions. The calculation formula of the nonlinear local thermoelectric calibration model is:

[0130]

[0131] wherein, is the calibrated equivalent heat capacity; is the wire temperature; is the time; is the calibrated equivalent thermal conductivity, representing the ability of the material to conduct heat; x is the spatial position coordinate; is the calibrated air convection heat transfer coefficient, representing the ability of the convection heat transfer between the wire and the air; is the wire perimeter; is the ambient temperature; is the calibrated corrected resistivity; is the ice-melting current; is the temperature spatial second derivative; is the temperature rate of change.

[0132] S5. According to the composite core wire icing state evaluation parameter, the target ice-melting current waveform is obtained by using the nonlinear local thermal-electric calibration model.

[0133] In some embodiments, the step of obtaining the target ice-melting current waveform according to the composite core wire icing state evaluation parameter by using the nonlinear local thermal-electric calibration model comprises:

[0134] The icing state multi-dimensional feature vector is analyzed by using a principal component analysis method to obtain a composite core wire icing state evaluation parameter;

[0135] The nonlinear local thermal-electric calibration model is compensated online according to the composite core wire icing state evaluation parameter to obtain a wire ice-melting thermal-electric control model that meets the current icing condition;

[0136] According to the ice thickness distribution and the wire material characteristics, the initial ice-melting current waveform of the local ice rib area of the composite core wire is calculated by using the wire ice-melting thermal-electric control model;

[0137] The initial ice-melting current waveform is adjusted according to the real-time monitored ambient temperature data and ice layer state data to obtain a target ice-melting current waveform of the local ice rib area of the composite core wire; the target ice-melting current waveform includes a target ice-melting current frequency and a target ice-melting current amplitude.

[0138] Specifically, the embodiment performs principal component analysis on the icing state multi-dimensional feature vector, extracts the main influencing factors as the icing state evaluation parameters, for example, the embodiment finds through principal component analysis that the ice thickness and the conductor sag are the main factors affecting the ice melting effect, so these two parameters are taken as the icing state evaluation parameters, and the non-linear local thermoelectric calibration model is compensated online according to the composite core conductor icing state evaluation parameters (such as ice thickness and conductor sag), specifically, the icing state evaluation parameters are fed back to the non-linear local thermoelectric calibration model in real time, the key parameters (such as corrected resistivity, equivalent thermal conductivity and equivalent heat capacity, etc.) in the non-linear local thermoelectric calibration model are adjusted, and the non-linear calibration model parameters are dynamically updated, for example, when the ice thickness increases, the resistivity and thermal conductivity parameters of the model will be adjusted accordingly to reflect the influence of ice on the thermoelectric properties of the conductor, after parameter compensation, the conductor ice melting thermoelectric control model that meets the current icing conditions is obtained, which can more accurately simulate the thermoelectric behavior of the conductor under the current icing state, then, the embodiment calculates the ice melting current of the local ice ridge region at a specific conductor temperature using the conductor ice melting thermoelectric control model according to the ice thickness distribution and the conductor material characteristics , and the ice melting current is decomposed into fundamental wave and harmonic components by Fourier decomposition to obtain the initial ice melting current waveform, at the same time, the embodiment monitors the environmental temperature data and ice layer state data in real time, and adjusts the initial ice melting current waveform according to the monitored environmental temperature data and ice layer state data, if the environmental temperature is low (for example, -10 degrees Celsius), the current amplitude or pulse width can be increased to improve the ice melting efficiency, if the ice layer surface temperature is close to the melting point (for example, -1 degree Celsius), the current amplitude can be reduced to avoid overheating, by comprehensively considering these real-time data, the target ice melting current waveform that adapts to the current environment and ice layer state is finally obtained.

[0139] S6. According to the target ice melting current waveform, control the mobile high-voltage ice melting device to output real-time ice melting current pulses, and dynamically adjust the pulse output timing and energy distribution ratio of the composite core conductor according to the real-time ice melting current pulses.

[0140] According to the target ice melting current waveform, the embodiment controls the mobile high-voltage ice melting device to output real-time ice melting current pulses, specifically, the parameters of the target ice melting current waveform are converted into specific control signals, which are sent to the mobile high-voltage ice melting device, so that it outputs current pulses according to the set waveform, and dynamically adjusts the pulse output timing and energy distribution ratio of the composite core conductor according to the ice state and temperature feedback of each region along the conductor, for example, more energy is allocated to the region with thicker ice (i.e. the duration or amplitude of the current pulse is increased), while the energy distribution to the region with thinner ice is reduced to avoid unnecessary thermal damage to the conductor, thereby realizing accurate control of the target ice melting current waveform and ensuring the efficiency and safety of the ice melting process.

[0141] The embodiment of the present application provides a line ice melting dynamic control method based on high-voltage large-capacity movement, the method collects conductor surface temperature distribution data, conductor impedance spectrum data and conductor sag data of a composite core conductor in real time to form an icing state multi-dimensional feature vector; determines a local ice rib area on the composite core conductor based on the icing state multi-dimensional feature vector, and obtains local ice rib morphological characteristics according to a temperature gradient deviation amount and an impedance spectrum characteristic deviation amount of the local ice rib area; performs integral calculation on the local ice rib morphological characteristics along a length direction of the composite core conductor to obtain ice rib distribution parameters, and calculates local conductor thermoelectric characteristic parameters according to the ice rib distribution parameters and the conductor sag data; constructs a nonlinear local thermoelectric calibration model by using a least square method based on the local conductor thermoelectric characteristic parameters and actual measured values; obtains a target ice melting current waveform by using the nonlinear local thermoelectric calibration model according to a composite core conductor icing state evaluation parameter; controls a real-time ice melting current pulse output by a mobile high-voltage ice melting device according to the target ice melting current waveform, and dynamically adjusts pulse output timing and energy distribution proportion of the composite core conductor by the real-time ice melting current pulse. Compared with the prior art, the method realizes accurate positioning of a local ice rib area of a high-voltage large-capacity transmission line under complex working conditions and adaptive ice melting current output control by using multi-dimensional data and target ice melting current waveform optimization, realizes rapid and safe ice melting of a local icing of a high-voltage large-capacity composite core conductor, and ensures safe and stable operation of a power grid under extreme freezing weather.

[0142] It should be noted that the size of the serial number of each process does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0143] In one embodiment, as shown in FIG. 1, Figure 2 The embodiment of the present application provides a line ice melting dynamic control system based on high-voltage large-capacity movement, the system comprises:

[0144] The data acquisition module 101 is used for collecting conductor surface temperature distribution data, conductor impedance spectrum data and conductor sag data of a composite core conductor in real time to form an icing state multi-dimensional feature vector;

[0145] The ice rib analysis module 102 is used for determining a local ice rib area on the composite core conductor based on the icing state multi-dimensional feature vector, and obtaining local ice rib morphological characteristics according to a temperature gradient deviation amount and an impedance spectrum characteristic deviation amount of the local ice rib area;

[0146] The thermal analysis module 103 is configured to integrate the local ice ridge shape features along the length direction of the composite core conductor to obtain ice ridge distribution parameters, and to calculate local conductor thermal-electric characteristic parameters according to the ice ridge distribution parameters and the conductor sag data;

[0147] The model construction module 104 is configured to measure real-time thermal-electric actual values of the composite core conductor, and to construct a nonlinear local thermal-electric calibration model based on the local conductor thermal-electric characteristic parameters and the real-time thermal-electric actual values by using a least square method;

[0148] The waveform generation module 105 is configured to obtain a target ice melting current waveform according to the composite core conductor icing state evaluation parameters and by using the nonlinear local thermal-electric calibration model.

[0149] The ice melting control module 106 is configured to control a mobile high-voltage ice melting device to output real-time ice melting current pulses according to the target ice melting current waveform, and to dynamically adjust the pulse output timing and energy distribution ratio of the composite core conductor by using the real-time ice melting current pulses.

[0150] The specific limitations of the line ice melting dynamic control system based on high-voltage and large-capacity movement can refer to the limitations of the line ice melting dynamic control method based on high-voltage and large-capacity movement described above, and will not be repeated here. Those skilled in the art can realize that the various modules and steps described in combination with the embodiments disclosed in the present application can be realized in hardware, software or a combination of both. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0151] The embodiment of the application provides a line ice-melting dynamic control system based on high-voltage and large-capacity movement, the system acquires the conductor surface temperature distribution data, the conductor impedance frequency spectrum data and the conductor sag data of the composite core conductor in real time through a data acquisition module, and forms an icing state multi-dimensional feature vector; an ice ridge analysis module determines the local ice ridge area on the composite core conductor based on the icing state multi-dimensional feature vector, and obtains the local ice ridge morphological feature according to the temperature gradient deviation amount and the impedance frequency spectrum characteristic deviation amount of the local ice ridge area; a thermoelectric analysis module performs integral calculation on the local ice ridge morphological feature along the length direction of the composite core conductor, obtains the ice ridge distribution parameter, and calculates the local conductor thermoelectric characteristic parameter according to the ice ridge distribution parameter and the conductor sag data; a model construction module constructs a nonlinear local thermoelectric calibration model by using the least square method based on the local conductor thermoelectric characteristic parameter and the actual thermoelectric measurement value; a waveform generation module obtains the target ice-melting current waveform by using the nonlinear local thermoelectric calibration model according to the composite core conductor icing state evaluation parameter; and an ice-melting control module controls the mobile high-voltage ice-melting device to output real-time ice-melting current pulses according to the target ice-melting current waveform, and dynamically adjusts the pulse output timing and energy distribution ratio of the composite core conductor through the real-time ice-melting current pulses. Compared with the prior art, the system realizes the accurate positioning of the local ice ridge area of the high-voltage and large-capacity transmission line under complex working conditions and the adaptive ice-melting current output control by using multi-dimensional data and target ice-melting current waveform optimization, realizes the rapid and safe ice-melting of the local icing of the high-voltage and large-capacity composite core conductor, and ensures the safe and stable operation of the power grid under extreme freezing weather.

[0152] The above-described embodiments only express several preferred embodiments of the application, and the description is relatively specific and detailed, but it should not be understood as limiting the scope of the patent. It should be noted that for ordinary skilled in the art, several improvements and replacements can be made without departing from the technical principles of the application, and these improvements and replacements should be regarded as the protection scope of the application. Therefore, the protection scope of the patent of the application should be subject to the protection scope of the claims.

Claims

1. A dynamic control method for de-icing of power lines based on high-voltage, high-capacity mobile transmission, characterized in that, Includes the following steps: Real-time acquisition of conductor surface temperature distribution data, conductor impedance spectrum data, and conductor sag data of composite core conductors to form a multi-dimensional feature vector of icing state; Based on the multidimensional feature vector of the icing state, the local ice ridge region on the composite core conductor is determined, and the local ice ridge morphology characteristics are obtained according to the temperature gradient deviation and impedance spectrum characteristic deviation of the local ice ridge region. The local icicle morphology features are integrated along the length of the composite core conductor to obtain the icicle distribution parameters, and the local conductor thermoelectric characteristic parameters are calculated based on the icicle distribution parameters and the conductor sag data. The actual thermoelectric values ​​of the composite core conductor are measured in real time, and a nonlinear local thermoelectric calibration model is constructed using the least squares method based on the local conductor thermoelectric characteristic parameters and the actual thermoelectric values. Based on the icing state evaluation parameters of the composite core conductor, the target de-icing current waveform is obtained using the nonlinear local thermoelectric calibration model. The mobile high-voltage de-icing device is controlled to output real-time de-icing current pulses based on the target de-icing current waveform, and the pulse output timing and energy distribution ratio of the composite core conductor are dynamically adjusted through the real-time de-icing current pulses.

2. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 1, characterized in that, The step of determining the local ice spike region on the composite core conductor based on the multidimensional feature vector of the icing state includes: The multidimensional feature vector of the icing state is preprocessed to obtain normalized feature data, and the Euclidean distance between the normalized feature data and the preset non-icing state benchmark data is calculated to generate the icing state deviation quantization value of each monitoring area. The quantized value of the icing state deviation is compared with a preset icing deviation threshold, and when the quantized value of the icing state deviation exceeds the preset icing deviation threshold, the corresponding monitoring area is marked as a potential icing area. The potential icing areas are clustered and grouped using the K-means clustering algorithm to determine the coordinates of potential ice ridge boundaries, and the ice ridge temperature gradient change rate and ice ridge impedance spectrum characteristics at the coordinates of the potential ice ridge boundaries are extracted. Calculate the correlation coefficient of the ice ridge feature between the rate of change of the temperature gradient of the ice ridge and the impedance spectrum feature of the ice ridge. If the correlation coefficient of the ice ridge feature exceeds the preset correlation coefficient threshold, the corresponding potential ice ridge boundary position coordinates are determined as the ice ridge region feature association point. By using a pre-constructed physical model of the conductor to spatially interpolate the feature correlation points of the icicle region, the local icicle region can be located.

3. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 2, characterized in that, The step of preprocessing the multidimensional feature vector of the icing state to obtain normalized feature data includes: The time-domain average filtering of the conductor surface temperature distribution data is performed using the sliding window method to obtain smooth temperature distribution data. The smoothed temperature distribution data is subjected to first-order difference calculation along the conductor length direction to obtain temperature gradient data between adjacent monitoring areas; Wavelet transform is used to perform multi-scale decomposition on the conductor impedance spectrum data to extract the impedance spectrum feature vector of each monitoring area; the impedance spectrum feature vector includes the resonant frequency and impedance magnitude corresponding to the local maximum value of the impedance amplitude. The smoothed temperature distribution data, the temperature gradient data, and the impedance spectrum feature vector are normalized to obtain normalized feature data for each monitoring area.

4. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 3, characterized in that, The impedance spectrum characteristic deviation includes the resonant frequency deviation and the impedance magnitude deviation. The step of obtaining the local icicle morphology characteristics based on the temperature gradient deviation and impedance spectrum characteristic deviation of the local icicle region includes: The temperature gradient deviation, resonant frequency deviation, and impedance modulus deviation corresponding to the local icicle region are weighted and fused to obtain the multi-source fused deviation of each monitoring region. The multi-source fusion deviation of each monitoring area is input into the pre-constructed ice ridge thickness prediction model to obtain ice ridge thickness distribution data; Based on the icicle thickness distribution data, the boundary coordinates of continuous coverage areas with icicle thickness greater than a preset thickness threshold are extracted. Based on the boundary coordinates of the continuous coverage areas, principal component analysis is used to determine the main direction of icicle coverage. The main direction of icicle coverage is the eigenvector direction corresponding to the largest eigenvalue of the covariance matrix of the boundary coordinates of the continuous coverage areas. Calculate the angle between the main direction of the ice ridge coverage and the axis of the composite core conductor to obtain the ice ridge coverage angle, and perform projection integration on the ice ridge thickness distribution data along the main direction of the ice ridge coverage to calculate the ice ridge extension length; By integrating the ice ridge thickness distribution data, the ice ridge coverage angle, and the ice ridge extension length, local ice ridge morphological characteristics are obtained.

5. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 4, characterized in that, The step of integrating the local icicle morphology characteristics along the length of the composite core conductor to obtain the icicle distribution parameters includes: The composite core conductor is divided into equally spaced conductor segments along its length, and the ice ridge thickness distribution data in each conductor segment is integrated along the axial direction of the composite core conductor to generate the ice ridge thickness integral value for each conductor segment. The average thickness of the icicles in each conductor segment is obtained by the ratio of the integral value of the icicle thickness to the length of the conductor segment. The average value of the average thickness of the icicles in all conductor segments is then calculated to obtain the basic icicle thickness index. The uniformity index of icing distribution is obtained based on the relative deviation between the average thickness of icicles and the basic thickness of icicles in each conductor section. The density gradient of ice ridges along the line of adjacent conductor segments is obtained by the ratio between the average thickness difference of ice ridges in adjacent conductor segments and the length of the conductor segments. The density gradient along the ice ridge is directionally corrected based on the ice ridge coverage angle to obtain the corrected density gradient along the ice ridge. The ice distribution uniformity index and the corrected density gradient along the ice ridge are used as ice ridge distribution parameters.

6. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 1, characterized in that, The step of calculating the local thermoelectric characteristic parameters of the conductor based on the icicle distribution parameters and the conductor sag data includes: The conductor sag data is filtered to obtain the conductor sag change, and the composite core conductor is divided into several equally spaced micro-segments along the axial direction. Based on the ice ridge distribution parameters and the material properties of the composite core conductor, an ice accretion mass-sag increment coupling equation is established for each micro-element segment. Based on the change in conductor sag, the ice mass-sag increment coupling equation is solved by iterative method to obtain the equivalent temperature rise of each micro-segment. Based on the equivalent temperature rise, the reference dynamic resistivity of each micro-element segment is calculated using linear interpolation, and the skin effect correction factor is superimposed on the reference dynamic resistivity of the icing section to obtain the corrected resistivity. The equivalent thermal conductivity and equivalent heat capacity of the composite core conductor are calculated based on the equivalent temperature rise and environmental parameters. The corrected resistivity, equivalent thermal conductivity and equivalent heat capacity are then standardized to obtain the local thermoelectric characteristic parameters of the conductor.

7. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 1, characterized in that, The step of constructing a nonlinear local thermoelectric calibration model using the least squares method based on the local conductor thermoelectric characteristic parameters and the actual thermoelectric measurement values ​​includes: The theoretical temperature distribution of the conductor is calculated based on the local conductor thermoelectric characteristic parameters and the actual thermoelectric measurement values. The theoretical temperature distribution of the conductor is spatiotemporally aligned with the measured temperature data to obtain temperature deviation distribution data. With minimizing the root mean square error of the temperature deviation distribution data as the optimization objective, the least squares method is used to perform global iterative calibration of the heat exchange coefficient to obtain the calibrated heat exchange coefficient; the heat exchange coefficient includes the air convection heat transfer coefficient and the radiation heat transfer coefficient. Based on the temperature deviation distribution data, the nonlinear temperature effect of the composite core conductor material properties is analyzed to obtain the conductor material property correction factor for different temperature ranges. The thermoelectric characteristic parameters of the local conductor are corrected according to the conductor material property correction factor to obtain the calibrated conductor thermoelectric characteristic parameters; A nonlinear local thermoelectric calibration model is constructed based on the thermoelectric characteristic parameters of the calibration conductor and the calibration heat exchange coefficient.

8. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 1, characterized in that, The step of obtaining the target de-icing current waveform using the nonlinear local thermoelectric calibration model based on the icing state evaluation parameters of the composite core conductor includes: Principal component analysis was used to analyze the multidimensional feature vector of the icing state to obtain the icing state evaluation parameters of the composite core conductor. Based on the icing state evaluation parameters of the composite core conductor, the nonlinear local thermoelectric calibration model is compensated online to obtain a conductor de-icing thermoelectric control model that meets the current icing conditions. Based on the ice thickness distribution and conductor material properties, the initial ice-melting current waveform of the local ice ridge region of the composite core conductor is calculated using the conductor ice-melting thermoelectric control model. The initial de-icing current waveform is adjusted based on real-time monitored ambient temperature data and ice layer status data to obtain the target de-icing current waveform for the local ice ridge area of ​​the composite core conductor.

9. The dynamic control method for line de-icing based on high-voltage, high-capacity mobile transmission as described in claim 8, characterized in that: The target de-icing current waveform includes the target de-icing current frequency and the target de-icing current amplitude.

10. A dynamic control system for de-icing of high-voltage, high-capacity mobile power lines, characterized in that: The system includes: The data acquisition module is used to collect real-time data on the surface temperature distribution, impedance spectrum, and sag of the composite core conductor, forming a multi-dimensional feature vector of the icing state. The icicle analysis module is used to determine the local icicle region on the composite core conductor based on the multidimensional feature vector of the icing state, and to obtain the local icicle morphology characteristics based on the temperature gradient deviation and impedance spectrum characteristic deviation of the local icicle region. The thermoelectric analysis module is used to perform integral calculations on the local icicle morphology characteristics along the length direction of the composite core conductor to obtain icicle distribution parameters, and to calculate the local conductor thermoelectric characteristic parameters based on the icicle distribution parameters and the conductor sag data. The model building module is used to measure the actual thermoelectric values ​​of the composite core conductor in real time, and to build a nonlinear local thermoelectric calibration model based on the local conductor thermoelectric characteristic parameters and the actual thermoelectric values ​​using the least squares method. The waveform generation module is used to obtain the target de-icing current waveform based on the evaluation parameters of the icing state of the composite core conductor and the nonlinear local thermoelectric calibration model. The de-icing control module is used to control the mobile high-voltage de-icing device to output real-time de-icing current pulses according to the target de-icing current waveform, and to dynamically adjust the pulse output timing and energy distribution ratio of the composite core conductor through the real-time de-icing current pulses.

Images