High-strength anti-icing OPGW sag dynamic prediction and optimization system

Abstract

The application discloses a high-strength anti-icing OPGW sag dynamic prediction and optimization system, and particularly relates to the field of power transmission monitoring in high-cold regions, and is used for solving the problem of sudden change of OPGW sag caused by icing superposition and difficult dynamic prediction. The application is a load-deformation joint sequence relying on multi-source load response data set and accurate correlation of segmented icing distribution, identifies tension rapid transfer nodes and fuses temperature drop shrinkage trend, quantifies risks through machine learning, drives local ice-melting power distribution with risk boundary, effectively suppresses peak tension and leaves controllable residual strain, and continuously corrects the prediction model by using time series increment assimilation; the application adopts synchronous iteration of data driving and model correction, eliminates prediction deviation caused by prediction deviation caused by uneven icing space and temperature shrinkage time sequence coupling; real-time ice-melting scheduling output significantly delays sag overrun caused by tension accumulation, guarantees optical fiber communication performance and conductor safe operation, and provides quantitative decision basis for high-strength anti-icing OPGW line operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of power transmission monitoring in high-altitude and cold regions, and more specifically, to a dynamic prediction and optimization system for high-intensity anti-icing OPGW sag. Background Technology

[0002] In high-altitude and cold mountain power transmission corridors, aluminum-clad steel composite OPGW (optical fiber composite overhead ground wire, a special cable combining optical fiber communication and power transmission functions, widely used in high-voltage transmission lines. OPGW serves as an overhead ground wire for lightning protection and grounding, while also containing optical fibers for high-speed data transmission and communication, enabling dual functions of power transmission and information communication in the power system) frequently encounters the triple effects of rapid ice thickening, sudden drops in nighttime temperatures, and gusty winds. The weight of the ice layer adds extra weight to the conductor, the low temperature causes a sudden reduction in conductor length, and the periodic swaying of gusts generates alternating tension at different span positions. The combined effect of these three factors causes the conductor sag to drop by tens of centimeters in a short time, exceeding the lower limit of the safety gap, and the internal optical fibers are subsequently stretched to the upper limit of the design strain, leading to increased signal attenuation and the risk of flashover discharge.

[0003] However, the current tension-sag check (which calculates the sag value of the conductor crossing section using the catenary equation under predetermined tension conditions and compares the sag with the design safety limit to determine whether the conductor sag height meets the safety clearance requirements) uses a static catenary model and a fixed safety margin to estimate conductor deformation. It does not dynamically analyze the temporal coupling effects of icing weight gain, temperature drop contraction, and wind load impact. The elastic modulus of aluminum-clad steel composite materials is limited, and they cannot provide sufficient rebound when the load step arrives. The static margin is difficult to absorb instantaneous deformation, resulting in a lag in sag over-limit prediction. Often, de-icing or adjustment strategies are only initiated after instability occurs, making it difficult to ensure the safe and stable operation of the line and communication.

[0004] To address the aforementioned problems, a technical solution is provided. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of existing technologies, embodiments of the present invention provide a dynamic prediction and optimization system for sag of high-strength anti-icing OPGW lines. This system relies on a multi-source load response dataset and a load-deformation joint sequence precisely correlated with segmented ice distribution. It identifies nodes with rapid tension transfer and integrates temperature-induced contraction trends. Risk is quantified through machine learning, and local de-icing power allocation is driven by risk boundaries, effectively suppressing peak tension and leaving controllable residual strain. The prediction model is then continuously corrected using time-series incremental assimilation. Data-driven and model calibration are iteratively synchronized to eliminate prediction biases caused by spatial unevenness of ice accumulation and the temporal coupling of temperature contraction. Real-time de-icing scheduling output significantly delays sag exceeding limits caused by tension accumulation, ensuring fiber optic communication performance and conductor safety. It also provides quantitative decision-making basis for the operation and maintenance of high-strength anti-icing OPGW lines, thus solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: Data fusion module: continuously aggregates multi-source load response datasets along the line, aligns them by timestamp to construct a load-deformation joint sequence, and marks the ice distribution map of the conductor segment in the sequence; Tension Analysis Module: Calls the load-deformation joint sequence, uses piecewise degradation mapping to identify tension fast transfer nodes, extracts the tension progression chain triggered by non-uniform ice attachment, and writes it into the sag prediction buffer; Risk assessment module: The deformation trend of conductor caused by temperature drop is superimposed on the sag prediction buffer. By analyzing the spatial and temporal variation of tension, a cross-span sag risk envelope is formed, and the correction amount is written back to the tension rapid transfer node. Intervention and optimization module: compare the corrected cross-span sag risk envelope with the design safety gap. If a compression segment occurs, calculate the minimum strain equilibrium path based on the correction amount, allocate distributed ice melting power to weaken the peak tension, and simultaneously transmit the residual strain back. Prediction update module: Receives the combined load-deformation sequence after residual strain reorganization, performs time-series incremental assimilation to refresh the span sag risk envelope, and pushes the optimized span tension-sag guidance table to the scheduling end.

[0007] In a preferred embodiment, the data fusion module processes the following: Meteorological remote sensing data, conductor vibration acoustic emission data, and fiber optic strain records were collected along the route, aligned with a unified timestamp, and cleaned and preprocessed to ensure data accuracy and consistency. The preprocessed meteorological remote sensing data and conductor vibration acoustic emission data were integrated into the load portion of the load-deformation joint sequence, and the fiber optic strain records were integrated into the deformation portion. Conductor segment information was introduced and the icing status was marked. The icing thickness and weight of each conductor segment were estimated using the meteorological remote sensing data and fiber optic strain records. The results were corrected to ensure consistency with the actual state and segment icing distribution maps were generated. The segment icing distribution maps were embedded into the load-deformation joint sequence to form an expanded joint sequence, which was stored in a database and supported for rapid querying in both time and spatial dimensions.

[0008] In a preferred embodiment, the tension analysis module processes the following: Meteorological remote sensing data, conductor vibration acoustic emission data, fiber optic strain records, and ice distribution maps of conductor segments within a specific time window are extracted from the load-deformation joint sequence to construct a time-space two-dimensional matrix. The rate of change of fiber optic strain records for each conductor segment within the selected time window is calculated to obtain the strain change rate. By calculating the fluctuation intensity of the strain change rate and combining it with a threshold judgment, the nodes of rapid tension transfer are identified.

[0009] In a preferred embodiment, the tension analysis module further includes the following processing steps: Based on the tension fast transfer node and the normalized value of icing mass in the conductor segment icing distribution map, the tension transfer intensity between adjacent nodes is calculated, a tension transfer network is constructed, and a path search algorithm is used to extract the tension progression chain; the attributes of the tension fast transfer node and the tension progression chain are written into the sag prediction buffer as key-value pairs.

[0010] In a preferred embodiment, the risk assessment module processes the following: Data from the tension fast transfer node and tension progression chain are extracted from the sag prediction buffer. Combined with temperature data, the thermal shrinkage of the conductor due to temperature drop is calculated and superimposed onto the sag prediction buffer. Based on the tension progression chain, the tension increment and duration between adjacent nodes are calculated to generate a step steepness field. The tension step robustness index is obtained by integrating along the span. The fiber strain record is decomposed into thermal shrinkage and viscoelastic relaxation components. The contributions of the thermal shrinkage and viscoelastic relaxation components to the conductor displacement are calculated to generate the offset rate. The low-temperature viscoelastic offset factor is calculated by combining the weighted average of icing mass.

[0011] In a preferred embodiment, the risk assessment module further includes the following processing: The tension step robustness index and low-temperature viscoelastic offset factor are input into the support vector regression model to train and predict the sag intensification coefficient. The sag value of each span is calculated based on the catenary model to form an initial sag risk envelope. The initial sag risk envelope is adjusted in combination with the sag intensification coefficient. The correction amount of the tension rapid transfer node is calculated based on the adjustment difference and node weight, and the correction amount is written back to the sag prediction buffer.

[0012] In a preferred embodiment, the intervention optimization module processes the following: Extract the corrected span sag risk envelope and design safety clearance from the sag prediction buffer. Identify spans with sag exceeding the limit by calculating the difference between the span sag risk envelope and the design safety clearance, and generate a set of compression segments. For the spans in the set of compression segments, calculate the strain exceedance. Based on the strain exceedance, construct a network and apply the minimum spanning tree algorithm to generate the minimum strain equilibrium path.

[0013] In a preferred embodiment, the intervention optimization module further includes the following processing: Based on the minimum strain equilibrium path and the total available melting power, the distributed melting power is calculated and allocated to the spans in the compression section set to reduce the tension peak. The real-time strain value after melting is recorded by the fiber optic strain monitoring device, the residual strain is calculated and transmitted back to the sag prediction buffer to update the attribute fields of the tension fast transfer node.

[0014] In a preferred embodiment, the prediction update module processes the following: The residual strain data is received and integrated into the load-deformation joint sequence to generate an extended load-deformation joint sequence. Based on the extended load-deformation joint sequence, the residual strain data is used as observations to update the strain estimate of the conductor. The sag value of each span is calculated based on the updated strain estimate of the conductor, and the refreshed cross-span sag risk envelope is generated.

[0015] In a preferred embodiment, the prediction update module further includes the following processing: Based on the refreshed cross-gear sag risk envelope, for each gear, the optimal tension value and the corresponding optimal sag value are found through numerical optimization methods, and an optimized cross-gear tension-sag guide table is generated and pushed to the scheduling terminal.

[0016] The technical effects and advantages of the high-strength anti-icing OPGW sag dynamic prediction and optimization system of this invention are as follows: This invention relies on a load-deformation joint sequence precisely correlated with a multi-source load response dataset and segmented icing distribution. The system identifies nodes with rapid tension transfer and integrates temperature-induced contraction trends. A cross-span sag risk envelope is constructed using machine learning excitation factors. This risk boundary drives local de-icing power allocation, effectively suppressing peak tension and leaving controllable residual strain. The prediction model is then continuously corrected using time-series incremental assimilation. The entire process employs data-driven and model-calibrated iterations simultaneously, eliminating prediction biases caused by spatial unevenness of icing and the temporal coupling of temperature contraction. Real-time de-icing scheduling significantly delays sag exceeding limits caused by tension accumulation, ensuring fiber optic communication performance and conductor safety, and providing quantitative decision-making basis for the operation and maintenance of high-strength anti-icing OPGW lines. Continuous iteration optimizes the prediction scheme, minimizing the intervention response window and improving overall maintenance efficiency and economic benefits. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the structure of the high-strength anti-icing OPGW sag dynamic prediction and optimization system of the present invention; Figure 2 This is a flowchart illustrating the risk assessment module of the high-strength anti-icing OPGW sag dynamic prediction and optimization system of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Figure 1 The present invention provides a dynamic prediction and optimization system for the sag of high-strength anti-icing OPGW, comprising: Data fusion module: continuously aggregates multi-source load response datasets along the line, aligns them by timestamp to construct a load-deformation joint sequence, and marks the ice distribution map of the conductor segment in the sequence.

[0020] Tension Analysis Module: Calls the load-deformation joint sequence, uses piecewise degradation mapping to identify tension fast transfer nodes, extracts the tension progression chain triggered by non-uniform ice attachment, and writes it into the sag prediction buffer.

[0021] Risk assessment module: The deformation trend of the conductor caused by temperature drop is superimposed on the sag prediction buffer. By analyzing the spatial and temporal variation of tension, a cross-span sag risk envelope is formed, and the correction amount is written back to the tension rapid transfer node.

[0022] Intervention and optimization module: Compare the corrected cross-span sag risk envelope with the design safety gap. If a compression segment occurs, calculate the minimum strain equilibrium path based on the correction amount, allocate distributed ice melting power to weaken the peak tension, and simultaneously transmit the residual strain back.

[0023] Prediction update module: Receives the combined load-deformation sequence after residual strain reorganization, performs time-series incremental assimilation to refresh the span sag risk envelope, and pushes the optimized span tension-sag guidance table to the scheduling end.

[0024] In power transmission corridors in high-altitude and cold mountainous areas, aluminum-clad steel composite fiber optic overhead ground wires (OPGW) bear the dual functions of power transmission and communication, and must cope with the challenges posed by extreme environments. In particular, rapid ice thickening increases conductor weight, sudden temperature drops at night cause conductor material shrinkage, and gusts of wind induce periodic tension fluctuations. The combined effect of these three factors significantly alters conductor sag (the degree of conductor droop). Excessive sag can compress safety gaps, leading to excessive fiber strain, accelerated signal attenuation, and even the risk of flashover, threatening line safety and communication stability. Traditional static tension-sag verification methods rely on fixed models and cannot reflect the dynamic coupling effects of icing, temperature drop, and wind load in real time, resulting in prediction lag and difficulty in taking early intervention measures. Therefore, a high-strength anti-icing OPGW sag dynamic prediction and optimization system has emerged. Through multi-source data fusion and real-time analysis, it accurately predicts sag change trends and optimizes response strategies to ensure the safe and stable operation of conductors under extreme conditions.

[0025] This data fusion module serves as the starting point for system operation and aims to provide an accurate data foundation for subsequent dynamic predictions.

[0026] The goal of the data fusion module is to continuously aggregate multi-source load response datasets along the route, align them by timestamp to construct a joint load-deformation sequence, and annotate the ice distribution map of the traverse segments within the sequence. The detailed technical logic is as follows: Data Acquisition and Preprocessing: Before analyzing the relationship between conductor load and deformation, comprehensive and reliable raw information is required. Therefore, a multi-source load response dataset was first collected along the conductor line, including meteorological remote sensing data, conductor vibration acoustic emission data, and fiber optic strain records. Meteorological remote sensing data characterizes environmental conditions, such as the effects of temperature and humidity on the conductor; conductor vibration acoustic emission data reflects the characteristics of dynamic loads, such as wind-induced vibrations; and fiber optic strain records directly provide the real-time state of conductor stress and deformation. These data collectively constitute the fundamental information source for the analysis.

[0027] Next, to ensure the synchronization of data from different sources, all data were aligned according to a unified time benchmark. Using the time points of the meteorological remote sensing data as a reference standard, for each time point of other data, the closest meteorological remote sensing data time point was found, and the data value at that time point was used as the matching result. If any time points were missing, they were filled by calculating the average of the data from two adjacent time points. This alignment method ensured the consistency of the time series and facilitated subsequent joint analysis of multi-source data.

[0028] After data alignment, cleaning and preprocessing are necessary to improve data accuracy and reliability. First, thresholds are set based on physical ranges to remove outliers exceeding reasonable limits, such as strain values ​​exceeding the material's tolerance. Then, smoothing is applied to the time series data. Specifically, the average of each data point and several preceding and following data points is calculated and used to replace the original value, thereby eliminating high-frequency noise while preserving the main trend of the data. The cleaned and preprocessed data provides a clean and consistent foundation for constructing the joint sequence.

[0029] Constructing a load-deformation combined sequence: To directly correlate external loads with the conductor's response, the preprocessed data needs to be integrated into a load-deformation joint sequence. Specifically, meteorological remote sensing data and conductor vibration acoustic emission data are combined to form the load component, representing the combined impact of the external environment and dynamic forces on the conductor; fiber optic strain records are used as the deformation component, representing the actual stress and deformation state of the conductor. By mapping the load and deformation components one-to-one in chronological order, a joint sequence containing the time dimension is formed. This integration method clearly reflects the relationship between load changes and deformation response, facilitating the analysis of the influence of external factors on the conductor's condition.

[0030] Building upon this foundation, to enhance the spatial resolution of the data, segmentation information of the conductor is further introduced. The entire conductor is divided into multiple continuous segments, the length of which is determined according to actual needs. For each segment, its icing state is marked based on its corresponding fiber strain record, such as the presence or initial degree of icing. This marking process integrates the spatial dimension into the joint sequence, enabling the data to not only reflect temporal changes but also the state differences at different locations on the conductor, providing support for subsequent detailed analysis of icing distribution.

[0031] Generation of segmented ice distribution maps: To quantify the icing situation in each segment of the conductor, it is necessary to estimate the icing thickness and weight based on existing data. First, the icing thickness is calculated using temperature and humidity information from meteorological remote sensing data. This estimation considers the physical process of water condensation when the temperature is below freezing. Specifically, the amount of condensable water in the air is determined based on humidity levels, and then the thickness formed by condensation on the conductor surface is estimated by combining this with the duration of the temperature remaining below freezing. Next, the icing weight is calculated by multiplying the estimated icing thickness by the surface area of ​​each conductor segment to obtain the icing volume, and then multiplying this by the density of ice to obtain the additional mass of each segment. This estimation method provides a specific contribution of icing to the conductor load.

[0032] To ensure consistency between the calculated results and the actual conditions, correction is necessary using fiber optic strain records. The specific method involves comparing the calculated impact of ice weight on conductor strain with the actual measured strain value. If a discrepancy exists, the initial estimate of the ice thickness is adjusted to match the calculated strain with the measured strain. This adjustment process is iterative until the deviation converges to an acceptable range. This correction method improves the accuracy of ice thickness and weight, reflecting the true stress state of the conductor.

[0033] After calculation and correction, the ice thickness and weight of each conductor segment are recorded to create a segmented ice distribution map. Specifically, each segment is labeled with its corresponding ice thickness and weight values, and arranged according to the conductor's location to form a clear distribution description. This distribution map clearly shows the details of icing at different locations on the conductor, providing crucial information for subsequent comprehensive analysis of loads and deformations.

[0034] Data integration and storage: To create a data structure that incorporates both temporal and spatial dimensions, the segmented icing distribution map needs to be embedded into the load-deformation joint sequence. Specifically, at each time point in the joint sequence, the icing thickness and weight information for each segment at that time are added, forming an expanded joint sequence. This expanded joint sequence not only preserves the temporal variations of load and deformation but also includes the icing status of each conductor segment, thus providing a more comprehensive data perspective and facilitating the analysis of the impact of icing on the overall performance of the conductor.

[0035] Finally, to ensure efficient data access and management, the expanded joint sequence is stored in the database. A combined temporal and spatial index structure is used for storage, specifically with time points as the primary index and spatial segments as sub-indexes. For each time point, the corresponding load, deformation, and icing information for each segment are recorded; for each segment, its state at different time points can be quickly retrieved through the spatial index. This storage method supports fast queries by time or spatial dimensions, providing convenient data extraction capabilities for subsequent analysis and prediction.

[0036] Through the processing logic of the aforementioned data fusion module, a load-deformation joint sequence containing temporal and spatial information was successfully constructed, and a segmented ice distribution map of the conductor was generated. The acquisition and preprocessing of multi-source data ensured the comprehensiveness and accuracy of the information; the construction of the joint sequence linked the relationship between load and deformation; the generation of the segmented ice distribution map quantified the spatial distribution of ice accumulation; and the integration and storage of the data provided a complete data structure for efficient access.

[0037] The data fusion module has constructed a load-deformation joint sequence through multi-source data fusion and labeled the icing distribution map of the conductor segments, laying the foundation for dynamic sag prediction. However, the load-deformation joint sequence alone cannot reveal the transmission and evolution of tension within the conductor, especially under non-uniform icing conditions, where tension may accumulate or transfer rapidly at certain nodes, causing local stress concentration and affecting the accuracy of sag prediction. Therefore, the tension analysis module focuses on identifying nodes with rapid tension transfer and extracting the tension progression chain, providing crucial information for subsequent sag risk assessment.

[0038] Call the load-deformation combined sequence: Before analyzing the dynamic changes in conductor tension, key information needs to be extracted from existing data. Therefore, the load-deformation joint sequence is first extracted from the data storage generated by the data fusion module. This sequence includes meteorological remote sensing data, conductor vibration acoustic emission data, fiber optic strain records, and icing distribution maps of conductor segments arranged in chronological order. To facilitate analysis of tension trends over a short period, a specific time window is selected, such as data from the past hour. This data is organized into a two-dimensional matrix, where each row corresponds to a time point, and each column corresponds to different variables or icing information for conductor segments. Specifically, each element in the matrix is ​​the value of a variable or segment at a given time point, such as the value of the fiber optic strain record or the icing thickness. In this way, data changes in both the temporal and spatial dimensions are visually presented, facilitating subsequent analysis of the dynamic evolution of tension over time and location.

[0039] Identifying tension-fast transfer nodes using piecewise degradation mapping: To accurately pinpoint the critical locations of rapid tension changes in a conductor, a detailed analysis of strain changes in each segment is necessary. First, for each segment, the rate of change of strain value within a selected time window is calculated. Specifically, the strain values ​​at two adjacent time points are taken, the difference between them is calculated, and then the difference is divided by the time interval between the two time points to obtain the strain rate of change. The strain rate of change reflects the rate of tension accumulation or release in that segment and is a core indicator for measuring dynamic tension changes. Next, rapid tension transfer nodes are further identified, i.e., segments where the strain rate of change changes drastically within a short period. The identification process involves first calculating the difference between the maximum and minimum strain rate of change for each segment within the time window to obtain the fluctuation intensity. The fluctuation intensity characterizes the severity of tension fluctuations. Then, a threshold is set based on the characteristics of the conductor material, for example, 10% of the design strain upper limit. If the fluctuation intensity of a segment exceeds this threshold, and the sign of the strain rate of change switches between positive and negative at adjacent time points, then that segment is marked as a rapid tension transfer node. By quantifying the amplitude and direction of strain changes, the key points of rapid tension transfer can be accurately located, providing a reliable basis for analyzing the tension transmission path.

[0040] Extracting the tension progression chain triggered by non-uniform ice adhesion: To reveal the transmission pattern of tension in conductors, it is necessary to extract the tension progression chain based on identified tension fast transfer nodes. The tension progression chain is a path from a certain node where tension is gradually transmitted and accumulated segment by segment along the conductor, driven by non-uniform icing distribution. The extraction process first calculates the tension transmission intensity between adjacent nodes. Specifically, it calculates the difference in strain change rate between two adjacent segments, and then uses the normalized value of the icing mass of that segment as a weight to perform a weighted average of the difference. The normalized value of the icing mass is the ratio of the icing weight of that segment to the maximum icing weight within the time window, used to quantify the impact of non-uniform icing on tension transmission. A positive tension transmission intensity indicates increased tension, while a negative value indicates decreased tension. Next, a tension transmission network is constructed based on the tension transmission intensity, where tension fast transfer nodes are the vertices of the network, and the tension transmission intensity is the edge weight connecting the vertices. In this network, a path search algorithm is used, such as successively selecting paths with positive and increasing edge weights from the starting point, to find the tension progression chain. The tension transmission intensity along this path is consistently positive and increases progressively, reflecting the gradual increase in tension from one node to another. This extraction method, by combining icing distribution and strain changes, can clearly reveal the progressive law of tension under non-uniform icing conditions, providing important information for sag risk assessment.

[0041] Write to the sag prediction buffer: To facilitate subsequent use of the analysis results, the identified tension rapid transfer nodes and extracted tension progression chains need to be stored. Specifically, the location, time point, and strain change rate of the tension rapid transfer nodes, along with the node sequence and corresponding transmission intensity of the tension progression chain, are organized into key-value pair structures. The key in each key-value pair is a timestamp, and the value is a set of attributes containing node location, time point, strain change rate, progression chain node sequence, and transmission intensity. During storage, the data is arranged in timestamp order to ensure data orderliness. The writing process involves recording these key-value pairs into a sag prediction buffer, which serves as temporary storage for the analysis results.

[0042] Through the processing logic of the aforementioned tension analysis module, the rapid tension transfer nodes in the conductor are identified, and the tension progression chain triggered by non-uniform icing is extracted. The entire process relies on the fiber optic strain record and icing distribution map in the load-deformation joint sequence, and by calculating the strain change rate and tension transmission intensity, clearly presents the dynamic transmission law of tension in the conductor.

[0043] The tension analysis module identified nodes of rapid tension transfer and extracted the tension progression chain, providing fundamental data for sag risk assessment. However, relying solely on tension transmission patterns is insufficient for a comprehensive assessment of sag risk; it is necessary to further superimpose the conductor deformation trend caused by temperature drop to capture the dynamic changes under load-temperature coupling. Therefore, as... Figure 2 The risk assessment module shown focuses on incorporating the temperature drop and contraction effect into the sag prediction buffer. By analyzing the evolution of tension in the spatiotemporal dimension, it forms a cross-sag risk envelope and writes back the correction amount to the tension rapid transfer node to improve prediction accuracy and optimize the ice melting strategy.

[0044] The temperature drop contraction trend is superimposed on the sag prediction buffer: In analyzing the impact of temperature drop on conductor sag, the primary task is to quantify the change in conductor length caused by temperature variation. To this end, data from the tension rapid transfer node and tension progression chain needs to be extracted from the sag prediction buffer generated by the tension analysis module, combined with temperature data from the load-deformation joint sequence of the data fusion module. Based on this data, the length shortening of the conductor due to thermal expansion and contraction is calculated. The specific calculation process involves first obtaining the conductor's coefficient of linear expansion, initial length, reference temperature, and real-time temperature. The coefficient of linear expansion is an inherent property of the conductor material, reflecting its sensitivity to temperature changes; the initial length is the conductor's length at installation; the reference temperature is the ambient temperature at installation; and the real-time temperature is obtained from meteorological remote sensing data. Then, the difference between the reference temperature and the real-time temperature is multiplied by the coefficient of linear expansion, and then multiplied by the initial length to obtain the conductor's shrinkage at the current temperature. This shrinkage represents the length shortened by the conductor due to the temperature drop. Next, the calculated shrinkage is used as a temperature-dependent shrinkage trend and superimposed on the sag prediction buffer. Specifically, a new field is added to the data record at each time point in the sag prediction buffer to store the conductor shrinkage at that time point. In this way, the sag prediction buffer not only contains information on the tension rapid transfer node and the tension progression chain, but also integrates the influence of temperature changes on conductor length, thus providing more comprehensive data support for subsequent sag analysis.

[0045] Calculate the tension step robustness index: To assess the stability of tension transmission in a conductor, an index reflecting tension change characteristics, namely the tension step robustness index, needs to be calculated. First, data from adjacent nodes are extracted from the tension progression chain, and the tension increment between adjacent nodes is calculated. Specifically, the tension values ​​of two adjacent nodes at the same time point are obtained, and the tension value of the latter node is subtracted from the tension value of the former node to obtain the tension increment. The tension value is calculated by multiplying the conductor's elastic modulus by the fiber strain record, where the elastic modulus is an inherent property of the conductor material, and the strain record is obtained from the data fusion module. Next, the duration for which the tension increment remains positive is recorded, i.e., the duration of sustained tension increase. Then, the tension increment is divided by this duration to obtain the rate of tension change, called the step steepness field, which reflects the rate of tension change per unit time. Further, the step steepness field is integrated along the conductor span to obtain the tension step robustness index. The specific operation involves multiplying the step steepness field value between each pair of adjacent nodes by the span between the two nodes to obtain the product of that segment. Then, the products of all pairs of adjacent nodes are summed to obtain a scalar value, namely the tension step robustness index. This scalar value integrates the amplitude, duration, and spatial distribution of tension changes. The larger the value, the more stable the tension transmission and the less prone it is to sudden changes, thus providing a reliable quantitative basis for subsequent risk assessment.

[0046] Calculate the low-temperature viscoelastic shift factor: In low-temperature environments, the viscoelastic changes of conductor materials significantly affect sag, necessitating the calculation of a low-temperature viscoelastic migration factor to quantify this effect. First, the fiber strain record is decomposed into thermal contraction and viscoelastic relaxation components. The thermal contraction component is calculated by multiplying the conductor's linear expansion coefficient by the difference between the reference and real-time temperatures, yielding the strain directly caused by temperature changes. The viscoelastic relaxation component is separated from the total strain by analyzing the strain's time-varying trend; for example, an exponential decay model can be used to fit the strain's time-varying changes, extracting the time-dependent relaxation portion. Next, the contributions of these two components to conductor displacement are calculated separately. Specifically, the thermal contraction and viscoelastic relaxation components are multiplied by the length of each conductor segment, and the results for each segment are summed to obtain the total displacement caused by thermal contraction and viscoelastic relaxation. Then, the displacement caused by viscoelastic relaxation is divided by the displacement caused by thermal contraction to obtain the migration rate, which reflects the relative influence of the viscoelastic effect on the thermal contraction effect. Finally, based on the icing mass of each conductor segment, a weighted average of the migration rates is calculated to obtain the low-temperature viscoelastic migration factor. The weighting process involves multiplying the offset rate of each segment by the icing mass of the corresponding segment, summing the products of all segments, and finally dividing by the total icing mass to obtain the final result. Segments with larger icing masses contribute more to the offset factor, reflecting the amplifying effect of icing on the viscoelastic effect.

[0047] Obtain the sag excitation coefficient using a machine learning model: To correlate the tension step robustness index and the cryogenic viscoelastic shift factor with sag risk, a machine learning model is needed to generate a sag intensification coefficient. Specifically, a support vector regression model is employed, which can handle nonlinear relationships and is suitable for capturing the complex correlation between tension, temperature, and sag changes. First, historical data is prepared, including past tension step robustness indices, cryogenic viscoelastic shift factors, and corresponding actual sag changes. These sag changes can be obtained through field measurements or simulations. Then, the input features are normalized to eliminate the influence of different dimensions. Specifically, for each feature value, the minimum value from the historical data is subtracted, and then divided by the difference between the maximum and minimum values, mapping it to a range of 0 to 1. Next, the support vector regression model is trained using a radial basis function as the kernel function. During training, the penalty parameter and tolerance are adjusted to minimize the model's prediction error on historical data while ensuring its predictive ability on new data. After training, the current tension step robustness index and cryogenic viscoelastic shift factor (normalized) are input into the model to obtain the output sag intensification coefficient. This coefficient is a dimensionless scalar; a value greater than 1 indicates an increased risk of sag, while a value less than 1 indicates a decreased risk. Transforming a complex physical process into a quantifiable risk indicator makes the dynamic assessment of sag risk more scientific and efficient.

[0048] Form the risk envelope of the sag across the span and write back the correction amount: To comprehensively assess and adjust the sag risk of conductors across different spans, a cross-span sag risk envelope needs to be formed and correction amounts written back to the tension rapid transfer nodes. First, based on the catenary model, and combining strain and icing mass in the load-deformation joint sequence, the sag value of each span at the current time point is calculated. The specific process of the catenary model calculating sag based on tension and load is as follows: using the conductor tension value and unit length load, the suspension shape of each conductor segment is determined, and then the sag value is obtained. The sag values ​​of all spans are collected to form an initial sag risk envelope. Next, based on the sag intensification coefficient obtained in the previous steps, the initial envelope is adjusted. Specifically, each sag value in the initial envelope is multiplied by 1 and the sum of the sag intensification coefficients is added to obtain the adjusted sag risk envelope, reflecting the combined effects of temperature drop and tension progression. Then, the difference between the adjusted envelope and the initial envelope is calculated, and correction amounts are allocated according to the number and weight of the tension rapid transfer nodes. The specific operation involves dividing the envelope difference by the total number of tension fast-transfer nodes to obtain the average correction amount. Then, the average correction amount is multiplied by the weight of each node in the tension progression chain (e.g., the proportion of node tension to total tension) to obtain the correction amount for each node. Finally, the calculated correction amount is recorded in the attribute field of the corresponding node in the sag prediction buffer. This processing method, by dynamically adjusting the sag risk envelope and feeding back the correction amount, ensures the real-time nature and accuracy of risk assessment, providing precise intervention support for subsequent ice-melting strategies.

[0049] Through the processing logic of the aforementioned risk assessment module, starting with the superposition of temperature drop and contraction trends, the tension step robustness index and low-temperature viscoelastic offset factor are calculated stepwise. A machine learning model is then used to generate the sag exacerbation coefficient, ultimately forming a cross-span sag risk envelope and writing back the correction amount. This entire process, targeting the coupling scenario of icing and temperature drop in fiber-optic composite overhead ground wires in high-altitude and cold mountainous areas, achieves spatiotemporal risk quantification, providing a scientific basis for optimizing de-icing strategies.

[0050] The risk assessment module overlays the temperature drop contraction trend into the sag prediction buffer, generating a corrected cross-span sag risk envelope. However, relying solely on the risk envelope to assess sag over-limit risk is insufficient for real-time intervention. Further comparison between the risk envelope and the design safety clearance is necessary to identify compression sections and implement targeted measures to reduce tension peaks and ensure conductor safety. The intervention optimization module focuses on this objective, achieving dynamic control of sag risk through precise calculations and power allocation.

[0051] Compare the corrected span sag risk envelope with the design safety clearance: In transmission line operation, excessive conductor sag can lead to insufficient distance from the ground or ground features, affecting safety. Therefore, numerical comparison is needed to assess whether the sag meets design requirements. The specific process involves first obtaining the corrected span-to-span sag risk envelope generated by the risk assessment module. This envelope is a predicted sag value for the conductor at each span, considering the effects of icing, temperature drop, and tension progression, reflecting the actual degree of conductor sag under current environmental conditions. Simultaneously, the design safety clearance is obtained. This clearance is the lower limit of permissible sag set for each span according to the line design specifications, ensuring that the distance between the conductor and the ground or ground features remains within a safe range. During comparison, for each span, the value of the corrected span-to-span sag risk envelope is subtracted from the value of the design safety clearance, resulting in a difference. This difference represents the degree to which the sag exceeds the design safety clearance, expressed in meters. If the difference is positive, it indicates that the corrected span-to-span sag risk envelope exceeds the design safety clearance, the conductor sag is excessive, and there is a compressed section, meaning the safety clearance is compressed, indicating a risk at that span. If the difference is negative or zero, it indicates that the corrected span-to-span sag risk envelope is within the design safety clearance, and the conductor sag is within the safe range. This direct numerical comparison allows for the rapid identification of spans with excessive sag, providing a clear basis for subsequent risk management, while ensuring a simple and efficient assessment process that is easy to apply in real time.

[0052] Identify the compression segment and calculate the minimum strain equilibrium path: Since the presence of compressed sections may cause abnormal conductor tension or safety hazards, it is necessary to further quantify the strain reduction required to restore sag to the safe clearance in order to optimize the implementation of intervention measures. The specific process is as follows: First, for each compressed section, its strain excess is calculated. The strain excess is calculated based on the sag excess value, span length, conductor elastic modulus, and cross-sectional area. During the calculation, the sag excess value (i.e., the aforementioned difference) is divided by the span length to obtain the displacement change per unit length; then, this displacement change is multiplied by the ratio of the conductor's elastic modulus to its cross-sectional area to obtain the strain excess. Here, the elastic modulus is an inherent property of the conductor material, reflecting its ability to resist deformation; the cross-sectional area is a geometric characteristic of the conductor, affecting its load-bearing capacity. The strain excess is a dimensionless scalar, representing the strain reduction per unit length of conductor required to restore sag to the design safe clearance. Next, based on the strain excess of all compressed sections, a network is constructed, where the compressed sections are nodes in the network, and the strain excess is the weight of each node. Then, a minimum spanning tree algorithm is applied, for example, by sequentially comparing and selecting the connection path with the smallest weight, to generate a minimum strain equilibrium path connecting all compressed segments. This path ensures that the total strain exceedance reaches a minimum when connecting all compressed segments, thus providing a reasonable order and priority for ice melting intervention. This method, by optimizing the intervention path, can effectively coordinate the processing needs of multiple compressed segments, improving the targeting and overall efficiency of the intervention.

[0053] Adjusting distributed ice melting power to reduce peak tension: The occurrence of conductor tension peaks under icing conditions can lead to mechanical damage. Therefore, it is necessary to reduce tension and restore sag by rationally allocating de-icing power. The specific method involves distributing de-icing power according to the minimum strain equilibrium path, ensuring that the power allocation is proportional to the strain exceedance of each compression segment to match the intervention intensity and risk level. The calculation process is as follows: first, calculate the sum of strain exceedances for all compression segments along the minimum strain equilibrium path; then, for each compression segment, divide its strain exceedance by the sum to obtain a proportionality coefficient; finally, multiply the total available de-icing power by this proportionality coefficient to obtain the de-icing power allocated to that compression segment. The total available de-icing power is determined by the capacity of the de-icing equipment and is a fixed value. The allocated de-icing power is used to heat the conductor, reducing the icing mass, thereby reducing the conductor's weight and tension, and weakening the tension peak. This proportional allocation method prioritizes compression segments with larger strain exceedances, ensuring that resources are concentrated in high-risk areas while avoiding power waste, improving the accuracy and economy of de-icing intervention, and ensuring the safe operation of the conductor.

[0054] Synchronous transmission of residual strain: Following ice-melting intervention, changes in conductor strain need to be monitored and recorded in real time to assess the intervention's effectiveness and update system data, providing a basis for subsequent sag prediction. The specific process involves recording the real-time strain values ​​of each conductor segment after ice-melting using fiber optic strain monitoring equipment, comparing them with reference strain values ​​before ice-melting, and calculating the residual strain. The residual strain is calculated by subtracting the reference strain value from the real-time strain value. This difference represents the change in conductor strain after ice-melting intervention. A negative residual strain indicates strain reduction and effective intervention; a positive residual strain indicates insufficient strain reduction, potentially requiring further intervention. The calculated residual strain is then fed back to the sag prediction buffer to update the attribute fields of the tension rapid transfer node. This feedback mechanism promptly reflects the intervention's effectiveness, ensuring data real-time performance and accuracy, while providing updated strain information for subsequent sag risk assessment. This closed-loop control allows for continuous optimization of the intervention strategy, enhancing the safety and stability of conductor operation.

[0055] Through the processing logic of the aforementioned intervention and optimization module, starting from comparing the corrected cross-span sag risk envelope with the design safety gap, the module progressively identifies the compression section, calculates the minimum strain equilibrium path, allocates distributed de-icing power, and transmits residual strain back. This achieves dynamic sag control of the high-strength anti-icing OPGW in the context of icing and temperature drop coupling. The entire process specifically addresses the sag exceeding limit problem. Through precise intervention and real-time feedback, it ensures the safe operation of the conductor in complex environments, providing reliable guarantees for stable power supply and communication in high-altitude and cold mountain transmission corridors.

[0056] The aforementioned data fusion module and intervention optimization module, through multi-source data fusion, dynamic risk assessment, and real-time ice melting intervention, have initially controlled sag risk. However, icing and temperature changes are continuously evolving processes. Residual strain and environmental changes after ice melting intervention still need to be updated in real time and fed back to the prediction model to continuously optimize the sag management strategy. The prediction update module focuses on this objective. By receiving residual strain and performing time-series incremental assimilation, it refreshes the cross-span sag risk envelope and pushes the optimized cross-span tension-sag guidance table to the scheduling end, ensuring the safe and stable operation of the conductor in a dynamic environment.

[0057] Receive residual strain and reorganize the load-deformation sequence: Following icing intervention, the strain state of the conductor changes due to ice melting. To enable the prediction model to capture this change in real time and reflect the current state of the conductor, the residual strain data after intervention needs to be integrated into the existing load-deformation joint sequence. Specifically, the residual strain data is first obtained from the intervention optimization module. This data represents the strain change of each segment of the conductor after the ice melting intervention, reflecting the impact of the intervention on the conductor strain. Next, based on the load-deformation joint sequence generated in the data fusion module, a new field is added for each conductor segment and each time point to record the corresponding residual strain value. This reorganization operation fuses the original load data (including meteorological remote sensing data and conductor vibration acoustic emission data), deformation data (including fiber optic strain recording data), and residual strain data together to generate an expanded load-deformation joint sequence. This expanded sequence fully preserves environmental load information, conductor deformation information, and strain feedback information after intervention, thus providing comprehensive and up-to-date data support for subsequent prediction model updates, enabling the model to accurately reflect the actual operating state of the conductor.

[0058] Execution time-incremental assimilation refreshes the cross-gap sag risk envelope: The dynamic changes in conductor condition require timely adaptation of sag risk assessment. Therefore, time-series incremental assimilation techniques are used to update the sag risk envelope across spans in real time. Specifically, the process involves first using Kalman filtering to optimize and adjust the conductor strain estimate using residual strain data as observed values. During calculation, the conductor strain value predicted based on historical data is taken as the initial estimate, and then combined with residual strain data to calculate the corrected strain estimate using a weighted average. The weights of the weighted average are determined by both model prediction error and observation error; the portion with smaller error has a higher weight in the calculation, resulting in a strain estimate closer to reality. Next, based on the corrected strain estimate, the sag value for each span is calculated. The sag calculation process involves first determining the parameters of the catenary model based on the conductor's unit length load (determined by adding the icing mass to the conductor's self-weight) and tension (calculated by multiplying the strain value by the conductor's elastic modulus), and then solving for the sag value using the catenary's characteristics. Specifically, the sag value is directly proportional to the square of the load per unit length and inversely proportional to the tension, while also considering the minor adjustments to sag caused by strain changes. Finally, the sag values ​​for all spans are integrated to form a refreshed cross-span sag risk envelope. This envelope reflects the latest sag state of the conductor after intervention, providing real-time risk assessment information to the dispatching authority and ensuring the accuracy and timeliness of decision-making.

[0059] Push the optimized span tension-sag guidance table to the scheduling terminal: Following intervention, to enable the dispatching end to continue optimizing conductor tension and sag management, an optimized span-span tension-sag guidance table needs to be generated and pushed out. The specific process is as follows: First, based on the updated span-span sag risk envelope, an optimal tension value is found for each span to ensure the conductor sag value is as close as possible to the designed safety clearance, while avoiding excessive tension that could damage the conductor. During optimization, the tension value is gradually adjusted through numerical iteration. After each adjustment, the corresponding sag value is calculated, and the difference between the sag value and the designed safety clearance, as well as the magnitude of the tension value, are comprehensively evaluated. During the evaluation process, a penalty coefficient is applied to the magnitude of the tension value for weighting, balancing sag safety and tension rationality. The tension value with the best overall evaluation result is selected as the optimal tension value. Subsequently, the corresponding optimal sag value is calculated based on this optimal tension value. Finally, the optimal tension value and optimal sag value for each span are combined into a guidance table and pushed to the dispatching end. This guidance table provides the dispatching end with quantitative operational suggestions, ensuring the conductor maintains a safe and stable operating state after intervention, while reserving management space to cope with future environmental changes.

[0060] Through the processing logic of the aforementioned prediction update module, starting from receiving residual strain and reconstructing the load-deformation joint sequence, the module progressively executes time-series incremental assimilation to refresh the span sag risk envelope, and generates an optimized span tension-sag guidance table which is then pushed to the scheduling end. This achieves continuous monitoring and optimized management of the sag risk of high-strength anti-icing fiber-optic composite overhead ground wires. The entire process ensures that the prediction model can reflect changes in conductor condition in real time, providing precise support for the safe and stable operation of transmission lines in high-altitude and cold mountainous areas.

[0061] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0062] It should be noted that the system of the present invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting a variety of hardware environments and usage requirements.

[0063] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0064] It should be noted that, in this document, the use of relational terms such as "first" and "second" is merely to distinguish one entity or operation from another, and does not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0065] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A high-strength anti-icing OPGW sag dynamic prediction and optimization system, characterized in that, Including the following steps: Data fusion module: continuously aggregates multi-source load response datasets along the line, aligns them by timestamp to construct a load-deformation joint sequence, and marks the ice distribution map of the conductor segment in the sequence; The tension analysis module calls the load-deformation joint sequence, uses piecewise degradation mapping to identify rapid tension transfer nodes, extracts the tension progression chain triggered by non-uniform icing, and writes it into the sag prediction buffer. This includes: extracting meteorological remote sensing data, conductor vibration acoustic emission data, fiber optic strain records, and icing distribution maps of conductor segments within a specific time window from the load-deformation joint sequence, constructing a time-space two-dimensional matrix; calculating the rate of change of fiber optic strain records for each conductor segment within the selected time window to obtain the strain change rate; identifying rapid tension transfer nodes by calculating the fluctuation intensity of the strain change rate and combining it with a threshold judgment; the tension analysis module also includes the following processing: Based on the tension fast transfer node and the normalized value of icing mass in the conductor segment icing distribution map, the tension transfer intensity between adjacent nodes is calculated, a tension transfer network is constructed, and a path search algorithm is used to extract the tension progression chain; the attributes of the tension fast transfer node and the tension progression chain are written into the sag prediction buffer as key-value pairs. Risk assessment module: The deformation trend of conductor caused by temperature drop is superimposed on the sag prediction buffer. By analyzing the spatial and temporal variation of tension, a cross-span sag risk envelope is formed, and the correction amount is written back to the tension rapid transfer node. Intervention and optimization module: compare the corrected cross-span sag risk envelope with the design safety gap. If a compression segment occurs, calculate the minimum strain equilibrium path based on the correction amount, allocate distributed ice melting power to weaken the peak tension, and simultaneously transmit the residual strain back. Prediction update module: Receives the combined load-deformation sequence after residual strain reorganization, performs time-series incremental assimilation to refresh the span sag risk envelope, and pushes the optimized span tension-sag guidance table to the scheduling end.

2. The high-strength anti-icing OPGW sag dynamic prediction and optimization system according to claim 1, characterized in that, The data fusion module processes the following: Meteorological remote sensing data, conductor vibration acoustic emission data, and fiber optic strain records were collected along the route, aligned with a unified timestamp, and cleaned and preprocessed to ensure data accuracy and consistency. The preprocessed meteorological remote sensing data and conductor vibration acoustic emission data were integrated into the load part of the load-deformation joint sequence, and the fiber optic strain records were integrated into the deformation part. Conductor segmentation information was introduced and the icing status was marked. The thickness and weight of ice on each segment of the conductor are estimated by using meteorological remote sensing data and fiber optic strain records. The results are then corrected to ensure consistency with the actual conditions and to generate segment ice distribution maps. The segment ice distribution maps are then embedded into the load-deformation joint sequence to form an extended joint sequence, which is stored in the database and supports fast querying in both time and space dimensions. 3.The high-strength anti-icing OPGW sag dynamic prediction and optimization system according to claim 1, characterized in that, The risk assessment module processes the following: Data from the tension fast transfer node and tension progression chain are extracted from the sag prediction buffer. Combined with temperature data, the thermal shrinkage of the conductor due to temperature drop is calculated and superimposed onto the sag prediction buffer. Based on the tension progression chain, the tension increment and duration between adjacent nodes are calculated to generate a step steepness field. The tension step robustness index is obtained by integrating along the span. The fiber strain record is decomposed into thermal shrinkage and viscoelastic relaxation components. The contributions of the thermal shrinkage and viscoelastic relaxation components to the conductor displacement are calculated to generate the offset rate. The low-temperature viscoelastic offset factor is calculated by combining the weighted average of icing mass.

4. The high strength anti-icing OPGW sag dynamic prediction and optimization system of claim 3, wherein, The risk assessment module also includes the following processing: The tension step robustness index and low-temperature viscoelastic offset factor are input into the support vector regression model to train and predict the sag intensification coefficient. The sag value of each span is calculated based on the catenary model to form an initial sag risk envelope. The initial sag risk envelope is adjusted in combination with the sag intensification coefficient. The correction amount of the tension rapid transfer node is calculated based on the adjustment difference and node weight, and the correction amount is written back to the sag prediction buffer.

5. The high strength anti-icing OPGW sag dynamic prediction and optimization system of claim 4, wherein, The intervention and optimization module processes the following: Extract the corrected span sag risk envelope and design safety clearance from the sag prediction buffer, identify the spans with excessive sag by calculating the difference between the span sag risk envelope and the design safety clearance, and generate a set of compressed sections. For the span in the compression segment set, the strain over-limit is calculated, a network is constructed based on the strain over-limit, and the minimum spanning tree algorithm is applied to generate the minimum strain equilibrium path.

6. The high strength anti-icing OPGW sag dynamic prediction and optimization system of claim 5, wherein, The intervention optimization module also includes the following processing: Based on the minimum strain equilibrium path and the total available melting power, the distributed melting power is calculated and allocated to the spans in the compression section set to reduce the tension peak. The real-time strain value after ice melting is recorded by fiber optic strain monitoring equipment, the residual strain is calculated and transmitted back to the sag prediction buffer to update the attribute fields of the tension rapid transfer node.

7. The high strength anti-icing OPGW sag dynamic prediction and optimization system of claim 6, wherein, The prediction update module processes the following: The residual strain data is received and integrated into the load-deformation joint sequence to generate an extended load-deformation joint sequence. Based on the extended load-deformation joint sequence, the residual strain data is used as observations to update the strain estimate of the conductor. The sag value of each span is calculated based on the updated strain estimate of the conductor, and the refreshed cross-span sag risk envelope is generated.

8. The high strength anti-icing OPGW sag dynamic prediction and optimization system of claim 7, wherein, The prediction update module also includes the following processing: Based on the refreshed cross-gear sag risk envelope, for each gear, the optimal tension value and the corresponding optimal sag value are found through numerical optimization methods, and an optimized cross-gear tension-sag guide table is generated and pushed to the scheduling terminal.

Images