A method and system for assessing the coupled risks of multiple disasters in power tower groups

Abstract

This invention discloses a method and system for assessing the coupled risk of multiple disasters in a group of power transmission towers, relating to the field of transmission tower technology. The method includes: obtaining key feature information for risk assessment based on the spatiotemporal correlation matching terms of multi-source heterogeneous data streams; determining the degree of impact of each individual disaster on the power tower based on the key feature information, and obtaining the unilateral disaster impact result; identifying the interaction between individual disasters, and obtaining the coupled disaster impact result; identifying unknown risk results based on the expected structural response baseline of key components of the power tower; calculating the comprehensive risk value of the power tower by combining the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result; and obtaining the risk assessment result by combining the interaction paths between disasters. This application systematically characterizes the complex enhancement or weakening effects between disasters, ensuring the accuracy of the description of the coupled response of multiple disasters while avoiding real-time detailed simulation of all operating conditions and all components, thus balancing the accuracy and efficiency of risk assessment.

Description

Technical Field

[0001] This application relates to the field of power transmission tower technology, specifically a method and system for assessing the coupled risks of multiple disasters in a group of power transmission towers. Background Technology

[0002] Conventional risk assessment methods for transmission towers that focus only on the impact of a single disaster are insufficient when dealing with transmission towers in coastal areas. This is because the interaction between multiple disasters is not a simple linear superposition, but involves complex nonlinear relationships. For example, when typhoons and salt spray corrosion occur simultaneously, their damage to the towers may be far greater than the sum of their individual effects. It is difficult to accurately reflect the comprehensive risk level under the combined effect of multiple disasters. Often, the mutual reinforcement or weakening effects between disasters are ignored, resulting in a significant deviation between the assessment results and the actual interaction of disasters, thus affecting the accuracy and comprehensiveness of the risk assessment.

[0003] For example, the patent application number CN202410036882.0, entitled "A Method, Device, Equipment and Storage Medium for Assessing the Ice Resistance Risk of Transmission Towers," uses a numerical simulation model to calculate the icing load on the transmission tower under a preset ice thickness, obtains the stress response results, calculates the stress ratio of each member to determine the target member, and obtains the actual stress and strain of the target member to determine its ice resistance risk. This scheme can only assess the risk under icy and snowy weather conditions, and the assessment results are affected by unilateral influences, resulting in coupling errors. In response, some related technologies have proposed coupling risk analysis, such as the patent publication number CN117973107A, entitled "A Method and System for Assessing the Resilience of Transmission Tower Line System under Wind-Ice Coupling Disasters." This method analyzes the wind-induced dynamic response by obtaining icing and tension monitoring data on the transmission line, thereby analyzing the probabilistic risk of the transmission tower line system under wind-ice coupling. Although this scheme considers the coupling effect of wind direction and icing, it still cannot accurately assess the associated effects of typhoons, salt spray corrosion, and geological subsidence on power towers in coastal areas. Therefore, how to efficiently fuse and unify multi-source heterogeneous data with significant differences in structure, scale, and mode, and accurately characterize the complex nonlinear interaction mechanism between various disaster factors, so as to improve the power grid's ability to withstand disasters and the safety of daily maintenance, is a research topic that needs to be addressed. Summary of the Invention

[0004] This application addresses the problem that conventional assessment methods for power tower groups facing multiple intersecting disasters struggle to balance accuracy and efficiency due to insufficient characterization of nonlinear interactions between disasters. It proposes a multi-disaster coupled risk assessment method and system for power tower groups. This method acquires the impact of each individual disaster separately, further identifies the coupled nonlinear effects between disasters, detects unknown risks, and finally integrates the three types of impacts and traces the disaster action paths to form a comprehensive assessment. It systematically characterizes the complex enhancement or weakening effects between disasters, ensuring accuracy in describing the coupled response of multiple disasters while avoiding real-time, detailed simulation of all operating conditions and components, thus balancing the accuracy and efficiency of risk assessment.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: In a first aspect, embodiments of this application provide a method for assessing the coupled risks of multiple disasters affecting power tower groups, the method comprising: Key feature information for risk assessment is obtained based on the spatiotemporal correlation matching terms of multi-source heterogeneous data streams; the impact of each single disaster on the power tower is determined based on the key feature information to obtain the unilateral disaster impact result; the interaction between single disasters is identified to obtain the coupled disaster impact result; unknown risk results are identified based on the expected structural response baseline of the key components of the power tower; the comprehensive risk value of the power tower is calculated by combining the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result, and the risk assessment result is obtained by combining the action path between disasters.

[0006] This approach acquires key feature information through spatiotemporal correlation matching of multi-source heterogeneous data streams. This unifies and integrates various heterogeneous and spatiotemporally asynchronous data into a reliable input that can be aligned and traced back to the causes of risks. This reduces data redundancy and verification overhead in subsequent analysis stages from the source, while avoiding assessment distortion caused by data mismatch. By obtaining the results of single-sided disaster impacts, the complex multi-hazard coupling environment is decomposed into basic disaster components with clear physical boundaries and independent calculation paths, providing quantifiable benchmark values ​​for subsequent analysis. Obtaining the results of coupled disaster impacts is specifically designed to characterize the complex physical effects that cannot be expressed by linear superposition, such as the mutual enhancement or weakening of typhoons and salt spray corrosion, thus compensating for the core deficiency of conventional assessment methods in their insufficient characterization of coupling mechanisms. By identifying unknown risk results, it proactively captures abnormal patterns not yet covered by existing knowledge bases and physical models, caused by combinations of unknown disasters or progressive damage, thereby eliminating assessment blind spots. By integrating the three types of results and tracing the path of disaster action to form a comprehensive assessment, single-hazard analysis, nonlinear coupling analysis and the discovery of unknown risks are unified in the same framework. This not only systematically depicts the complex enhancement or weakening effects between disasters, but also discovers damage patterns that traditional models cannot cover through baseline monitoring. While ensuring the accuracy of the description of multi-hazard coupled response, it avoids real-time detailed simulation of all working conditions and all components, thus balancing the accuracy and efficiency of risk assessment.

[0007] Optionally, the step of obtaining key feature information for risk assessment based on the spatiotemporal correlation matching item of the multi-source heterogeneous data stream includes: obtaining multi-source heterogeneous data streams including meteorological data, environmental monitoring data, equipment monitoring data, and geometric attribute data of the power tower's location; performing time registration on the multi-source heterogeneous data streams with the spatial location of each power tower as the center; extracting basic feature information based on the time-registered multi-source heterogeneous data streams; combining the power tower's geometric feature information with the physical principles of mass transfer and fluid dynamics to obtain local microenvironmental features; and using the basic feature information and the local microenvironmental features as key feature information.

[0008] Optionally, the step of determining the degree of impact of each single disaster on the power tower based on the key feature information and obtaining the unilateral disaster impact result includes: calculating the wind pressure on the tower body and conductors based on wind speed characteristics, wind direction characteristics and power tower structural parameters to obtain the typhoon impact result; calculating the corrosion rate and cumulative damage degree based on salt spray concentration and temperature and humidity information combined with power tower material properties to obtain the salt spray corrosion impact result; and assessing the tower foundation settlement trend based on geological feature data and meteorological feature data to obtain the geological settlement impact result; wherein the typhoon impact result, the salt spray corrosion impact result and the geological settlement impact result form the unilateral disaster impact result.

[0009] Optionally, identifying the interactions between individual disasters and obtaining the results of coupled disaster impacts includes: based on a preset causal association rule base, identifying the coupling effects between individual disasters according to the key feature information and the results of the single-sided disaster impacts, wherein the coupling effects include enhancement effects and weakening effects; and calculating the corresponding coupling effect quantification value using a nonlinear influence function based on the enhancement effect / weakening effect, thereby obtaining the results of coupled disaster impacts.

[0010] Optionally, the identification of unknown risk results based on the expected structural response baseline of key components of the power tower includes: establishing the expected structural response baseline of the key components of the power tower under corresponding environmental conditions; calculating in real time the deviation between the actual structural response value of each key component and the expected structural response baseline, continuously tracking the temporal changes of the deviation to identify persistent abnormal deviations below a threshold; forming an abnormal pattern fingerprint based on the spatiotemporal characteristics of the persistent abnormal deviation, wherein the spatiotemporal characteristics include at least the duration of the abnormality, the trend of change, and the spatial distribution location; and combining the abnormal pattern fingerprint with the key feature information to determine the causal path and novel damage mode of the current abnormal deviation in order to obtain unknown risk results.

[0011] Optionally, establishing the expected structural response baseline for key components of the power tower under corresponding environmental conditions includes: extracting a list of key components of the power tower based on the key feature information, and matching corresponding response indicators for each component; determining the unilateral response benchmark value of the response indicator under a single disaster based on the calculation process parameters of the unilateral disaster impact result, wherein the unilateral response benchmark value of each component and each response indicator under the individual action of each single disaster corresponds one-to-one with the unilateral disaster impact result; obtaining the linear sum of the unilateral response benchmark values ​​of all single disasters involved in the coupling, determining the corresponding coupling correction coefficient according to the coupling mode to correct the linear sum, and obtaining the expected nominal response value under multi-disaster coupling conditions; calculating the fluctuation standard deviation of the expected nominal response value, and determining the initial response baseline interval based on the expected nominal response value; calibrating the initial response baseline interval based on the historical health data of the power tower during periods of abnormal operation, and obtaining the expected structural response baseline, wherein the expected structural response baseline includes the boundaries of the theoretical stress accumulation rate, vibration frequency, strain amplitude, fatigue damage increment, and tower foundation settlement rate.

[0012] Optionally, the step of combining the abnormal pattern fingerprint and the key feature information to determine the causal path and novel damage mode of the current abnormal deviation to obtain unknown risk results includes: extracting and aggregating spatiotemporally aligned correlation clues from the key feature information for the abnormal pattern fingerprint, the correlation clues including abnormal pattern features, environmental conditions, local microenvironment features, and component attribute features; calling a preset basic action mechanism library to match the abnormal pattern features, environmental conditions, local microenvironment features, and component attribute features respectively to obtain candidate causal paths that are causally correlated with the abnormal pattern fingerprint; selecting the optimal path according to the credibility of the candidate causal paths to identify the corresponding novel damage mode; using the correlation clues as input to the damage evolution correction model corresponding to the novel damage mode, calculating the risk increment of the novel damage mode relative to the coupled disaster impact result, and using the risk increment as the unknown risk result.

[0013] Optionally, the step of calculating the comprehensive risk value of the power tower by combining the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result, and obtaining the risk assessment result by combining the interaction path between disasters, includes: weighting the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result to calculate the comprehensive risk value of the power tower group; comparing the comprehensive risk value with the risk level threshold to determine the risk level; and generating a full-link risk assessment report by combining the causal path corresponding to the unknown risk result and the coupled causal path corresponding to the coupling effect between single disasters.

[0014] Optionally, the step of weightedly calculating the comprehensive risk value of the power tower group by integrating the unilateral disaster impact results, the coupled disaster impact results, and the unknown risk results includes: weightedly calculating the unilateral disaster impact results, the coupled disaster impact results, and the unknown risk results based on the individual power tower components to obtain a single-component-level comprehensive risk value; weightedly integrating the single-component-level comprehensive risk values ​​of each power tower component based on the component importance weights to obtain a power tower-level comprehensive risk value; and weightedly integrating the power tower-level comprehensive risk values ​​of each power tower within the region based on the spatial location importance weights to obtain a comprehensive risk value for the power tower group.

[0015] Secondly, embodiments of this application provide a multi-hazard coupled risk assessment system for power tower groups, comprising: The feature acquisition module is used to acquire key feature information for risk assessment based on the spatiotemporal correlation matching items of multi-source heterogeneous data streams; the disaster identification module is used to determine the degree of impact of each single disaster on the power tower based on the key feature information, and to acquire the unilateral disaster impact result; to identify the interaction between single disasters and to acquire the coupled disaster impact result; and to identify the unknown risk result based on the expected structural response baseline of the power tower's key components; the risk assessment module is used to calculate the comprehensive risk value of the power tower by combining the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result, and to acquire the risk assessment result by combining the action path between disasters.

[0016] The beneficial effects of this application are: 1. By judging the impact of each single disaster on power towers, the complex multi-hazard environment is decomposed into multiple independent disaster components with clear boundaries. This not only ensures the accuracy of the calculation of each disaster effect, but also avoids the high computational cost of directly dealing with multi-field coupling problems, providing a reusable benchmark for coupling analysis. 2. By identifying the coupled disaster effects between individual disaster interactions, this approach replaces the conventional simplistic method that relies solely on linear superposition. It accurately captures the actual physical effects of enhancement or weakening between disasters. The identification process itself targets the coupling mechanism, eliminating the need for real-time coupled numerical simulations of all operating conditions and components. Therefore, it improves the accuracy of coupling characterization while reducing computational consumption. Furthermore, by actively detecting anomalies in the actual response that deviate from expectations using baseline comparison of expected structural responses, the assessment scope is expanded from known disaster patterns to previously unknown damage forms, significantly improving the comprehensiveness and accuracy of the assessment. Finally, by combining unilateral, nonlinear, and unknown risk results to calculate a comprehensive risk value and trace the disaster's action path, it achieves a complete fusion of known independent effects, known coupled effects, and potential unknown damage. This results in a clear causal chain for outputting assessment conclusions, ensuring the overall accuracy of the risk quantity while avoiding the efficiency loss caused by repeated recalculations due to fragmented information. The entire process adopts a progressive structure of "decomposing the baseline, identifying couplings, monitoring the unknown, and fusing backtracking" to analyze multidimensional complex problems into functional blocks that can be executed in parallel or step by step. Each block is implemented with a targeted analytical method rather than full-domain simulation, thereby ensuring the efficient operation of the assessment process while accurately depicting the nonlinear action mechanism of multiple disasters. Attached Figure Description

[0017] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings. The drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0018] Figure 1A flowchart of a method for assessing the coupled risks of multiple disasters in a power tower group, provided in an embodiment of this application; Figure 2 This is a schematic diagram of a multi-hazard coupling risk assessment system module for power tower groups provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of this application and are only used to explain this application. They do not limit the scope of protection of this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] Example 1: As Figure 1 As shown, a method for assessing the coupled risks of multiple disasters in a power tower group includes steps S1-S5, wherein: S1. Obtain key feature information for risk assessment based on the spatiotemporal correlation matching items of multi-source heterogeneous data streams; In an optional embodiment, step S1 includes: Acquire multi-source heterogeneous data streams, including meteorological data, environmental monitoring data, equipment monitoring data, and geometric attribute data of the power tower's location; Time registration is performed on multi-source heterogeneous data streams, centered on the spatial location of each power tower. Based on time-registered multi-source heterogeneous data streams, basic feature information is extracted. Combined with the geometric feature information of the power tower, local micro-environment features are obtained based on the physical principles of mass transfer and fluid mechanics. The basic feature information and the local micro-environment features are used as key feature information.

[0021] Specifically, meteorological data includes weather station data such as wind speed, wind direction, rainfall, temperature, and humidity; environmental monitoring data includes environmental monitoring data such as salt spray concentration, soil moisture content, and soil geological parameters; equipment monitoring data includes static attribute data of equipment such as power tower structural drawings, material properties, design parameters, historical maintenance records, and geological exploration reports, as well as high-frequency sampling data from sensors such as stress, vibration frequency, strain, and settlement rate of key power tower components; and power tower geometric attribute data includes local geometric features such as connection angles of power tower components and geometric dimensions of local grooves / gaps, as well as material attribute data such as corrosion fatigue curves, fracture mechanics parameters, and anti-corrosion coating status. The above data can be accessed in real-time / in batches through standardized interfaces such as the MQTT protocol and RESTful API. Preliminary parsing is performed according to data type to remove outliers, fill in missing values, and distinguish between numerical data, categorical data, and spatial geometric data, outputting a structured raw dataset with timestamps, geographic coordinates, equipment IDs, and sensor IDs.

[0022] Specifically, using the geographical coordinates of the power tower as the spatial reference and the minimum sampling frequency as the time reference, methods such as linear interpolation and nearest neighbor matching are used to achieve dual alignment of timestamps and spatial locations for data with different sampling frequencies and spatial coverage areas. This ensures that all data from the same power tower / component at the same time segment correspond one-to-one. Furthermore, macroscopic features (i.e., basic feature information) and local micro-environment features strongly correlated with power tower risks are extracted.

[0023] Specifically, the basic feature information includes meteorological features (average wind speed, maximum gust, cumulative rainfall, temperature and humidity time series, etc. over a period of time), environmental features (average salt spray exposure, cumulative deposition rate), and structural features (dominant vibration frequency and amplitude, peak stress and stress cycle number, strain change amplitude); local microenvironment features include rainwater retention and water volume in local grooves / cracks of the power tower, water evaporation rate, salt spray deposition rate, and local corrosion status; these are obtained through mass transfer and fluid mechanics principles, specifically: based on rainfall and the geometry of local grooves / cracks, the rainwater retention and water volume at specific locations on the structure are calculated; combined with wind speed and ambient temperature and humidity, the water evaporation rate is calculated, and the deposition rate of salt spray particles in the water is calculated simultaneously; the salt ion enrichment concentration in the local water is iteratively calculated, and the local corrosion microenvironment features are output.

[0024] In this embodiment, macroscopic monitoring data is mapped onto actual disaster-causing conditions such as wind pressure and salt spray deposition at the microscopic scale of the power tower location. This ensures that the input for subsequent disaster impact calculations is closer to the actual environmental conditions at the tower location, effectively reducing assessment bias caused by insufficient spatial resolution, thereby improving the accuracy of risk assessment at the data source.

[0025] S2. Based on the key feature information, determine the degree of impact of each single disaster on the power tower and obtain the single-sided disaster impact result.

[0026] In an optional embodiment, step S2 includes: Based on wind speed characteristics, wind direction characteristics and power tower structural parameters, the wind pressure on the tower body and conductor is calculated to obtain the typhoon impact results; Based on the information of salt spray concentration and temperature and humidity, combined with the material properties of the power tower, the corrosion rate and cumulative damage were calculated to obtain the results of the impact of salt spray corrosion. The settlement trend of the tower foundation is assessed based on geological and meteorological characteristic data, and the impact of geological settlement is obtained. Among them, the typhoon impact results, salt spray corrosion impact results, and geological subsidence impact results form a unilateral disaster impact result.

[0027] Specifically, the typhoon impact results are used to assess the risks of deformation and conductor misoperation; the salt spray corrosion impact results are used to assess the risks of reduced component strength and shortened fatigue life; and the geological settlement impact results are used to assess the tower foundation settlement trend and the risk of reduced bearing capacity. The assessment is based on soil moisture content, rainfall, and settlement monitoring data. For example, if the soil moisture content continues to increase and the settlement rate increases when rainfall decreases, the tower foundation settlement trend will increase, and there is a risk of geological settlement.

[0028] In this embodiment, quantitative impact models directly associated with key characteristic information were established for the typical disaster types faced by power tower clusters in coastal areas. This gives the degree of independent effect of each single disaster on power towers a clear physical meaning and calculable indicators, providing a solid and comparable benchmark impact value for subsequent identification of coupling effects and comprehensive risk measurement. This overcomes the shortcomings of previous single risk assessments where results were difficult to integrate.

[0029] S3. Identify the interactions between individual disasters and obtain the results of coupled disaster impacts.

[0030] In an optional embodiment, step S3 includes: Based on a pre-defined causal association rule base, the coupling effect between single disasters is identified by combining the key feature information with the single-sided disaster impact results. The coupling effect includes enhancement effect and weakening effect. Based on the enhancement effect / weakening effect, the corresponding coupling effect quantification value is calculated using a nonlinear influence function, thereby obtaining the coupling disaster impact result.

[0031] Specifically, the causal rule base contains the correlation between disaster types and their corresponding data features and single-disaster impact results. For the identification of enhancement effects: causal rules such as "typhoon wind load + structural resonance accelerates bolt fatigue damage" and "salt spray corrosion + periodic wind load accelerates component fatigue failure" are used to call nonlinear enhancement functions to calculate the quantitative value of the coupling enhancement effect, representing the pure risk increment of the enhanced coupling (positive value, representing risk amplification). For the identification of mitigation effects: causal rules such as "saturated rainfall + short-term mitigation of geological subsidence impact in soft soil foundations" are used to calculate the quantitative value of the coupling mitigation effect, representing the pure risk reduction of the weakened coupling (negative value, representing risk reduction). The set of the quantitative values ​​of the coupling enhancement effect and the coupling mitigation effect is constructed to obtain the coupled disaster impact results.

[0032] Specifically, the quantization value of the coupling enhancement effect Quantification of coupling weakening effect The calculation formula is expressed as: ; ; in, To enhance the coupling reference coefficient, To weaken the coupling baseline coefficient, the baseline coefficient is calibrated by the operating conditions; m is the number of disaster types involved in the coupling (e.g., typhoon + salt spray corrosion, m=2); n is the number of disaster combinations involved in weakening the coupling. Let be the enhanced sensitivity coefficient for the i-th type of disaster; Let be the standardized intensity index for the i-th type of disaster; The basic risk value for the i-th type of disaster is the basic risk value for the single disaster (that is, the standardized result of the single-sided disaster impact is used as the basic risk value for the single disaster). Let be the weakening strength coefficient of the j-th coupling group; The degree to which the triggering condition of the j-th coupling is satisfied (dimensionless, 0 to 1); Let be the attenuation exponent of the j-th coupling group. The exponential attenuation form characterizes the nonlinear attenuation magnitude, and a negative result indicates a reduction in risk. is the inhibition factor of the j-th coupling group (dimensionless, 0 to 1).

[0033] In this embodiment, the coupling type is determined by causal rules rather than purely statistical data relationships, making the capture of nonlinear synergistic or antagonistic effects between disasters such as typhoons and salt spray corrosion physically interpretable. The introduction of nonlinear functions directly produces quantitative coupling effect values. Compared with relying on single linear superposition or real-time coupling calculations of multiple physics fields, this not only more accurately characterizes the actual risk amplification or mitigation degree under the combined effects of multiple disasters, but also significantly reduces the computational complexity of the assessment and improves the assessment efficiency.

[0034] S4. Identification of unknown risk results based on the expected structural response baseline of key components of the power tower.

[0035] In an optional embodiment, step S4 includes: Establish baselines for the expected structural response of key components of power towers under corresponding environmental conditions; The deviation between the actual structural response value of each key component and the expected structural response baseline is calculated in real time, and the temporal changes of the deviation are continuously tracked to identify persistent abnormal deviations below a threshold. An anomaly pattern fingerprint is formed based on the spatiotemporal characteristics of the continuous abnormal deviation, wherein the spatiotemporal characteristics include at least the duration of the anomaly, the trend of change, and the spatial distribution location. By combining the abnormal pattern fingerprint with the key feature information, the causal path and novel damage mode of the current abnormal deviation can be determined to obtain unknown risk results.

[0036] Specifically, the expected structural response baseline is a dynamic temporal boundary with confidence intervals, segmented by component and index, rather than a fixed value. It characterizes the reasonable range of structural response that a target component with normal performance and no hidden damage should produce under the current real-time environment, disaster conditions, and component attributes, based on verified physical laws and preset coupling rules. Spatially, the expected structural response baseline is specific to key components of the power tower (such as main tower body materials, connecting bolts / node plates, tower foundation, conductor hardware, etc.), corresponding one-to-one with anomaly monitoring points. Temporally, it is synchronized with the minimum time step of the spatiotemporal alignment stage of key feature information, updating in real time according to operating conditions.

[0037] Specifically, the anomaly pattern fingerprint refers to the anomaly data dictionary formed by the duration, trend, and spatial distribution of the anomaly. Based on the anomaly pattern fingerprint, key feature information is aggregated to gather highly correlated full-dimensional clues, including environmental features, local microenvironment features, component geometric features, and material property features. This allows for the reverse derivation of the causal chain and novel damage mode that best explains the current anomaly deviation, such as "long-term compound disaster → water accumulation in a specific structure → local high salt enrichment → crevice corrosion + stress corrosion cracking (SCC) combined damage". The additional risk value brought by this novel damage mode is further quantified to obtain the unknown risk result.

[0038] In this embodiment, by identifying and quantifying the impact of unknown risks, the assessment system is not only able to calculate the risks under known combinations of disasters, but also has the ability to proactively perceive unknown, progressive, or atypical damage patterns. That is, regardless of how disasters are combined or whether there are unrecognized coupling effects, as long as the component response deviates from the known range of normal and known damage evolution, it can be captured in time and its source can be traced. This significantly enhances the comprehensiveness of risk assessment and the ability to discover new disaster-causing mechanisms in the complex context of multiple intertwined disasters, without the need to enumerate all possible damage modes in advance, thus maintaining assessment efficiency.

[0039] In an optional embodiment, establishing the expected structural response baseline for key components of the power tower under corresponding environmental conditions includes: Based on the key feature information, a list of key components of the power tower is extracted, and a corresponding response index is matched for each component. The calculation process parameters based on the single-sided disaster impact result determine the single-sided response benchmark value of the response index under the corresponding single disaster, wherein the single-sided response benchmark value of each component and each response index under the individual action of each single disaster corresponds one-to-one with the single-sided disaster impact result; Obtain the linear sum of the single-sided response baseline values ​​of all single disasters involved in the coupling, determine the corresponding coupling correction coefficient according to the coupling mode, and correct the linear sum to obtain the expected nominal response value under the multi-disaster coupling condition. Calculate the standard deviation of the fluctuation of the nominal value of the expected response, and determine the initial response baseline range based on the nominal value of the expected response; The initial response baseline range is calibrated based on historical health data from periods of normal operation of the power tower to obtain the expected structural response baseline, which includes the boundaries of theoretical stress accumulation rate, vibration frequency, strain amplitude, fatigue damage increment, and tower foundation settlement rate.

[0040] In this embodiment, key feature information provides the working condition boundary conditions for baseline calculation, including real-time meteorological environmental parameters, local microenvironment parameters, and component geometric / material / static property parameters; the single-sided disaster impact result provides the single-disaster driven response benchmark value, that is, the single-disaster impact result is reused to calculate the physical model, and the intermediate structural response output of the model is extracted as the basic component of the baseline; by reusing the coupling enhancement effect quantification value and the coupling weakening effect quantification value, the corresponding correction coefficient is obtained, and the coupling linear superposition of the single-sided response benchmark value is corrected to obtain the expected response nominal value considering disaster coupling.

[0041] Specifically, a list of key load-bearing components and vulnerable components of the power tower is extracted from key feature information, including: main tower body materials, connecting bolts / node plates, tower foundation, crossarm hardware, flanges, etc., which correspond one-to-one with the installation points of structural sensors; corresponding core response indicators are matched for each component, which are completely aligned with the sensor monitoring indicators, as shown in Table 1. Table 1 Component-Indicator Matching Table Connecting bolts / node plates Stress accumulation rate, strain amplitude, fatigue damage increment Strain gauges, stress sensors Main materials of the tower body Vibration characteristic frequency, transverse strain amplitude Vibration accelerometer, strain sensor Tower foundation Settlement rate, uneven settlement difference Settlement sensor, tilt sensor For each response index, a physical calculation model is matched that is completely consistent with the results of both single-sided disaster impact and coupled disaster impact. Specifically, for calculating the baseline response value of components driven by a single disaster, the following steps are taken: extract the operating parameters of the current time section (such as wind speed, salt spray concentration, soil moisture content, component geometry, material strength, etc.), and extract the intermediate structural response output from the physical model used to calculate the results of single-sided disaster impact as the baseline response value under single disaster. This is because the results of single-sided disaster impact are essentially the quantified results of the "structural response baseline value" after risk level mapping; therefore, the nominal value of the structural response can be extracted in reverse / synchronously using the same physical model. For example, when calculating the stress amplitude of connectors under typhoon wind load, since the tower wind pressure has already been calculated when calculating the typhoon impact results, this wind pressure value is directly reused here and substituted into the structural mechanics model to calculate the theoretical stress amplitude of the target connector. The calculation formula is expressed as: ; in, This represents the stress amplitude of the connector under typhoon wind load. For windward area, The wind pressure on the tower is given by [value], and L is the lever arm length. Let be the net section modulus of the component, and s be the safety factor. Simultaneously calculate the baseline values ​​for stress accumulation rate and fatigue damage increment under a single disaster, serving as the response baseline for a single typhoon disaster. Similarly, using the same calculation logic, calculate the response baseline for each component and each index under the individual effects of salt spray corrosion and geological subsidence, serving as the unilateral response baseline value.

[0042] Furthermore, enhancement / weakening correction coefficients are calculated based on the quantized values ​​of the coupling enhancement effect and the coupling weakening effect. The calculation formula is expressed as follows: ; ; in, , respectively, are the enhancement / weakening correction coefficients for the j-th group of coupling under the i-th type of disaster, and m is the number of disaster types participating in this group of coupling; Quantification value of coupling enhancement effect, Quantification of the coupling weakening effect; This represents the pure risk increment that will enhance coupling. This is converted to an amplification factor relative to the linear sum of individual disaster risks, with a value greater than 1. A larger value indicates a stronger enhancement effect. The specific derivation process is as follows: Total Coupled Risk = Linear Sum of Individual Disaster Risks + Pure Enhancement Increment = Correction factor = total coupled risk / single-disaster linear sum = . This indicates a reduction in the pure risk of weakening coupling. This is converted to a reduction factor relative to the linear sum of single-hazard risks, with a value ranging from (0,1). A smaller value indicates a stronger mitigation effect. The specific derivation process is as follows: Total Coupled Risk = Linear Sum of Single-Hazard Risks + Pure Mitigation Reduction = Correction factor = total coupled risk / single-disaster linear sum = .

[0043] Furthermore, the sum of the baseline values ​​of the unilateral responses of all participating single disasters is corrected based on the correction coefficient. The correction calculation formula is expressed as follows: ; in, To account for the nominal value of the expected response to the interaction of multiple disasters, Let i be the baseline value for the one-sided response to the i-th type of disaster. This is the sum of the individual disaster response baseline values ​​for all disasters participating in this group of coupling events. For correction factor, It should be noted that the same set of disaster coupling can only occur in one of the enhancement or weakening modes, and cannot occur simultaneously. and The system will first determine which mode the coupling belongs to based on the preset causal association rules, and then call the corresponding formula to calculate the correction coefficient.

[0044] In some possible embodiments, if a component is simultaneously affected by multiple independent coupling relationships (such as the simultaneous existence of "typhoon-salt spray corrosion" enhancing coupling and "rainfall-geological subsidence" weakening coupling), the nominal expected response value of each independent coupling is weighted and summed according to the weight coefficient of the coupling path to obtain the final nominal expected response value. The weighted calculation formula is as follows: ; in, Let P be the nominal value of the expected response under multiple independent coupling effects, where P is the total number of independent coupling paths affecting the component. The weight coefficient of the p-th coupling path (determined by the causal association rule) is given. This is the correction coefficient corresponding to the p-th coupling path. This is the sum of the single-disaster response baseline values ​​corresponding to the p-th coupling path.

[0045] Furthermore, considering sensor measurement error, model calculation error, and random fluctuation error under operating conditions, the standard deviation of the baseline fluctuation is obtained. Specifically, the nominal accuracy of the sensor is extracted from the static attributes of the equipment, and the sensor measurement standard deviation is calculated; the model standard deviation is obtained by statistically analyzing the deviation between the calculated and measured values ​​of the expected response nominal value under historical no-abnormal operating conditions; the operating condition standard deviation is obtained by statistically analyzing the natural fluctuation range of parameters such as wind speed and rainfall; and the three types of standard deviations are added together to obtain the standard deviation of the expected response nominal value.

[0046] Furthermore, based on the standard deviation of the expected response nominal value, the expected response nominal value of each component is extended to a baseline boundary with confidence intervals (initial response baseline interval). For example, the upper and lower boundaries of the baseline are determined based on the 95% confidence interval of the normal distribution, forming a complete expected response baseline. The upper limit of the baseline is: ,in, The baseline lower limit is the standard deviation of the expected nominal response (i.e., the sum of the three types of standard deviations mentioned above); .

[0047] Furthermore, based on historical health data of power towers during periods of abnormal operation (such as stable operating conditions with wind speeds < 15 m / s, no heavy rainfall, no disaster warnings, and no maintenance records), the correction coefficients and the standard deviation of the nominal expected response values ​​are continuously optimized to obtain the final expected structural response baseline. If a systematic deviation between measured and nominal values ​​occurs over a long period, the physical calculation parameters of the unilateral disaster impact results and / or coupled disaster impact results are simultaneously corrected to optimize the correction coefficients, ensuring that the expected response baseline closely matches the actual response of healthy components.

[0048] In this embodiment, the expected structural response baseline of the key components of the power tower is calculated, thereby dynamically determining the reasonable response range under all current operating conditions (including disaster coupling). Compared to a fixed baseline, the expected structural response baseline is adjusted in real time with changes in operating conditions, always matching the current disaster coupling state, ensuring the accuracy and timeliness of the risk assessment results. Simultaneously, the process of constructing the expected structural response baseline organically integrates single-hazard mechanical analysis, multi-hazard nonlinear coupling correction, and actual on-site health data. This ensures that the baseline reflects both the physical expectations of multiple hazard superposition and adapts to the actual state and normal aging drift of the specific power tower. This provides a high-precision benchmark for identifying unknown risks, determines the reliability of anomaly judgment, ensures the accuracy of subsequent anomaly deviation identification, and avoids repeated detailed numerical simulations for each coupled operating condition through analytical correction, maintaining the overall efficiency of the assessment.

[0049] In an optional embodiment, the step of combining the anomalous pattern fingerprint with the key feature information to determine the causal path and novel damage mode of the current anomalous deviation in order to obtain unknown risk results includes: For the abnormal pattern fingerprint, spatiotemporally aligned association clues are extracted and aggregated from the key feature information. The association clues include abnormal pattern features, environmental conditions, local microenvironment features, and component attribute features. The preset basic action mechanism library is invoked to match the abnormal mode features, environmental conditions, local microenvironment features, and component attribute features respectively, and to obtain candidate causal paths that are causally related to the abnormal mode fingerprint. The optimal path is selected based on the credibility of the candidate causal paths, thereby identifying the corresponding novel damage mode. The correlation clues are used as input to the damage evolution correction model corresponding to the novel damage pattern to calculate the risk increment of the novel damage pattern relative to the coupled disaster impact result, and the risk increment is used as the unknown risk result.

[0050] Specifically, for the identified anomalous pattern fingerprints, four types of spatiotemporally aligned correlation clues are extracted and aggregated from key feature information. These include: anomalous pattern features (at least the duration, trend type, spatial distribution location, and fluctuation characteristics of the anomalous deviation); macro-environmental conditions (at least the duration, cumulative value, and extreme value of wind speed, rainfall, and salt spray concentration within a preset time period before and after the anomaly); local micro-environmental features (at least the amount of rainwater retained at specific locations on the structure, the volume of accumulated water, and the local salt ion concentration); and component attribute features (at least the component's material characteristics (e.g., stress corrosion cracking sensitivity coefficient) and geometric characteristics (e.g., the component's local geometric stress concentration coefficient, local periodic stress amplitude, and remaining effective thickness of the anti-corrosion coating). A basic action mechanism library contains fundamental physicochemical principles and damage evolution laws from materials science, corrosion electrochemistry, fracture mechanics, and structural mechanics. The four types of correlation clues are then matched with the principles in the basic action mechanism library using multi-dimensional features to select a set of candidate action mechanisms that are causally correlated with the anomalous pattern fingerprints.

[0051] Furthermore, based on the set of candidate action mechanisms and combining the temporal and spatial correspondences of four types of related clues, multiple candidate causal paths capable of fully explaining the process of abnormal deviation generation are synthesized in reverse: taking the final manifestation of the abnormal deviation as the starting point of reasoning, each possible intermediate evolutionary step is traced backward, and candidate action mechanisms are matched one-to-one with related clues to eliminate evolutionary steps with logical contradictions or lack of corresponding clue support, thus forming logically consistent candidate causal paths. For each candidate causal path, credibility is evaluated according to three dimensions: clue matching degree, temporal consistency, and physical rationality. A comprehensive credibility score is calculated based on preset weights, and the candidate causal path with the highest comprehensive credibility score is selected as the optimal causal path. The corresponding novel damage mode is then determined based on the damage evolution mechanism of the optimal causal path. Among them, the clue matching degree is the ratio of the number of clues corresponding to each link in the candidate causal path to the total number of related clues; the temporal consistency is the degree of agreement between the occurrence time of each link in the candidate causal path and the timestamp of the corresponding clue. If the timestamp difference is within the preset time interval, it is considered to be consistent; the physical rationality is the degree to which the candidate causal path conforms to the basic physical and chemical laws. It is understandable that physical and rational assessment are conventional techniques in the field and will not be explained further here.

[0052] Furthermore, the damage evolution correction model corresponding to the novel damage mode is invoked. The expected structural response baseline and the local microenvironmental characteristics and component attribute characteristics in the associated clues are substituted into the correction model to quantify the additional risk increment of the novel damage mode relative to the preset coupling effect. This additional risk increment is treated as an unknown risk outcome. The damage evolution correction model is a parameterized model constructed based on fracture mechanics and corrosion fatigue theory, and its output is the structural damage acceleration coefficient and additional risk increment caused by the novel damage mode.

[0053] In some examples, the damage evolution correction model is represented as: ; ; in, As an additional risk increment, The time step for risk assessment and baseline updates is aligned with the spatiotemporal alignment of the features. This represents the local damage acceleration factor. The stress corrosion cracking sensitivity coefficient of the component material. The stress concentration factor is the local geometric stress concentration factor of the component. This refers to the localized salt ion concentration at a specific location within the structure. This is the standard corrosion salt concentration threshold (material experimental calibration constant). This refers to the local periodic stress amplitude at critical locations of the component. This is the material's fatigue limit stress (a material experimental calibration constant). This is the nonlinear effect index of salt concentration (experimental calibration). is the nonlinear influence index of stress amplitude (experimental calibration), and k is the attenuation coefficient of anti-corrosion coating (experimental calibration). This refers to the remaining effective thickness of the anti-corrosion coating on the component surface.

[0054] In this embodiment, by transforming ambiguous abnormal signals into explicit damage patterns and their risk increments through clue convergence and mechanism reasoning, not only is the unknown risk advanced from "being detected" to "being explained and quantified," providing an incremental value comparable to traditional risk results, but the reasoning process is based on an extensible mechanism library, which not only ensures the causal rationality of the new damage identification but also avoids blindly increasing the computational scale, ensuring that the accuracy and efficiency of the assessment are not compromised when unknown risks are discovered.

[0055] S5. Calculate the comprehensive risk value of the power tower by combining the single-sided disaster impact results, the coupled disaster impact results, and the unknown risk results, and obtain the risk assessment results by combining the action paths between disasters.

[0056] In an optional embodiment, step S5 includes: The comprehensive risk value of the power tower group is calculated by weighting the results of the single-sided disaster impact, the coupled disaster impact, and the unknown risk. The comprehensive risk value is compared with the risk level threshold to determine the risk level, and a full-link risk assessment report is generated by combining the causal path corresponding to the unknown risk result with the coupled causal path corresponding to the coupling effect between single disasters.

[0057] In an optional embodiment, the comprehensive risk value of the power tower group is calculated by weighting the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result, including: The single-component-level comprehensive risk value is obtained by weighting the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result of a single power tower component. Based on the importance weight of the components of the power tower, the single-component-level comprehensive risk value of each component of the power tower is weighted and fused to obtain the power tower-level comprehensive risk value; Based on the spatial importance weight of the power towers, the comprehensive risk value of each power tower in the region is weighted and fused to obtain the comprehensive risk value of the power tower group.

[0058] Specifically, the unilateral disaster impact results, coupled disaster impact results, and unknown risk results of each key component of the power tower are weighted and fused to obtain the single-component-level comprehensive risk value of a single power tower. The calculation formula is expressed as follows: ; in, , , The weighting coefficients are for the results of unilateral disaster impact, coupled disaster impact, and unknown risk, respectively. These coefficients are determined by engineering case studies and expert scoring and can be dynamically adjusted according to different regions and different types of power towers. When there is no coupling and no unknown risks, It degenerates into a traditional single-disaster linear superposition risk; This is the quantization value for the p-th enhanced coupling path. The quantization value for the p-th weakening coupling path is given; the same coupling path can only belong to the enhancement class (included in the calculation). ) or weakening category (included) ), The result is positive if it enhances, negative if it weakens, and automatically cancels out if both occur, serving as the final weighted net risk of known coupling.

[0059] Furthermore, the comprehensive risk value of a single power tower at the single component level is weighted and fused based on the structural importance weight to obtain the comprehensive risk value at the power tower level; the comprehensive risk value at the power tower level is weighted and fused based on the spatial location importance weight of the power tower to obtain the comprehensive risk value of the power tower group area.

[0060] In some examples, structural importance weights are used to quantify the impact of the failure of a single critical component of the power tower on the overall structural safety of the tower. These weights are determined based on the severity of the consequences of the component failure, its contribution to the stress, and the probability of failure, as shown in Table 2. Table 2 Structural Importance Weighting Analysis Table

[0061] The spatial location importance weight of the power tower is used to quantify the spatial impact caused by the failure of the power tower. It is determined through multi-dimensional indicators. All indicators are directly obtained or calculated from spatial geographic information (GIS) data, as shown in Table 3. Table 3. Spatial Location Importance Weighting Analysis Table

[0062] In this embodiment, a hierarchical risk assessment strategy is adopted at the component level, tower level, and region level. First, at the component level, the combined impact of single-disaster risk, known coupling effects, and unknown local damage is quantified to accurately capture the enhancing / weakening effects between disasters. Then, structural importance weights are used to fuse these into tower-level risks, pinpointing weak points in individual towers. Finally, spatial location importance weights are used to aggregate these into region-level risks, reflecting the differences in impact between tower topology and geographical space. This approach retains the accuracy of component-level assessments while achieving efficient calculations through hierarchical weighting, significantly improving the disaster resilience of the power grid and the efficiency of operation and maintenance decisions.

[0063] Based on the same inventive concept, this application also provides a multi-hazard coupled risk assessment system for power tower groups, corresponding to a method for assessing the coupled risk of multiple hazards in power tower groups, such as... Figure 2 As shown, the system includes: The feature acquisition module is used to obtain key feature information for risk assessment based on the spatiotemporal correlation matching items of multi-source heterogeneous data streams; The disaster identification module is used to determine the degree of impact of each individual disaster on the power tower based on the key feature information, and obtain the unilateral disaster impact results; identify the interaction between individual disasters and obtain the coupled disaster impact results; and identify unknown risk results based on the expected structural response baseline of the key components of the power tower. The risk assessment module is used to calculate the comprehensive risk value of the power tower by combining the single-sided disaster impact results, the coupled disaster impact results, and the unknown risk results, and to obtain the risk assessment results by combining the action paths between disasters.

[0064] In this embodiment, the various modules work together to achieve functions such as spatiotemporal alignment of multi-source data and inversion of local micro-environment, identification of single disaster and coupled impact, detection of unknown risks and generation of comprehensive risk reports. Thus, in actual engineering scenarios, the problem of balancing the characterization and assessment efficiency of nonlinear interactions of multiple disasters is solved in a systematic way, realizing integrated automatic assessment from data acquisition to end-to-end risk output.

[0065] The above-described embodiments are preferred embodiments of this application and are not intended to limit the specific scope of this application. The scope of this application includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with the shape, structure, and method of this application are within the protection scope of this application.

Claims

1. A method for risk assessment of multi-hazard coupling of an electric tower group, characterized in that: Includes the following steps: Key feature information for risk assessment is obtained based on the spatiotemporal correlation matching items of multi-source heterogeneous data streams; Based on the key feature information, the degree of impact of each single disaster on the power tower is determined, and the single-sided disaster impact results are obtained; Identify the interactions between individual disasters and obtain the results of coupled disaster impacts; Identifying unknown risk results based on the expected structural response baseline of key components of power towers involves: establishing the expected structural response baseline of key components under corresponding environmental conditions; calculating the deviation between the actual structural response value of each key component and the expected structural response baseline in real time, continuously tracking the temporal changes of the deviation to identify persistent abnormal deviations below a threshold; forming an abnormal pattern fingerprint based on the spatiotemporal characteristics of the persistent abnormal deviations, wherein the spatiotemporal characteristics include at least the duration of the abnormality, the trend of change, and the spatial distribution location; and combining the abnormal pattern fingerprint with the key feature information to determine the causal path and novel damage mode of the current abnormal deviation in order to obtain unknown risk results. The comprehensive risk value of the power tower is calculated by combining the results of the single-sided disaster impact, the results of the coupled disaster impact, and the results of the unknown risk, and the risk assessment results are obtained by combining the action paths between disasters. 2.The method of claim 1, wherein: The step of obtaining key feature information for risk assessment based on the spatiotemporal correlation matching terms of multi-source heterogeneous data streams includes: Acquire multi-source heterogeneous data streams, including meteorological data, environmental monitoring data, equipment monitoring data, and geometric attribute data of the power tower's location; Time registration is performed on multi-source heterogeneous data streams, centered on the spatial location of each power tower. Based on time-registered multi-source heterogeneous data streams, basic feature information is extracted. Combined with the geometric feature information of the power tower, local micro-environment features are obtained based on the physical principles of mass transfer and fluid mechanics. The basic feature information and the local micro-environment features are used as key feature information.

3. The method of claim 1, wherein: The process of determining the impact of each individual disaster on the power tower based on the key feature information and obtaining the unilateral disaster impact results includes: Based on wind speed characteristics, wind direction characteristics and power tower structural parameters, the wind pressure on the tower body and conductor is calculated to obtain the typhoon impact results; Based on the information of salt spray concentration and temperature and humidity, combined with the material properties of the power tower, the corrosion rate and cumulative damage were calculated to obtain the results of the impact of salt spray corrosion. The settlement trend of the tower foundation is assessed based on geological and meteorological characteristic data, and the impact of geological settlement is obtained. Among them, the typhoon impact results, salt spray corrosion impact results, and geological subsidence impact results form a unilateral disaster impact result.

4. The method of claim 3, wherein the method further comprises: Identify the interactions between individual disasters and obtain the results of coupled disaster impacts, including: Based on a pre-defined causal association rule base, the coupling effect between single disasters is identified by combining the key feature information with the single-sided disaster impact results. The coupling effect includes enhancement effect and weakening effect. Based on the enhancement effect / weakening effect, the corresponding coupling effect quantification value is calculated using a nonlinear influence function, thereby obtaining the coupling disaster impact result.

5. The method of claim 1, wherein: The establishment of the expected structural response baseline for key components of the power tower under corresponding environmental conditions includes: Based on the key feature information, a list of key components of the power tower is extracted, and a corresponding response index is matched for each component. The calculation process parameters based on the single-sided disaster impact result determine the single-sided response benchmark value of the response index under the corresponding single disaster, wherein the single-sided response benchmark value of each component and each response index under the individual action of each single disaster corresponds one-to-one with the single-sided disaster impact result; Obtain the linear sum of the single-sided response baseline values ​​of all single disasters involved in the coupling, determine the corresponding coupling correction coefficient according to the coupling mode, and correct the linear sum to obtain the expected nominal response value under the multi-disaster coupling condition. Calculate the standard deviation of the fluctuation of the nominal value of the expected response, and determine the initial response baseline range based on the nominal value of the expected response; The initial response baseline range is calibrated based on historical health data from periods of normal operation of the power tower to obtain the expected structural response baseline, which includes the boundaries of theoretical stress accumulation rate, vibration frequency, strain amplitude, fatigue damage increment, and tower foundation settlement rate.

6. The method of claim 1, wherein: The step of combining the abnormal pattern fingerprint with the key feature information to determine the causal path and novel damage mode of the current abnormal deviation in order to obtain unknown risk results includes: For the abnormal pattern fingerprint, spatiotemporally aligned association clues are extracted and aggregated from the key feature information. The association clues include abnormal pattern features, environmental conditions, local microenvironment features, and component attribute features. The preset basic action mechanism library is invoked to match the abnormal pattern features, environmental conditions, local microenvironment features, and component attribute features respectively, and to obtain candidate causal paths that are causally related to the abnormal pattern fingerprint. The optimal path is selected based on the credibility of the candidate causal paths, thereby identifying the corresponding novel damage mode. The correlation clues are used as input to the damage evolution correction model corresponding to the novel damage pattern to calculate the risk increment of the novel damage pattern relative to the coupled disaster impact result, and the risk increment is used as the unknown risk result.

7. The method of claim 1, wherein: The calculation of the comprehensive risk value of the power tower by combining the unilateral disaster impact results, the coupled disaster impact results, and the unknown risk results, and obtaining the risk assessment results by combining the action paths between disasters, includes: The comprehensive risk value of the power tower group is calculated by weighting the results of the single-sided disaster impact, the coupled disaster impact, and the unknown risk. The comprehensive risk value is compared with the risk level threshold to determine the risk level, and a full-link risk assessment report is generated by combining the causal path corresponding to the unknown risk result with the coupled causal path corresponding to the coupling effect between single disasters.

8. The method of claim 7, wherein the method further comprises: The weighted calculation of the comprehensive risk value of the power tower group by integrating the unilateral disaster impact results, the coupled disaster impact results, and the unknown risk results includes: The single-component-level comprehensive risk value is obtained by weighting the unilateral disaster impact result, the coupled disaster impact result, and the unknown risk result of a single power tower component. Based on the importance weight of the components of the power tower, the single-component-level comprehensive risk value of each component of the power tower is weighted and fused to obtain the power tower-level comprehensive risk value; Based on the spatial importance weight of the power towers, the comprehensive risk value of each power tower in the region is weighted and fused to obtain the comprehensive risk value of the power tower group.

9. An electric tower group multi-disaster coupling risk assessment system, characterized in that: The method for assessing the coupled risks of multiple disasters in a power tower group as described in any one of claims 1-8 includes: The feature acquisition module is used to obtain key feature information for risk assessment based on the spatiotemporal correlation matching items of multi-source heterogeneous data streams; The disaster identification module is used to determine the impact of each individual disaster on the power tower based on the key feature information, and obtain the unilateral disaster impact results; identify the interaction between individual disasters and obtain the coupled disaster impact results; identify unknown risk results based on the expected structural response baseline of the key components of the power tower, wherein the expected structural response baseline of the key components of the power tower under the corresponding environmental conditions is established; the deviation between the actual structural response value of each key component and the expected structural response baseline is calculated in real time, and the temporal change of the deviation is continuously tracked to identify continuous abnormal deviations below a threshold; an abnormal pattern fingerprint is formed based on the spatiotemporal characteristics of the continuous abnormal deviation, wherein the spatiotemporal characteristics include at least the duration of the abnormality, the trend of change, and the spatial distribution location; and the causal path and novel damage mode of the current abnormal deviation are determined by combining the abnormal pattern fingerprint and the key feature information to obtain unknown risk results. The risk assessment module is used to calculate the comprehensive risk value of the power tower by combining the single-sided disaster impact results, the coupled disaster impact results, and the unknown risk results, and to obtain the risk assessment results by combining the action paths between disasters.

Images