Methodology for Long-Term Typhoon Disaster Risk Assessment of Power Distribution Network Towers under Climate Change Scenarios

Abstract

This invention discloses a method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios. The method includes: obtaining the annual typhoon formation frequency sequence and spatial formation location based on historical typhoon data; simulating the evolution of typhoon movement trajectories and maximum wind speeds near the typhoon center based on the formation locations of each typhoon; generating a high-resolution wind and rain-induced disaster field based on the simulated typhoon trajectory and intensity; extracting the disaster-induced field intensity index, and constructing a deep ensemble learning model by combining the geospatial environmental factors and equipment attribute parameters of the power distribution network towers, outputting the physical failure probability of each tower; and using climate model output data under different shared socio-economic pathways to drive the quantification of the medium- and long-term damage scale and spatial distribution of power distribution network towers under different carbon emission pathways. This invention achieves systematic quantification of the scale and spatial pattern of tower disasters under multiple greenhouse gas emission pathways by establishing a cross-scale refined mapping of climate-meteorological-physical damage.

Description

Technical Field

[0001] This invention belongs to the field of power system disaster prevention, mitigation and risk assessment technology, specifically relating to a method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios. Background Technology

[0002] With the non-stationary climate change caused by global greenhouse gas emissions, the intensity, path, and frequency of extreme weather events such as typhoons are exhibiting high uncertainty. Distribution network towers, as the weakest link in the resilience system of coastal power grids, are highly susceptible to clustered and cascading damage under extreme wind and rain conditions. However, existing technologies are mostly based on the assumption of historical stationary distribution, focusing on short-term early warning of single disasters, and lack a cross-scale dynamic mapping mechanism from large-scale long-term climate evolution to micro-asset damage. In the context of building new power systems, existing static assessment models are insufficient to support the iterative development of power grid planning and disaster prevention standards over decades. Bridging the assessment gap between macro-climate prediction and micro-asset risk has become a common problem urgently needing to be solved in the industry. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for assessing the risk of typhoon disasters to power distribution network towers under climate change scenarios, thereby solving the problem of predicting extreme fluctuations in electricity prices.

[0004] The present invention achieves the above objectives through the following technical solutions: In a first aspect, the present invention provides a method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios, the method comprising: Step 1: Modeling the Spatiotemporal Distribution of Typhoon Formation: Based on historical typhoon data, the probability distribution of the annual typhoon formation frequency is characterized by the negative binomial distribution, and a spatial probability density model of the typhoon formation location is established by kernel density estimation to obtain the annual typhoon formation frequency sequence and spatial formation location. Step 2: Typhoon Path and Intensity Dynamics Simulation: Based on the formation location of each typhoon, the beta of the large-scale environmental guiding wind field and beta drift is superimposed. The advection path model simulates the typhoon's movement trajectory; at the same time, an autoregressive model is used to simulate the evolution of the maximum wind speed near the typhoon's center. This autoregressive model is coupled with the sea surface temperature forcing effect, the vertical wind shear suppression effect, and the land surface friction dissipation effect. Step 3: Reconstruction of near-surface wind and rain disaster field: Based on the simulated typhoon trajectory and intensity, the gridded near-surface wind field is calculated using a parametric wind field model, and a nonlinear rainfall model based on the maximum wind speed and translation speed near the typhoon center is constructed to generate a high-resolution wind and rain disaster field. Step 4, Cross-scale Risk Coupling Assessment: Extract the intensity index of the disaster-causing field, combine it with the geospatial environmental factors of the distribution network towers and the equipment attribute parameters in the ledger, construct a deep integrated learning model, and output the physical failure probability of each tower; and use the climate model output data under different shared socio-economic paths to drive the quantification of the medium- and long-term damage scale and spatial distribution of distribution network towers under different carbon emission paths; wherein, the intensity index includes maximum wind speed, gusts and rainfall.

[0005] Following the above technical solution, the probability mass function of the negative binomial distribution is: ; In the formula: The number of typhoons formed, with values ​​ranging from 1 to 2. ; These are discrete parameters, reflecting the degree of dispersion of the interannual variability of typhoons; This is a success probability parameter, i.e., the tendency for typhoon formation; The distribution parameters are determined using the second-order moment estimation method based on historical observations, as follows: ; ; In the formula: and These are the parameter estimation solutions for the discrete parameters and the success probability parameters, respectively. This represents the average number of typhoons in historical years. The sample variance of the number of typhoons in historical years; For the synthesis period length is The simulation of the year generates an independent and identically distributed annual generation frequency sequence. : ; In the formula: It follows a negative binomial distribution.

[0006] Following the above technical solution, the spatial probability density model function for the typhoon formation location is: ; In the formula: Locations where historical typhoons formed; The number of historical samples is represented by i, and the sample number is represented by i. Bandwidth controls the smoothness. The kernel function is a two-dimensional Gaussian kernel; Selecting the optimal bandwidth from the bandwidth candidate set using cross-validation : ; In the formula: For cross-validation folds; For the first The likelihood function of the fold; Using the average center of gravity to measure the spatial concentration tendency of typhoons: for typhoons containing The average centroid of a typhoon along its path in both longitude and latitude. for: ; ; In the formula, and For the first The longitude and latitude of the typhoon's path; The standard deviation of the latitude and longitude distribution is used to assess the spatial dispersion and coverage of typhoons. The formula for calculating the standard deviation of the latitude and longitude distribution is shown below: ; ; In the formula: The standard deviation of the longitude distribution. represents the standard deviation of the latitudinal distribution.

[0007] Following the above technical solution, the beta-advection path model will translate the typhoon's shifting wind field. This refers to the large-scale environmental guiding wind field, i.e., the control wind field. With Beta Drift Linear superposition: ; Under known control wind field With Beta Drift Given the location of the typhoon's formation, the typhoon's movement path is obtained through time integration, and the typhoon's center position is updated progressively: ; ; In the formula: Longitude; For time; The radius of the Earth; , These are the typhoon shifting wind fields. The longitudinal and latitudinal components; Latitude; Among them, controlling the wind field Defined as the vertically layered mass-weighted average wind field from low-level inflow to high-level outflow; beta drift The Coriolis parameter is generated by varying with the zonal direction and is obtained by analyzing the evolution of the typhoon vorticity field. Considering two-dimensional divergence-free flow, the horizontal velocity field is derived from the stream function. express: ; ; In the formula: and These are the eastward and northward velocity components, respectively; Pointing east, Pointing north, velocity field vorticity Defined as: ; In order to control the wind field In a reference frame with shifted coordinates, the equation for non-divergent single-layer vorticity in the beta plane is: ; ; In the formula: Planetary vorticity gradient; It is the Earth's rotational angular velocity.

[0008] Following the above technical solution, the spatial correlation coefficient is used to assess the consistency between the recreated typhoon path and the historical typhoon path: ; In the formula: is the spatial correlation coefficient, with a value range of [-1, 1]; This represents the number of spatial grids. and The first Historical and recurring path point density values ​​for each grid point; and These represent the average density of historical and recurring path spaces, respectively.

[0009] Following the above technical solution, the autoregressive model uses a discrete-time autoregressive first-order process to describe the stochastic evolution of the maximum wind speed near the center. The discrete update form is as follows: ; In the formula: The autoregressive coefficient represents the memory of the maximum wind speed near the center of the typhoon. For a moment Maximum wind speed near the center; The amplitude of random disturbances reflects the cumulative effect of environmental uncertainty; For time step; It is Gaussian noise and follows a standard normal distribution; This is due to the sea surface temperature forcing effect. This is a vertical wind shear suppression effect; This is due to the dissipation effect of land surface friction; The sea surface temperature forcing effect is: ; In the formula: The sea surface temperature forcing coefficient; For a moment Sea surface temperature; The critical sea surface temperature threshold for typhoon development; The vertical wind shear suppression effect is: ; In the formula: The wind shear attenuation coefficient; For a moment Vertical wind shear; The threshold value for critical wind shear; The land surface friction dissipation effect is as follows: ; In the formula: The land attenuation coefficient; For a moment Land indicator coefficient; when The typhoon's center was located over land; when The typhoon's center is located over the ocean; The simulation will terminate when the maximum wind speed near the center of the typhoon decreases to below the tropical storm standard. ; In the formula: It meets the standards for tropical storms.

[0010] Following the above technical solution, typhoon wind field From tangential wind field Mobile wind farm With control of wind field Three parts linearly superimposed: ; The Holland model is used to calculate the tangential wind speed distribution. The specific formula is as follows: ; In the formula: Tangential wind speed; For a moment Maximum wind speed near the center; For a moment Radius of maximum wind speed near the center; This refers to the distance from the center of the typhoon. The wind speed profile shape index; time Tangential wind azimuth : ; In the formula: , For a moment Longitude and latitude of the target point; , For a moment Longitude and latitude of the typhoon center.

[0011] Following the above technical solution, a nonlinear enhancement function based on the maximum wind speed near the typhoon center is used to represent the original peak rainfall rate. : ; In the formula: Baseline rainfall rate; This is the rainfall amplification factor; For a moment Maximum wind speed near the center; For reference wind speed; It is a non-linear amplification index; Introducing a time-dependent retention enhancement factor : ; In the formula: This refers to the typhoon's translational speed; The moving speed threshold; This represents the maximum relative enhancement. Final peak rainfall rate for: ; In the formula: The lower limit of the rainfall rate; These are global calibration coefficients; Instantaneous rainfall rate Radial distance With time The structural parameters jointly determine that the radial distribution adopts Gaussian attenuation: ; Rainfall characteristic radius The calculation formula is: ; In the formula: This is the radial scale factor.

[0012] In a second aspect, the present invention provides a computer device / apparatus / system, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the first aspect.

[0013] Thirdly, the present invention provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0014] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: This invention provides a method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios. By coupling macro-climate evolution trends, mesoscale micro-meteorological wind field models, and micro-level tower structural physical vulnerabilities, a cross-scale digital twin simulation and damage risk quantification assessment strategy is constructed. This method integrates climate scenario-driven typhoon synthesis methods with structural damage prediction models. By establishing a cross-scale refined mapping of "climate-meteorological-physical damage," it achieves systematic quantification of the scale and spatial pattern of tower damage under multiple greenhouse gas emission pathways, filling the assessment gap between macro-climate evolution and micro-level asset damage, and possesses significant engineering and social value. Attached Figure Description

[0015] Figure 1 This is a block diagram of a method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under a climate change scenario, according to an embodiment of the present invention. Figure 2 This is a schematic diagram illustrating the typhoon risk formation path of power distribution network towers under a climate change scenario according to an embodiment of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments provided by this invention without inventive effort are within the scope of protection of this invention.

[0017] Obviously, the accompanying drawings described below are merely some examples or embodiments of the present invention. Those skilled in the art can apply the present invention to other similar scenarios based on these drawings without any inventive effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this invention, modifications to design, manufacturing, or production based on the technical content disclosed in this invention are merely conventional technical means and should not be construed as insufficient disclosure of the present invention.

[0018] In this invention, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this invention may be combined with other embodiments without conflict.

[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "a," "an," "an," "the," and similar words used in this invention do not indicate quantity limitation and may indicate singular or plural. The terms "comprising," "including," "having," and any variations thereof used in this invention are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms "connected," "linked," "coupled," and similar words used in this invention are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "A plurality" used in this invention refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships may exist; for example, "A and / or B" can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects have an "or" relationship. The terms "first," "second," and "third" used in this invention are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0020] This invention provides a method for assessing the medium- and long-term typhoon disaster risk of distribution network towers under climate change scenarios, specifically a method for assessing the medium- and long-term typhoon disaster risk of distribution network towers for the construction of climate-resilient power grids. This invention constructs a cross-scale digital twin simulation and damage risk quantification assessment strategy by coupling macro-climate evolution trends, mesoscale micro-meteorological wind field models, and micro-level tower structural physical vulnerabilities.

[0021] This invention aims to solve the following technical problems: 1. Lack of dynamic closed-loop cross-scale mapping mechanism: Existing technologies are mostly limited to disaster intensity assessment and cannot achieve a refined mapping from large-scale climate loads to microscopic tower physical failures. This invention establishes a complete "climate-meteorology-physical damage" closed loop through a deep learning architecture.

[0022] 2. Ignoring the non-stationary characteristics of climate evolution: In response to the new normal of typhoon extreme values ​​shifting upwards and unstable paths, this invention abandons the limitations of traditional patents that rely on historical static data. By introducing the CMIP6 multipath climate scenario, it solves the pain point of inaccurate quantification of damage over long periods in the future.

[0023] 3. Balancing the contradiction between physical mechanisms and computational efficiency: Existing dynamic models have excessively high computational costs. This invention employs a statistical dynamic hybrid synthesis mechanism, which, while ensuring physical consistency, supports efficient simulation of disaster sequences on a ten-thousand-year scale, enabling large-scale risk extrapolation.

[0024] 4. Lack of a multi-source heterogeneous data fusion and evaluation framework: By integrating multi-dimensional feature vectors such as meteorological, geographical, underlying surface roughness and equipment ledgers, a deep integrated learning evaluation system was constructed, which solved the problems of single feature and poor generalization of disaster loss prediction models.

[0025] As a forward-looking solution to extreme weather, this invention integrates a climate scenario-driven typhoon synthesis method with a structural damage prediction model. By establishing a cross-scale refined mapping of "climate-meteorological-physical damage", it achieves systematic quantification of the scale and spatial pattern of tower damage under multiple greenhouse gas emission pathways, filling the assessment gap between macro-climate evolution and micro-asset damage, and has significant engineering and social value.

[0026] In terms of commercialization and practical application, this invention possesses extremely high commercialization value and cross-industry application potential. Firstly, in the field of power engineering, this solution can be directly packaged into a modular medium- to long-term disaster prevention decision-making platform, serving the new power grid planning and reinforcement market worth hundreds of billions of yuan. Secondly, in the construction of smart city emergency response systems, this invention can serve as a foundational algorithm module, providing vulnerability projection data for urban critical infrastructure over a decades-long period.

[0027] The most explosive commercial application lies in the climate finance sector: the risk quantification results of this invention can directly serve the actuarial pricing of catastrophe insurance in the insurance industry, climate risk derivatives trading, and ESG compliance ratings for large enterprises. By accurately depicting the scale of asset damage under different emission pathways, it assists financial institutions in achieving precise penetration and hedging of climate risks. This cross-industry operation model of "technology + finance" greatly enhances the asset premium and operational activity of this invention in the intellectual property market.

[0028] like Figure 1 and Figure 2 As shown in the figure, the method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios according to an embodiment of the present invention specifically includes the following steps: Step 1: Characterizing the spatiotemporal distribution characteristics of typhoon formation based on statistical models.

[0029] First, in order to assess the long-term risks under future climate change scenarios, it is necessary to construct a set of disaster events that can reflect the statistical regularity and randomness of typhoon activity, and to model the annual frequency and spatial location of typhoons respectively.

[0030] (1) Typhoon annual formation frequency modeling: The number of typhoons formed is affected by factors such as El Niño and the Southern Oscillation, and large-scale circulation anomalies, exhibiting significant interannual variability. Therefore, a negative binomial distribution is used to characterize the annual frequency variation. This distribution can be regarded as a mixture of Poisson distribution parameters following a Gamma distribution. Statistically, it has a good fitting ability for overly discrete data. Its probability mass function is defined as: ; In the formula: The number of typhoons formed, with values ​​ranging from 1 to 2. ; These are discrete parameters, reflecting the degree of dispersion of the interannual variability of typhoons; This is the success probability parameter, i.e., the tendency for typhoon formation.

[0031] The second-order moment estimation method is used to determine the distribution parameters based on historical observations. The moment estimation equations are as follows: ; The parameter estimation solution is: ; ; In the formula: and These are the parameter estimation solutions for the discrete parameters and the success probability parameters, respectively. This represents the average number of typhoons in historical years. The sample variance of the number of typhoons in historical years.

[0032] For the synthesis period length is The simulation of the year generates an independent and identically distributed annual generation sequence. : ; In the formula: It follows a negative binomial distribution; iid indicates independent and identically distributed.

[0033] (2) Typhoon formation location modeling: The formation location of typhoons is spatially non-uniform due to factors such as sea surface temperature gradients. Kernel density estimation (KDE) is used for modeling. Latitude and longitude are projected onto the Lambert azimuth equal-area projection coordinate system to eliminate area distortion. Its probability density function is: ; In the formula: Locations where historical typhoons formed; This represents the number of historical samples. Bandwidth controls the smoothness. For the kernel function, use a two-dimensional Gaussian kernel. .

[0034] The optimal bandwidth is selected from the bandwidth candidate set through cross-validation. It can balance overfitting and undersmoothing issues, and improve the spatial realism of typhoon generation points: ; In the formula: For cross-validation folds; For the first The likelihood function of the fold.

[0035] The average center of gravity is used to measure the spatial concentration tendency of typhoons. For those containing The average centroid of a typhoon along its path in both longitude and latitude. for: ; ; In the formula, and For the first The longitude and latitude of the typhoon's path.

[0036] The standard deviation of the latitude and longitude distribution is used to assess the spatial dispersion and coverage of typhoons. The formula for calculating the standard deviation of latitude and longitude distribution is as follows: ; ; In the formula: The standard deviation of the longitude distribution. represents the standard deviation of the latitudinal distribution.

[0037] Step 2: Simulation of the dynamic mechanism of typhoon movement trajectory and intensity evolution.

[0038] Based on the results of step one, a dynamic model incorporating physical mechanisms was used to simulate the evolution of the typhoon's path and maximum wind speed near its center.

[0039] (1) Beta-advection path modeling driven by climate background wind field: Typhoon motion can be regarded as a linear superposition of two physical mechanisms: large-scale environmental guidance wind field, i.e. control wind field. The zonal variation of the geostrophic parameter causes self-drift, i.e., beta drift. . This refers to the advection component of typhoons along with large-scale circulation; The self-drift component caused by the vorticity effect induced by the variation of the Coriolis parameter with latitude. Typhoon translational wind field. It can be represented as: .

[0040] In the known and Given the location of the typhoon's formation, the typhoon's movement path can be obtained through time integration, and the typhoon's center position can be updated progressively. ; ; In the formula: Longitude; For time; The radius of the Earth is taken as 6.371 × 10⁶ m; , They are respectively The longitude and latitudinal components are positive in the east and north directions, respectively; Latitude.

[0041] The controlling wind field is defined as the vertically layered mass-weighted average wind field from the 850 hPa low-level inflow to the 200 hPa high-level outflow: ; To avoid the complexity of integrating the wind field for each layer vertically, a linear combination of a few wind layers is typically used to approximate the above equation. The control wind field is represented by a weighted linear combination of the 850 hPa and 200 hPa wind fields in the meridional and zonal directions, respectively: ; ; In the formula: , These are the zonal and meridional components of the wind field at the 200 hPa level, respectively. , These are the zonal and meridional components of the wind field at the 850 hPa level, respectively. , These are the weighting coefficients.

[0042] The beta drift caused by the zonal variation of the Coriolis parameter can be obtained by analyzing the evolution of the typhoon vorticity field. Considering a two-dimensional divergence-free flow, its horizontal velocity field can be derived from the stream function. express: ; ; In the formula: and These are the eastward and northward velocity components, respectively; Pointing east, Pointing north, velocity field vorticity Defined as: ; In order to control the wind field In a reference frame with shifted coordinates, the equation for non-divergent single-layer vorticity in the beta plane is: ; ; In the formula: Planetary vorticity gradient; It is the Earth's rotational angular velocity, taken as 7.292 × 10⁻⁵ rad / s.

[0043] Using spatial correlation coefficients to assess the consistency between recreated and historical typhoon tracks: ; In the formula: is the spatial correlation coefficient, with a value range of [-1, 1]; This represents the number of spatial grids. and The first Historical and recurring path point density values ​​for each grid point; and These represent the average density of historical and recurring path spaces, respectively. The closer the value is to 1, the higher the similarity between the historical and repeated typhoon movement paths.

[0044] (2) Intensity autoregressive modeling considering inland attenuation effect: The stochastic evolution of the near-center maximum wind speed is described by a discrete-time autoregressive first-order process, and the discrete update form is as follows: ; In the formula: The autoregressive coefficient represents the memory of the maximum wind speed near the center of the typhoon. For a moment Maximum wind speed near the center; The amplitude of random disturbances reflects the cumulative effect of environmental uncertainty; For time step; It is Gaussian noise and follows a standard normal distribution; This is due to the sea surface temperature forcing effect. This is a vertical wind shear suppression effect; This is due to the dissipation effect of land surface friction.

[0045] The sea surface temperature forcing effect is the nonlinear response of sea surface temperature to the maximum near-center wind speed of a typhoon. When the sea surface temperature is above a critical value, the typhoon draws energy from the ocean, and the maximum near-center wind speed increases; when it is below the critical value, the energy supply is insufficient, and the maximum near-center wind speed decreases. ; In the formula: The sea surface temperature forcing coefficient; For a moment Sea surface temperature; This refers to the critical sea surface temperature threshold for typhoon development.

[0046] The vertical wind shear suppression effect refers to the suppression of maximum near-center wind speeds by environmental vertical wind shear. Only when the wind shear exceeds a threshold does vertical wind shear disrupt the warm core structure of the typhoon and inhibit convection development. ; In the formula: The wind shear attenuation coefficient; For a moment Vertical wind shear; The threshold value is the critical wind shear.

[0047] The land surface friction dissipation effect is the decrease in maximum wind speed near the center of a typhoon after landfall due to increased surface friction and interruption of energy supply. ; In the formula: The land attenuation coefficient; For a moment Land indicator coefficient. When The typhoon's center was located over land; when The typhoon's center is located over the ocean.

[0048] The simulation will terminate when the maximum wind speed near the center of the typhoon decreases to below the tropical storm standard. ; In the formula: It meets the standards for tropical storms.

[0049] Step 3: Reconstruction of the spatial distribution characteristics of near-surface wind and rain-induced disaster sites.

[0050] Based on the generated trajectories and intensities, a wind-rain disaster field model with a physical resolution of grid level is constructed.

[0051] (1) Near-surface wind field modeling based on Holland parameters: Typhoon wind field From tangential wind field Mobile wind farm With control of wind field Three parts linearly superimposed: .

[0052] The Holland model was used to calculate the tangential wind speed distribution. This model derives the radial wind speed distribution based on the gradient wind balance equation and can capture the maximum near-center wind speed and the exponential decay characteristics in the outer region of the typhoon eyewall. The specific formula is as follows: ; In the formula: Distance from the typhoon center Tangential wind speed over distance; For a moment Radius of maximum wind speed near the center; This refers to the distance from the center of the typhoon. This is the wind speed profile shape index.

[0053] time Tangential wind azimuth : ; In the formula: , For a moment Longitude and latitude of the target point; , For a moment Longitude and latitude of the typhoon center.

[0054] (2) Modeling of heavy rainfall profiles integrating typhoon motion elements: Constructing a nonlinear enhanced base rainfall rate based on the maximum wind speed at the center: The radial rainfall structure of a typhoon is determined by its peak rainfall rate. Considering that the maximum wind speed near the typhoon center plays a dominant role in convective development and vertical transport, a nonlinear enhancement function based on the maximum wind speed near the typhoon center is used to represent the original peak rainfall rate. : ; In the formula: Baseline rainfall rate; This is the rainfall amplification factor; For reference wind speed; It is a non-linear amplification index.

[0055] Typhoon translation speed The time-varying nature of the velocity of movement alters the residence time in a given area, thus affecting rainfall intensity. To characterize the time-varying impact of movement velocity, a time-dependent residence enhancement factor is introduced. : ; In the formula: The moving speed threshold; This represents the maximum relative increase.

[0056] Final peak rainfall rate for: ; In the formula: The lower limit of the rainfall rate; These are the global calibration coefficients.

[0057] Instantaneous rainfall rate Radial distance With time The structural parameters jointly determine this. The radial distribution employs Gaussian attenuation. ; Rainfall characteristic radius The calculation formula is: ; In the formula: This is the radial scale factor.

[0058] Step 4: Cross-scale risk coupling assessment based on deep ensemble learning models.

[0059] After obtaining the maximum wind speed and rainfall of the gridded wind and rain field, cross-scale mapping from the climate disaster layer to the asset physical layer is realized.

[0060] First, the intensity indicators of the synthesized disaster field (maximum wind speed, gusts, rainfall, etc.), the geospatial environmental factors of the power distribution network towers (longitude, latitude, altitude, aspect, slope, underlying surface type, and surface roughness, etc.), and the attribute parameters of the tower ledger equipment are extracted. Among them, gusts refer to the instantaneous peak wind speed that occasionally breaks out suddenly on the basis of the average wind speed.

[0061] Secondly, a deep integrated learning architecture is constructed, which takes the fused multidimensional feature vector as input and optimizes the nonlinear structural response characteristics excited by the coupling of complex disaster-causing factors through the collaborative fusion of multi-base learners and meta-learners, and directly outputs the vulnerable physical failure probability of each base tower.

[0062] Finally, using the MIROC6 climate model output from the Sixth Coupled Model Intercomparison Project (CMIP6) as the driving source, future long-term sea surface temperature and vertical wind field parameters were extracted under four shared socio-economic pathways (SSP1-2.6, SSP2-4.5, SSP4-6.0, and SSP5-8.5). These parameters were then substituted into the above system for simultaneous solution, quantifying the damage scale, risk level evolution, and spatial migration characteristics of power grid towers under each carbon emission pathway, thus completing a closed-loop assessment framework for disaster prevention and mitigation.

[0063] In summary, the method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios provided by this invention not only has a significant technological advantage in basic algorithms compared with existing technologies, but also has huge market application potential in cross-industry commercialization, specifically reflected in the following five aspects: 1. A hybrid disaster simulation engine that balances physical mechanisms and computational efficiency was constructed, breaking through the computational bottleneck of ultra-long-cycle extreme climate simulation.

[0064] Results: The model successfully reproduced the complex movement trajectories and long-tail distribution characteristics of typical historical typhoons such as Talim and Mangkhut, with a spatial correlation coefficient as high as 0.92. Moreover, it can efficiently synthesize typhoon simulation sequences spanning up to 10,000 years under conventional computing power.

[0065] Causal reasoning: This invention completely abandons the limitations of single statistical models (which are difficult to reflect the non-stationary characteristics of climate) or single dynamic models (which are too computationally expensive and difficult to implement). It creatively utilizes the negative binomial distribution and KDE to characterize spatiotemporal probability and introduces a beta-advection path and intensity autoregressive model driven by large-scale climate background wind fields. Through the intervention of core physical mechanisms such as sea surface temperature forcing and vertical wind shear, the dynamic process of disaster evolution is accurately reproduced, achieving the best balance between simulation accuracy and commercial computing costs.

[0066] 2. Achieving a refined cross-scale mapping between macro-meteorology and micro-assets, and constructing a high-fidelity digital twin foundation for power grid disaster prevention.

[0067] Results: This invention can output the maximum wind speed and rainfall of wind fields at the 1 km × 1 km grid level in a refined manner, and successfully draw the spatial distribution map of the damage probability of power distribution network towers, clearly and intuitively identifying extremely high-risk areas (such as the southernmost tip of the Leizhou Peninsula).

[0068] Causal reasoning: The fundamental reason lies in the fact that this invention innovatively introduces a deep ensemble learning model after generating a high-resolution typhoon-induced disaster field. This architecture deeply integrates meteorological forcing, geographical environmental factors (such as altitude, slope, and surface roughness), and electrical parameters from tower records. Through collaborative optimization using multi-base learners, it accurately captures the nonlinear structural response characteristics excited by the coupling of complex disaster-causing factors, filling the key technological blind spots from meteorological disasters to physical damage.

[0069] 3. Accurately quantify the non-stationary derivative risks under multiple carbon emission pathways, providing a quantitative anchor for reconstructing climate defense benchmarks for new power systems.

[0070] Results: The numerical examples of this invention intuitively reveal the dramatic trends in disaster damage under different carbon emission backgrounds. Under the high emission scenario of SSP5-8.5, the maximum wind speed near the center of the typhoon, the maximum wind speed of the wind field, and the average rainfall increased by 17.49%, 19.27%, and 25.60% respectively compared with SSP1-2.6, resulting in a surge of 48.00% in the expected number of damaged power distribution network towers.

[0071] Causal reasoning: The dynamics module of this invention is directly driven by the output of the internationally authoritative CMIP6 (MIROC6 climate model), keenly capturing the effects of sea surface warming and accelerated water cycle caused by global warming. This ability to directly translate international climate consensus into grid capacity stress test indicators truly reflects the dynamic impact of non-stationary climate evolution on infrastructure.

[0072] 4. It possesses strong engineering feasibility and forward-looking planning capabilities, enabling targeted allocation and precise investment of disaster prevention resources for highly resilient power grids.

[0073] Results: The study not only quantified the absolute value of the damage scale, but also revealed the dynamic migration characteristics of the disaster spatial pattern (i.e., high-risk areas are advancing inland). The assessment results can be directly converted into work orders or decision-making reports, assisting power grid companies in identifying weak links in the power grid structure in advance in medium- and long-term (2026-2080) planning.

[0074] Causal reasoning: The final output of this invention is not obscure academic data, but an intuitive spatial distribution map of multi-scenario risk levels. This method transforms the highly complex coupled evolutionary process into engineering indicators that planners and emergency management departments can directly utilize, enabling targeted disaster prevention and repair material reserves and power grid reinforcement and upgrading, avoiding ineffective infrastructure investment, and possessing both high economic benefits and social public safety value.

[0075] 5. Transcending the boundaries of single power grid operation and maintenance, reconstructing a trillion-dollar business ecosystem encompassing climate projection, physical disaster prevention, and financial risk hedging.

[0076] Results: It completely breaks through the limitation of traditional power disaster prevention patents that can only be applied in regional power grid emergency repair scenarios, and successfully builds them into a standardized algorithm base with multi-industry compatibility, significantly improving the commercial operation activity and asset valuation value of the patents.

[0077] Causal reasoning: The high-precision, multi-scenario, long-term disaster risk spatial distribution data produced by this invention can not only be licensed as a SaaS service to State Grid and China Southern Power Grid for the upgrading and transformation of new distribution networks; its underlying algorithm can also be used as a core risk control model for cross-border technology transfer, providing authoritative technical support for the determination of catastrophe insurance rates by large insurance companies, disaster simulation of government smart city emergency brains, and ESG (climate risk disclosure) compliance rating of listed energy companies, greatly expanding the market size and commercial monetization channels of this patent.

[0078] Furthermore, the present invention also provides a computer device / apparatus / system, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method.

[0079] The present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0080] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or block diagrams.

[0081] These computer program instructions may also be stored in a computer-readable storage medium capable of directing a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more blocks in a block diagram.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, thereby providing steps for implementing the functions specified in one or more flowcharts and / or one or more blocks in a block diagram.

[0083] It should be noted that, depending on the implementation needs, the various steps / components described in this invention can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0084] Those skilled in the art will readily understand that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under a climate change scenario, characterized in that... The method includes: Step 1: Modeling the Spatiotemporal Distribution of Typhoon Formation: Based on historical typhoon data, the probability distribution of the annual typhoon formation frequency is characterized by the negative binomial distribution, and a spatial probability density model of the typhoon formation location is established by kernel density estimation to obtain the annual typhoon formation frequency sequence and spatial formation location. Step 2: Typhoon Path and Intensity Dynamics Simulation: Based on the formation location of each typhoon, the beta of the large-scale environmental guiding wind field and beta drift is superimposed. The advection path model simulates the typhoon's movement trajectory; at the same time, an autoregressive model is used to simulate the evolution of the maximum wind speed near the typhoon's center. This autoregressive model is coupled with the sea surface temperature forcing effect, the vertical wind shear suppression effect, and the land surface friction dissipation effect. Step 3: Reconstruction of near-surface wind and rain disaster field: Based on the simulated typhoon trajectory and intensity, the gridded near-surface wind field is calculated using a parametric wind field model, and a nonlinear rainfall model based on the maximum wind speed and translation speed near the typhoon center is constructed to generate a high-resolution wind and rain disaster field. Step 4, Cross-scale Risk Coupling Assessment: Extract the intensity index of the disaster-causing field, combine it with the geospatial environmental factors of the distribution network towers and the equipment attribute parameters in the ledger, construct a deep integrated learning model, and output the physical failure probability of each tower; and use the climate model output data under different shared socio-economic paths to drive the quantification of the medium- and long-term damage scale and spatial distribution of distribution network towers under different carbon emission paths; wherein, the intensity index includes maximum wind speed, gusts and rainfall.

2. The method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios as described in claim 1, characterized in that, The probability mass function of the negative binomial distribution is: ； In the formula: The number of typhoons formed, with values ​​ranging from 1 to 2. ; These are discrete parameters, reflecting the degree of dispersion of the interannual variability of typhoons; This is a success probability parameter, i.e., the tendency for typhoon formation; The distribution parameters are determined using the second-order moment estimation method based on historical observations, as follows: ； ； In the formula: and These are the parameter estimation solutions for the discrete parameters and the success probability parameters, respectively. This represents the average number of typhoons in historical years. The sample variance of the number of typhoons in historical years; For the synthesis period length is The simulation of the year generates an independent and identically distributed annual generation frequency sequence. : ； In the formula: It follows a negative binomial distribution.

3. The method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios as described in claim 1, characterized in that, The spatial probability density model function for the typhoon formation location is: ； In the formula: Locations where historical typhoons formed; The number of historical samples is represented by i, and the sample number is represented by i. Bandwidth controls the smoothness. The kernel function is a two-dimensional Gaussian kernel; Selecting the optimal bandwidth from the bandwidth candidate set using cross-validation : ； In the formula: For cross-validation folds; For the first The likelihood function of the fold; Using the average center of gravity to measure the spatial concentration tendency of typhoons: for typhoons containing The average centroid of a typhoon along its path in both longitude and latitude. for: ； ； In the formula, and For the first The longitude and latitude of the typhoon's path; The standard deviation of the latitude and longitude distribution is used to assess the spatial dispersion and coverage of typhoons. The formula for calculating the standard deviation of the latitude and longitude distribution is shown below: ； ； In the formula: The standard deviation of the longitude distribution. represents the standard deviation of the latitudinal distribution.

4. The method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios according to claim 1, characterized in that, The beta-advection path model translates the typhoon's shifting wind field. This refers to the large-scale environmental guiding wind field, i.e., the control wind field. With Beta Drift Linear superposition: ; Under known control wind field With Beta Drift Given the location of the typhoon's formation, the typhoon's movement path is obtained through time integration, and the typhoon's center position is updated progressively: ； ； In the formula: Longitude; For time; The radius of the Earth; , These are the typhoon shifting wind fields. The longitudinal and latitudinal components; Latitude; Among them, controlling the wind field Defined as the vertically layered mass-weighted average wind field from low-level inflow to high-level outflow; beta drift The Coriolis parameter is generated by varying with the zonal direction and is obtained by analyzing the evolution of the typhoon vorticity field. Considering two-dimensional divergence-free flow, the horizontal velocity field is derived from the stream function. express: ； ； In the formula: and These are the eastward and northward velocity components, respectively; Pointing east, Pointing north, velocity field vorticity Defined as: ； In order to control the wind field In a reference frame with shifted coordinates, the equation for non-divergent single-layer vorticity in the beta plane is: ； ； In the formula: Planetary vorticity gradient; It is the Earth's rotational angular velocity.

5. The method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios as described in claim 4, characterized in that, Using spatial correlation coefficients to assess the consistency between recreated and historical typhoon tracks: ； In the formula: is the spatial correlation coefficient, with a value range of [-1, 1]; This represents the number of spatial grids. and The first Historical and recurring path point density values ​​for each grid point; and These represent the average density of historical and recurring path spaces, respectively.

6. The method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios according to claim 1, characterized in that, The autoregressive model uses a discrete-time autoregressive first-order process to describe the stochastic evolution of the maximum wind speed near the center. The discrete update form is as follows: ； In the formula: The autoregressive coefficient represents the memory of the maximum wind speed near the center of the typhoon. For a moment Maximum wind speed near the center; The amplitude of random disturbances reflects the cumulative effect of environmental uncertainty; For time step; It is Gaussian noise and follows a standard normal distribution; This is due to the sea surface temperature forcing effect. This is a vertical wind shear suppression effect; This is due to the dissipation effect of land surface friction; The sea surface temperature forcing effect is: ； In the formula: The sea surface temperature forcing coefficient; For a moment Sea surface temperature; The critical sea surface temperature threshold for typhoon development; The vertical wind shear suppression effect is: ； In the formula: The wind shear attenuation coefficient; For a moment Vertical wind shear; The threshold value for critical wind shear; The land surface friction dissipation effect is as follows: ； In the formula: The land attenuation coefficient; For a moment Land indicator coefficient; when The typhoon's center was located over land; when The typhoon's center is located over the ocean; The simulation will terminate when the maximum wind speed near the center of the typhoon decreases to below the tropical storm standard. ； In the formula: It meets the standards for tropical storms.

7. The method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios according to claim 1, characterized in that, Typhoon wind field From tangential wind field Mobile wind farm With control of wind field Three parts linearly superimposed: ； The Holland model is used to calculate the tangential wind speed distribution. The specific formula is as follows: ； In the formula: Tangential wind speed; For a moment Maximum wind speed near the center; For a moment Radius of maximum wind speed near the center; This refers to the distance from the center of the typhoon. The wind speed profile shape index; time Tangential wind azimuth : ； In the formula: , For a moment Longitude and latitude of the target point; , For a moment Longitude and latitude of the typhoon center.

8. The method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under climate change scenarios according to claim 1, characterized in that, The original peak rainfall rate is represented by a nonlinear enhancement function based on the maximum wind speed near the typhoon center. : ； In the formula: Baseline rainfall rate; This is the rainfall amplification factor; For a moment Maximum wind speed near the center; For reference wind speed; It is a non-linear amplification index; Introducing a time-dependent retention enhancement factor : ； In the formula: This refers to the typhoon's translational speed; The moving speed threshold; This represents the maximum relative enhancement. Final peak rainfall rate for: ； In the formula: The lower limit of the rainfall rate; These are global calibration coefficients; Instantaneous rainfall rate Radial distance With time The structural parameters jointly determine that the radial distribution adopts Gaussian attenuation: ； Rainfall characteristic radius The calculation formula is: ； In the formula: This is the radial scale factor.

9. A computer device / equipment / system, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When executed by a processor, the computer program / instruction implements the steps of the method for assessing the medium- and long-term typhoon disaster risk of power distribution network towers under any one of claims 1 to 8.

Images