A method for dynamic assessment of the vulnerability of power grid facilities to flood disasters

Abstract

This invention belongs to the field of power grid flood disaster assessment technology. Specifically, it discloses a dynamic assessment method for the flood vulnerability of power grid facilities. This method acquires a historical flood event dataset of power grid facilities in a target area, extracts homogeneous flood sequence groups to calculate the failure rate growth rate, and comprehensively calculates and dynamically updates a multi-factor statistical disaster bearing vulnerability score. This enables quantitative analysis of the cumulative damage to power grid facilities in multiple flood events, accurately captures the cumulative degradation trend of power grid facility performance, and effectively reveals the vulnerability change patterns of facilities in continuous disasters. It provides a reliable basis for scientifically assessing the risk status of power grid facilities under long-term flood threats. Furthermore, by setting an elevation protection coefficient based on the facility's ground elevation and combining it with flood intensity influence factors to correct the original fault degree, this two-factor correction method can meticulously and accurately assess the flood vulnerability of facilities at different locations.

Description

Technical Field

[0001] This invention belongs to the field of power grid flood disaster assessment technology, and specifically relates to a dynamic assessment method for the vulnerability of power grid facilities to flood disasters. Background Technology

[0002] With rapid urbanization and increased density of power grid equipment, floods often result in widespread power outages and severe equipment damage, significantly impacting urban operations. To ensure timely early warning systems for power grid facilities in the face of floods, it is necessary to assess the extent of damage caused by such disasters.

[0003] Existing technologies, such as the flood forecasting method and system for power grid facilities disclosed in Chinese invention patent application No. 2023105426264, solve the problem of low forecast accuracy by objectively processing raw data using K-means clustering analysis and constructing a deep learning model using backpropagation algorithm, thereby achieving more accurate flood forecasting for power grid facilities and reducing the risk of equipment damage.

[0004] Existing technologies, such as the storm risk assessment method for power grid facilities disclosed in Chinese invention patent application No. 2016103179949, collect disaster, storm, socio-economic, geographical, and power grid data, and efficiently calculate four major elements based on these data: disaster-causing factors, disaster-prone environment, vulnerability of disaster-bearing bodies, and disaster prevention capabilities. Finally, risk rating and classification are carried out, which solves the problems of one-sided, inefficient, and abstract results in assessment, and achieves efficient, accurate, and comprehensive storm risk assessment. It also provides intuitive risk index and level output to facilitate decision-making by staff.

[0005] Regarding existing technical solutions, it is clear that current flood disaster assessments of power grid facilities mainly focus on real-time risk prediction and static structural vulnerability, which is a timely analysis, i.e., handling each flood event independently. It also has the following shortcomings: 1. It does not consider the quantitative analysis of the cumulative damage to facilities caused by multiple flood events, lacks the mining of the historical failure growth pattern of equipment, and fails to capture the cumulative degradation trend of power grid facility performance, which leads to the subsequent risk assessment underestimating the actual risk and making it difficult to reveal the vulnerability change pattern of facilities in continuous disasters.

[0006] 2. The correlation between flood severity and failure rate was ignored. The differences in failure rate growth for the same facility under different flood intensities were not analyzed. Furthermore, the reinforcing effect of consecutive disasters and the impact of disaster intervals on equipment recovery were not considered, which weakened the effectiveness of the early warning system.

[0007] 3. The evaluation process treated the equipment's elevation above the ground as a fixed parameter, failing to reflect the performance degradation differences of equipment at different elevations in the same flood, and failing to quantify the gradient of the impact of different submersion depths on the equipment failure rate. Summary of the Invention

[0008] In view of this, in order to solve the above problems, a dynamic assessment method for the vulnerability of power grid facilities to flood disasters is proposed.

[0009] The objective of this invention can be achieved through the following technical solution: This invention provides a dynamic assessment method for the vulnerability of power grid facilities to flood disasters. The method includes: S1, obtaining the ground elevation of each power grid facility in the target area and the historical flood event dataset of the target area. Each record includes facility identification, trigger time, single flood event characteristic data, and post-flood quantitative facility failure degree value and failure type.

[0010] S2. Based on the characteristic data of flood events, set the flood intensity influence factor and set the elevation protection coefficient based on the facility's ground elevation. After correcting the original fault degree with both, output the balanced fault degree.

[0011] S3. Extract continuous flood events belonging to the same preset flood intensity range from the flood intensity influencing factors, form a homogeneous flood sequence group, and calculate the failure rate of power grid facilities in the group based on the balanced fault degree setting.

[0012] S4. The disaster vulnerability score is obtained by statistically analyzing the overall failure rate and the number of groups whose failure rate exceeds the preset warning value, and feedback is provided.

[0013] S5. When a new flood event occurs, repeat steps S1-S4 to update the comprehensive disaster carrying capacity vulnerability score.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention obtains a dataset of historical flood events of power grid facilities in the target area, extracts homogeneous flood sequence groups to calculate the failure growth rate, and comprehensively calculates the disaster bearing vulnerability score based on multiple factors and updates it dynamically. It realizes the quantitative analysis of the cumulative damage of power grid facilities in multiple flood events, deeply explores the historical failure growth pattern of equipment, accurately captures the cumulative degradation trend of power grid facility performance, and thus avoids the situation of underestimating the actual risk in subsequent risk assessments. It effectively reveals the vulnerability change pattern of facilities in continuous disasters and provides a reliable basis for scientifically assessing the risk status of power grid facilities under long-term flood threats.

[0015] (2) This invention sets a flood intensity influencing factor based on flood event characteristic data and calculates the failure rate growth rate of power grid facilities within different flood intensity ranges. It fully considers the correlation between flood severity and failure rate growth, analyzing not only the differences in failure rate growth for the same facility under different flood intensities, but also reflecting the reinforcing effect of consecutive disasters and the impact of disaster intervals on equipment recovery. This significantly enhances the effectiveness of early warning and facilitates subsequent prediction of equipment failure risk changes based on different flood intensities, allowing for early preventative measures.

[0016] (3) This invention sets an elevation protection coefficient based on the facility's ground elevation and corrects the original failure degree by combining the flood intensity influence factor. By using a two-factor correction method, it quantifies the gradient of the influence of different flood depths on the equipment failure rate, which can accurately reflect the performance degradation differences of equipment at different elevations in the same flood, thereby accurately assessing the flood vulnerability of facilities at different locations. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the overall implementation process of the method of the present invention.

[0019] Figure 2 This is a schematic diagram of the process for setting the flood intensity influencing factor in this invention.

[0020] Figure 3 This is a schematic diagram of the process for setting the elevation protection coefficient in this invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 As shown, the present invention provides a dynamic assessment method for the vulnerability of power grid facilities to flood disasters. The method includes: S1, acquiring the ground elevation of each power grid facility in the target area and the historical flood event dataset of the target area. Each record includes facility identification, trigger time, single flood event characteristic data, and post-flood quantitative facility failure degree value and failure type.

[0023] S2. Based on the characteristic data of flood events, set the flood intensity influence factor and set the elevation protection coefficient based on the facility's ground elevation. After correcting the original fault degree with both, output the balanced fault degree.

[0024] S3. Extract continuous flood events belonging to the same preset flood intensity range from the flood intensity influencing factors, form a homogeneous flood sequence group, and calculate the failure rate of power grid facilities in the group based on the balanced fault degree setting.

[0025] S4. The disaster vulnerability score is obtained by statistically analyzing the overall failure rate and the number of groups whose failure rate exceeds the preset warning value, and feedback is provided.

[0026] S5. When a new flood event occurs, repeat steps S1-S4 to update the comprehensive disaster carrying capacity vulnerability score.

[0027] This invention acquires a dataset of historical flood events affecting power grid facilities in a target area, extracts homogeneous flood sequence groups to calculate the failure rate, and comprehensively analyzes and dynamically updates a multi-factor statistical disaster vulnerability score. This enables quantitative analysis of the cumulative damage to power grid facilities during multiple flood events, deeply mines historical failure growth patterns of equipment, accurately captures the cumulative degradation trend of power grid facility performance, and thus avoids underestimating actual risks in subsequent risk assessments. It effectively reveals the vulnerability patterns of facilities in continuous disasters, providing a reliable basis for scientifically assessing the risk status of power grid facilities under long-term flood threats.

[0028] It should be added that the characteristic data of a single flood event includes flow velocity sequence, inundation depth sequence, and total flood duration.

[0029] Please see Figure 2 As shown, the specific setting process for the flood intensity influencing factor mentioned in step S2 is as follows: A1. Extract the maximum value of the flow velocity sequence of a single flood as the target flood flow velocity, and at the same time calculate the standard deviation of the flow velocity sequence to obtain the flow velocity fluctuation.

[0030] A2. Extract the maximum value of the inundation depth sequence of a single flood as the target flood inundation depth, and obtain the inundation depth growth rate through differential calculation.

[0031] A3. If the rate of increase of flow velocity fluctuation or inundation depth exceeds the corresponding preset threshold, the flow velocity fluctuation and inundation depth growth rate are normalized and then imported into the Sigmoid function to output the flood intensity assessment compensation factor.

[0032] A4. Obtain the total duration of a single flood event as the target duration of that flood event.

[0033] A5. Weighting coefficients for target flood velocity, target flood inundation depth, and target duration assigned using the analytic hierarchy process (AHP).

[0034] A5. After normalizing the target flood flow velocity, target flood inundation depth and target duration, the basic flood intensity influence factor is obtained by weighted summation. After compensation by the flood intensity assessment compensation factor, the final flood intensity influence factor is output.

[0035] Understandably, the normalization adopts linear normalization, i.e., Min-Max normalization. Taking the normalization of flow velocity fluctuation as an example, the minimum and maximum values ​​of flow velocity fluctuation are found from historical flood data. The calculated flow velocity fluctuation is then imported into the Min-Max normalization setting formula to obtain the normalized flow velocity fluctuation. When the flow velocity fluctuation is lower than the historical minimum value, the normalization result is directly assigned to 0.

[0036] Understandably, the preset thresholds for flow velocity fluctuation and inundation depth growth rate are determined by comprehensively considering historical flood data and the extent of damage to power grid facilities. This involves collecting data on flow velocity fluctuation and inundation depth growth rate from a large number of historical flood events, performing statistical analysis on this data, and calculating statistical measures such as mean, standard deviation, and quantiles. For example, by calculating the 95th quantile, data exceeding this value are considered to have a high probability of being abnormal. Simultaneously, by referencing historical flood data that caused varying degrees of failure in power grid facilities, the boundaries for flow velocity fluctuation and inundation depth growth rate that would trigger significant disaster impacts are determined. Based on this, experts from the power industry and meteorology can be invited to evaluate and adjust the statistical analysis results based on their professional experience, ultimately determining a reasonable preset threshold.

[0037] It should be noted that in flood intensity assessment, when the fluctuation of flow velocity or the rate of increase in inundation depth exceeds a preset threshold, it signifies an abnormal change in flood characteristics that could have a more complex and severe impact on power grid facilities. The Sigmoid function, due to its unique S-shaped curve characteristics, can map the input normalized data to a range of 0 to 1, achieving smooth processing and nonlinear transformation of abnormal data. Furthermore, its nonlinear characteristics can better simulate the complexity of flood disaster impacts, avoiding the over-amplification or neglect of abnormal data by simple linear calculations. This makes the flood intensity assessment compensation factor more closely reflect the actual disaster impact, thereby improving the accuracy and reliability of flood intensity assessment and providing a more scientific basis for assessing the vulnerability of power grid facilities.

[0038] It is important to note that when both the flow velocity fluctuation and the inundation depth growth rate exceed the corresponding preset thresholds, the weighted sum of their normalized results should be used as input. The weights of the inundation depth growth rate for flow velocity fluctuation can be set using the analytic hierarchy process (AHP) and machine learning algorithms, such as random forests and support vector machines. These weights can be set by training the model using historical flood data and data on the damage to power grid facilities, and by evaluating the importance of model features. The AHP and machine learning algorithms are common weighting methods and will not be elaborated further.

[0039] This invention, through embodiments thereof, sets a flood intensity influencing factor based on flood event characteristic data and calculates the failure rate growth rate of power grid facilities within different flood intensity ranges. It fully considers the correlation between flood severity and failure rate growth, analyzing not only the differences in failure rate growth for the same facility under different flood intensities but also reflecting the amplifying effect of successive disasters and the impact of disaster intervals on equipment recovery. This significantly enhances the effectiveness of early warning and facilitates subsequent prediction of changes in equipment failure risk based on different flood intensities, enabling proactive preventative measures.

[0040] Please see Figure 3 As shown, the specific setting process of the elevation protection coefficient in step S2 includes: B1, matching the interference inundation depth of each power grid facility from the preset ground elevation and interference inundation depth mapping table based on the ground elevation.

[0041] B2. Assessment of the effectiveness of flood protection based on the depth of flood inundation.

[0042] B3. Calculate the ratio of the number of historical floods that were deemed to be effective protection for each power grid to the total number of historical floods to obtain the effective flood protection ratio.

[0043] B4. Calculate the average flood inundation depth of each historical flood that was judged to be effective in protection, and obtain the inundation depth difference by subtracting it from the disturbance inundation depth. At the same time, calculate the average of the total duration of each historical flood that was judged to be effective in protection to obtain the average effective flood protection duration.

[0044] B5. After calculating the normalized values ​​of the inundation depth difference and the average effective flood protection duration respectively, the protection correction factor is obtained by weighted summation based on the preset weights of the inundation depth difference and the average effective flood protection duration.

[0045] B6. After correcting the effective ratio of flood protection by the protection correction factor, the elevation protection coefficient is output.

[0046] It should be added that the normalized values ​​of the inundation depth difference and the average effective flood protection duration are calculated using the same normalization method as the flow velocity fluctuation and the inundation depth growth rate, so the example will not be repeated.

[0047] Understandably, the preset mapping table between ground elevation and disturbance inundation depth is set through historical data statistical analysis and machine learning algorithm training. Specifically, by collecting ground elevation data of a large number of power grid facilities within the target area, combined with inundation depth and facility damage information from historical flood events, the ground elevation is divided into multiple intervals, such as every 5 meters. For each elevation interval, the probability, type, and severity of facility failures at different inundation depths in history are statistically analyzed, and the average disturbance level is calculated. Then, machine learning algorithms such as regression analysis and decision trees are used, with ground elevation as the independent variable and disturbance inundation depth (i.e., the inundation depth corresponding to the quantified disturbance level) as the dependent variable, to train the model, optimize the parameters, and generate the mapping relationship.

[0048] It is important to note that all data in the table has passed cross-validation, and the parameters and mapping relationships in the mapping table need to be updated periodically based on newly occurring flood data and facility damage feedback to ensure that the data in the table dynamically reflects the actual impact of floods. Furthermore, the improved analysis and training methods and related algorithms described above are existing processing techniques and will not be described in detail here.

[0049] Understandably, for each power grid facility, all historical flood events with effective protection are first screened out, and the average flood inundation depth of these events is calculated. This average is then subtracted from the disturbance inundation depth obtained in step 1 to obtain the inundation depth difference. This difference reflects the deviation between the actual inundation depth and the disturbance threshold when the facility can effectively protect itself. The larger the difference, the stronger the protection capability, indicating that the facility can maintain effective protection even when facing deeper floods. Simultaneously, the average duration of the total duration of these effectively protected flood events is calculated to obtain the average effective flood protection duration. This duration reflects the facility's ability to withstand flood impacts under effective protection; the longer the duration, the better the facility's protection stability. Therefore, protection correction factors are set based on two factors: inundation depth difference and effective flood protection duration, to compensate for the current shortcomings of focusing on the frequency of events while neglecting the degree of protection.

[0050] In one specific embodiment, the protection correction factor is comprehensively adjusted based on the difference in inundation depth and the average effective flood protection duration to further correct the flood protection effectiveness ratio. Specifically, the correction method involves multiplying the sum of 1 and the protection correction factor by the flood protection effectiveness ratio to obtain the elevation protection coefficient. This coefficient comprehensively considers the elevation of the power grid facilities above ground, historical flood protection performance, and flood characteristics under effective protection conditions. It can accurately quantify the protection capability of the facilities in the face of floods, providing a crucial basis for flood risk assessment of power grid facilities.

[0051] Understandably, the preset weights for inundation depth difference and average effective flood protection duration need to take into account both subjective experience and objective data. They can be directly assigned by expert evaluation, or the weights can be calculated based on the correlation between inundation depth difference, average effective protection duration and facility failure in historical flood events. The final value of the preset weights is the average of the weights assigned by expert evaluation and the weights calculated.

[0052] Furthermore, the specific judgment process for determining the effectiveness of flood interference protection described in step B2 includes: B21, traversing the inundation depth data of each historical flood event, comparing it one by one with the interference inundation depth corresponding to each power grid facility, and extracting the total duration of each flood and comparing it with the preset flood interference duration threshold.

[0053] B22. For each flood event, the power grid facility is deemed to be effective in flood protection if the following conditions are met simultaneously: the inundation depth is less than or equal to the disturbance inundation depth of the facility.

[0054] The total duration of the flooding exceeded the preset flood disturbance duration.

[0055] No malfunction caused by flooding was triggered.

[0056] If any of the above conditions are not met, the power grid facility is deemed ineffective in its flood protection efforts.

[0057] Understandably, the inundation depth directly determines the degree of contact between power grid facilities and floodwaters. Different facilities, such as transformers and line towers, have different tolerances to water depth. By comparing the actual inundation depth with the preset disturbance inundation depth, it can be determined whether the flood has exceeded the basic protection threshold of the facilities. The duration of flood immersion is positively correlated with the aging of facility materials and the decline in insulation performance.

[0058] Regarding step S2, it should be added that the degree of balancing failure can be obtained by multiplying the original degree of failure by the flood intensity influence factor and then multiplying the result by 1 and the difference between the elevation protection coefficient.

[0059] Understandably, this formula accurately depicts the antagonistic relationship between disaster and facility protection by using the logic that flood intensity positively amplifies the severity of a fault and the protection coefficient negatively weakens its impact. Multiplying the flood intensity factor by the original fault severity aligns with the physical principle that stronger disasters lead to more severe facility damage. The difference between 1 and the protection coefficient quantifies the risk discount brought by facility elevation; the stronger the protection capability, the more significant the attenuation of the fault severity. Its calculation logic is consistent with the classic model in the power industry where risk is the product of disaster intensity and vulnerability. In extreme scenarios, such as when the protection coefficient is 1, an output of 0 can reasonably reflect the effectiveness of flood protection.

[0060] This invention, through setting an elevation protection coefficient based on the facility's elevation above ground and combining it with flood intensity influence factors to correct the original failure level, uses a two-factor correction method to quantify the gradient of the impact of different submersion depths on equipment failure rates. This can accurately reflect the performance degradation differences of equipment at different elevations in the same flood, thereby providing a detailed and accurate assessment of the flood vulnerability of facilities at different locations.

[0061] The extraction of homogeneous flood sequence groups in step S3 must meet the following conditions: the standard deviation of the flood intensity influence factor within the group is less than or equal to the preset fluctuation threshold.

[0062] Events within the group are consecutive in time and the interval between them is less than or equal to the preset maximum allowed interval.

[0063] The specific calculation process for the failure rate of the power grid facilities mentioned in step S3 includes: C1, taking the difference in the balance of fault degree between the first and last flood history within the group, and dividing it by the average triggering time interval of adjacent historical floods to obtain the failure rate of each power grid facility.

[0064] C2. Fit the balance failure rate change curve of each power grid setting with time as the horizontal axis and balance failure rate as the vertical axis to determine whether failure rate compensation is needed.

[0065] C3. If it is determined that failure rate compensation is required, the ratio of the total length of the curve above the average balanced failure level to the total length of the changing curve is extracted from the curve as the failure rate compensation factor.

[0066] C4. The failure rate of each power grid facility is obtained by compensating the failure rate with a failure rate compensation factor.

[0067] C5. The failure rate of each power grid facility is calculated by averaging the failure rate rates of each facility to obtain the final failure rate rate of the power grid facilities.

[0068] It should be added that plotting the curve of the change in the degree of balanced failure and determining whether compensation is needed aims to identify nonlinear trends. When the growth of failures accelerates, decelerates, or fluctuates, such as when equipment aging accelerates or temporary maintenance reduces failures, simple linear calculations will produce deviations. In this case, it is necessary to explore the potential patterns of change through curve characteristics to avoid underestimating or overestimating the risk of failure.

[0069] It should also be noted that the proportion of the curve length above the average equilibrium failure level is used as a compensation factor because the failure growth segment above the average level better reflects the accelerating trend or abnormal deterioration of equipment failure. The larger this proportion, the more severe the failure growth, requiring amplified compensation of the base growth rate. Conversely, the compensation is reduced to precisely adjust the growth rate to match the actual failure characteristics.

[0070] Understandably, using compensation factors to correct the base growth rate can eliminate the interference of nonlinear factors, making the growth rate closer to the actual failure process of facilities. Furthermore, averaging the growth rates of each facility to obtain the final result can balance individual differences and reflect the average failure rate of facilities within the group from an overall perspective, providing a unified risk assessment benchmark for power grid operation and maintenance.

[0071] It should be added that the failure growth assessment compensation factor and the failure growth anomaly coefficient of power grid facilities calculated above are used to correct the failure growth rate of power grid facilities. Specifically, the two are weighted and calculated with the original failure growth rate so that the corrected failure growth rate can more accurately reflect the failure trend of equipment under actual disaster environments.

[0072] Understandably, the weighting of the power grid facility failure rate rate and failure growth anomaly coefficient can be comprehensively set by combining data-driven approaches and expert experience. Specifically, by analyzing a large amount of historical flooding events and power grid facility failure data, the correlation between the failure rate rate and failure growth anomaly coefficient and actual failure situations is statistically analyzed, with indicators showing higher correlation being assigned higher weights. Regarding expert experience, experts in power engineering, disaster prevention and mitigation, and other fields are invited to score and evaluate the importance of the two indicators in reflecting facility failure risk, combining their professional knowledge and practical experience to form a comprehensive weighting.

[0073] Furthermore, the determination of whether failure rate compensation is needed in step C2 includes: extracting the total number of peaks and valleys from the balanced failure degree change curve, and using the ratio of the total number of peaks and valleys to the total length of the curve as the frequency of change.

[0074] If the frequency of change exceeds a preset threshold, it is determined that failure rate growth compensation is required; otherwise, it is determined that failure rate growth compensation is not required.

[0075] Understandably, the peaks and troughs in the balanced failure severity curve represent significant increases and decreases in failure severity, reflecting abrupt changes during equipment failure. More peaks and troughs indicate a more complex failure mode due to flooding, such as re-damage after emergency repairs or nonlinear responses under varying flood intensities. Using the ratio of the total number of peaks and troughs to the total curve length to measure the frequency of change transforms the curve's fluctuation characteristics into standardized values, avoiding subjective judgment. This indicator considers both the number of fluctuations and the time span (curve length), comprehensively reflecting the dynamics of failure changes. A higher ratio indicates more drastic fluctuations in failure severity per unit time, and a lower reliability of the baseline growth rate.

[0076] In one specific embodiment, the threshold for frequency of change can be set by comprehensively considering historical data, equipment characteristics, and engineering requirements. Typically, it is achieved by statistically analyzing the peak and trough distribution of historical curves for similar equipment and selecting the median or 90th percentile as the threshold. Alternatively, it can be based on a physical model of equipment failure to simulate the growth rate error under different fluctuation scenarios, and the frequency corresponding to an error exceeding 10% can be set as the threshold.

[0077] Regarding step S4, the specific statistical process for the disaster bearing vulnerability score includes: D1, calculating the failure growth assessment compensation factor based on the triggering time and flood intensity influence factors of each historical flood within the same group.

[0078] D2. Based on the balance failure degree and failure type of different power grid facilities in the same group during each historical flood, calculate the abnormal coefficient of power grid facility failure growth.

[0079] D3. Correct the failure rate of the power grid facilities based on the failure growth assessment compensation factor and the failure growth anomaly coefficient of the power grid facilities.

[0080] D4. Calculate the difference in average flood intensity influence factors and the difference in the corrected failure growth rate for different groups, and match the corresponding reference failure growth rate difference based on the difference in flood intensity influence factors.

[0081] D5. The ratio of the number of adjacent groups whose statistically corrected failure growth rate difference exceeds the reference failure growth rate difference to the total number of sequence groups is called the failure growth rate difference over-limit ratio.

[0082] D6. The ratio of the number of groups whose failure growth rate exceeds the preset warning value to the total number of sequence groups is recorded as the failure growth rate over-limit ratio.

[0083] D7. Based on the weights corresponding to the failure growth rate difference exceeding the limit ratio and the failure growth rate exceeding the limit ratio, the growth rate difference exceeding the limit ratio and the effective growth rate exceeding the limit ratio are weighted and summed to obtain a comprehensive evaluation coefficient. The comprehensive evaluation coefficient is multiplied by the preset disaster bearing vulnerability score to obtain the disaster bearing vulnerability score.

[0084] In one specific embodiment, the preset total score for disaster vulnerability can be either out of 10 or out of 100, and is not limited here.

[0085] It should be added that the weights corresponding to the failure rate difference exceeding the limit ratio and the failure rate growth rate exceeding the limit ratio are usually determined by combining professional experience and historical data verification. That is, experts in the fields of power engineering and disaster prevention and mitigation directly assign weights based on their professional judgment on the impact of the failure rate difference and failure rate on vulnerability. They also use the real results of vulnerability scores from historical flood events to backfit the calculation so that the score best reflects reality.

[0086] It should be added that the difference in the average flood intensity influence factor reflects the difference in disaster intensity among different homogeneous flood sequence groups, and the difference in the corrected failure growth rate reflects the difference in the failure growth rate of equipment in different groups. The correction of the failure growth rate of the power grid facilities can be obtained by multiplying the compensation factor and the anomaly coefficient by the corresponding allocation weights determined based on expert experience and historical data, and then multiplying the sum of the results with 1 and the failure growth rate of the power grid facilities.

[0087] Understandably, based on the difference in the average flood intensity influence factor, a corresponding reference failure rate difference is matched using a pre-established lookup table or mathematical mapping relationship. This lookup table or mapping relationship can be trained using a large amount of historical data and is used to measure the normal range of the equipment failure rate rate difference under the same intensity difference.

[0088] Furthermore, the specific calculation of the failure growth assessment compensation factor mentioned in step D1 is as follows: D11, calculate the ratio of consecutive floods to total floods within the same group.

[0089] D12. Calculate the mean value of the flood intensity influence factors corresponding to each historical flood in the group to obtain the average flood intensity influence factor.

[0090] D13. Calculate the sequence of intervals between adjacent events based on the trigger time;

[0091] D14. The flood trigger density is obtained by comparing the number of continuous floods with a trigger interval shorter than the preset interval with the total number of historical floods.

[0092] D15. Calculate the average trigger interval of historical flood events within the group and compare it with the preset reference interval. If the average interval is greater than the reference interval, set the continuity to 0. Otherwise, calculate the ratio of the absolute value of the difference between the two to the reference interval, and use it as the flood trigger continuity.

[0093] D16. The failure assessment compensation factor is obtained by weighted summation of the ratio of consecutive floods, the average merit intensity influence factor, the flood triggering density, and the flood triggering continuity.

[0094] Understandably, consecutive flood events lead to a lack of maintenance and recovery time for equipment, resulting in exponentially accumulating damage. This indicator quantifies the accelerating effect of the continuity of the disaster sequence on failure. The intensity of a single flood determines the immediate damage level of the equipment, while the average intensity reflects the overall damage level of the sequence group, avoiding assessment bias caused by individual high-intensity or low-intensity events. Flood events within short intervals can exceed the fatigue limit of equipment. This indicator quantifies the amplifying effect of the density of disaster temporal distribution on failure. Comparing the average trigger interval with a preset reference value can identify whether the flood has formed a long-term continuous threat scenario where the density of continuous pressure replenishment is not covered.

[0095] It should be noted that the weighting of the consecutive flood frequency ratio, average flood intensity influence factor, flood triggering density, and flood triggering continuity should be based on the comprehensive impact of each indicator on equipment failure. The consecutive flood frequency ratio reflects the cumulative effect of the disaster; multiple consecutive floods easily exacerbate equipment fatigue damage and can be assigned a higher weight, such as 0.3. The average flood intensity influence factor reflects the overall level of disaster damage and directly determines the immediate degree of equipment damage; its weight is secondary, such as 0.25. The flood triggering density measures the frequency of disaster occurrences in the short term; dense flood impacts compress equipment recovery time, and its weight can be set to 0.25. The flood triggering continuity is used to quantify the sustained pressure of the disaster, and its weight is set to 0.2.

[0096] Furthermore, the calculation process for the abnormal coefficient of power grid facility failure growth mentioned in step D2 is as follows: D21, sort the historical floods within the same group according to their time sequence.

[0097] D22. Match the preset penalty coefficient according to the fault type, and correct the fault degree through the penalty coefficient to obtain the comprehensive fault degree.

[0098] D23. Perform differential calculation on the comprehensive fault severity sequence of the same power grid facility to obtain the rate of change of comprehensive fault severity between adjacent events.

[0099] D24. For each power grid facility, calculate the maximum and average of the rates of change of all adjacent events, and determine whether the maximum is less than or equal to 0.

[0100] D25. If the maximum value is less than or equal to 0, then the abnormal growth coefficient of the power grid infrastructure failure is assigned to 0; otherwise, the maximum rate of change is normalized to obtain the abnormal growth coefficient of the infrastructure.

[0101] D26. If the average value is greater than 0, the ratio of the average value to the maximum value is used as the abnormal growth correction factor, and the failure growth abnormal coefficient is corrected by the abnormal growth correction factor.

[0102] D27. Calculate the mean of the corrected failure growth anomaly coefficients for all power grid facilities, and use it as the failure growth anomaly coefficient for the entire homogeneous sequence group.

[0103] Understandably, chronologically sorting historical floods within the same group provides a clear picture of the damage evolution of power grid facilities during successive floods. Based on fault type matching and pre-set penalty coefficients, and considering the significant differences in the impact of different fault types on power grid facility performance and subsequent operation—for example, structural damage to critical components poses a greater potential threat of facility failure than surface corrosion—the penalty coefficients quantify this difference, making the assessment more closely reflect actual risks.

[0104] It should be added that the preset penalty coefficient is obtained by classifying the impact of power grid facility fault types on dimensions such as safety, reliability, repair cost, and life loss. Then, it is combined with historical flood fault data statistics, such as fault occurrence frequency, repair time, and subsequent failure probability, and obtained by referring to power industry standards and expert experience and quantifying and normalizing the comprehensive impact of each fault type through methods such as the analytic hierarchy process. The normalization is the same as the aforementioned normalization method and will not be described again.

[0105] Understandably, by performing differential calculations on the failure severity of the same facility across different historical floods to obtain the rate of change in failure severity between adjacent floods, it is possible to capture the specific changes in facility performance after each flood. Compared to simply focusing on the absolute value of failure severity, the rate of change better reflects the degradation trend of the facility in successive disasters, providing dynamic data support for failure growth assessment.

[0106] The maximum failure rate of change and the average failure rate of change are selected. When the maximum failure rate of change is less than or equal to 0, it indicates that the facility performance has not deteriorated due to flooding, and may even have improved. In this case, the foundation failure growth anomaly coefficient is assigned a value of 0 to reasonably reflect its stable state. If it is greater than 0, the foundation coefficient is obtained by normalizing the maximum failure rate of change, highlighting the degree of facility degradation under the worst conditions.

[0107] When the average rate of change in failure severity is greater than 0, its ratio to the maximum rate of change in failure severity is used as a correction factor. This is because the maximum rate of change may be affected by extreme conditions and thus have an element of chance, while the average rate of change reflects the overall trend. By correcting for the ratio of the two, the relationship between extreme values ​​and the overall trend is balanced, ensuring that the assessment considers both sudden severe degradation situations and long-term stable degradation trends.

[0108] Finally, by averaging the corrected failure growth anomaly coefficients of each power grid facility, the degree of failure growth anomaly of the power grid facilities in the target area under a specific flood sequence can be reflected from an overall perspective. This avoids the interference of the particularity of a single facility on the assessment results and provides a more reliable and unified quantitative indicator for the disaster-bearing vulnerability assessment of power grid facilities.

[0109] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for dynamic assessment of the vulnerability of power grid facilities to flood disasters, characterized in that, include: S1. Obtain the ground elevation of each power grid facility in the target area and the historical flood event dataset of the target area. Each record includes facility identification, trigger time, single flood event characteristic data, and post-flood quantitative facility failure degree value and failure type. S2. Set flood intensity influence factors based on flood event characteristic data and set elevation protection coefficient based on facility ground elevation. After correcting the original fault degree with both, output the balanced fault degree. S3. Extract continuous flood events that belong to the same preset flood intensity range and form a homogeneous flood sequence group. Calculate the failure rate of power grid facilities in this group based on the balanced fault degree setting. S4. The overall failure rate and the number of groups whose failure rate exceeds the preset warning value are statistically analyzed to obtain a disaster bearing vulnerability score and feedback is provided. S5. When a new flood event occurs, repeat steps S1-S4 to update the comprehensive disaster vulnerability score. The specific calculation process for the failure rate of the power grid facilities includes: taking the difference in the degree of balance failure between the first and last floods in the group, dividing it by the average triggering time interval between adjacent historical floods to obtain the failure rate of each power grid facility; fitting a balance failure degree change curve for each power grid facility with time as the horizontal axis and balance failure degree as the vertical axis to determine whether failure rate compensation is needed; if it is determined that failure rate compensation is needed, extracting the ratio of the total length of the curve above the average balance failure degree to the total length of the change curve as the failure rate compensation factor; compensating the failure rate of the power grid facilities with the failure rate compensation factor to obtain the failure rate of each power grid facility; and calculating the final failure rate of the power grid facilities by averaging the failure rate of each power grid facility. The specific statistical process for the disaster vulnerability assessment includes: The failure growth assessment compensation factor is calculated based on the triggering time and flood intensity impact factors of each historical flood within the same group. Based on the degree and type of balance failure of different power grid facilities within the same group during various historical floods, calculate the abnormal coefficient of power grid facility failure growth. The failure rate of the power grid facilities is corrected based on the failure growth assessment compensation factor and the failure growth anomaly coefficient of the power grid facilities. Calculate the difference in average flood intensity influence factors and the difference in corrected failure growth rate for different groups, and match the corresponding reference failure growth rate difference based on the difference in flood intensity influence factors. The ratio of the number of adjacent groups whose statistically corrected failure growth rate difference exceeds the reference failure growth rate difference to the total number of sequence groups is denoted as the failure growth rate difference excess ratio. The ratio of the number of groups whose failure rate exceeds the preset warning value to the total number of sequence groups is denoted as the failure rate exceeding limit ratio. A comprehensive evaluation coefficient is obtained by weighted summation of the failure rate growth rate difference exceeding the limit ratio and the failure rate growth rate exceeding the limit ratio. The comprehensive evaluation coefficient is then multiplied by the preset disaster bearing vulnerability score to obtain the disaster bearing vulnerability score.

2. The method for dynamic assessment of the vulnerability of power grid facilities to flood disasters as described in claim 1, characterized in that: The specific process for setting the flood intensity influencing factor is as follows: The maximum value of the velocity sequence of a single flood is extracted as the target flood velocity, and the standard deviation of the velocity sequence is calculated to obtain the velocity fluctuation. The maximum value of the inundation depth sequence of a single flood is extracted as the target flood inundation depth, and the inundation depth growth rate is obtained by difference calculation. If the rate of increase of flow velocity fluctuation or inundation depth exceeds the corresponding preset threshold, the flow velocity fluctuation and inundation depth growth rate are normalized and then imported into the Sigmoid function to output the flood intensity assessment compensation factor. Obtain the total duration of a single flood event as the target duration of that flood event; The weighting coefficients for target flood velocity, target flood inundation depth, and target duration are assigned using the analytic hierarchy process (AHP). After normalizing the target flood flow velocity, target flood inundation depth and target duration, the basic flood intensity influence factor is obtained by weighted summation. After compensation by the flood intensity assessment compensation factor, the final flood intensity influence factor is output.

3. The method for dynamic assessment of the vulnerability of power grid facilities to flood disasters as described in claim 1, characterized in that: The specific process for setting the elevation protection coefficient includes: Based on the ground elevation, the matching interference inundation depth of each power grid facility is obtained by matching from the preset ground elevation and interference inundation depth mapping table; Assessment of the effectiveness of flood protection based on the depth of inundation. The effective flood protection ratio is obtained by statistically analyzing the ratio of the number of historical floods that were deemed effective for each power grid to the total number of historical floods. The average flood inundation depth of each historical flood that was judged to be effective in protection was calculated, and the difference between the flood inundation depth and the disturbance inundation depth was obtained. At the same time, the average effective flood protection duration was obtained by calculating the average of the total duration of each historical flood that was judged to be effective in protection. After calculating the normalized values ​​of the inundation depth difference and the average effective flood protection duration respectively, the protection correction factor is obtained by weighted summation based on the preset weights of the inundation depth difference and the average effective flood protection duration. The elevation protection coefficient is output after correcting the effective ratio of flood protection by the protection correction factor.

4. The method for dynamic assessment of the vulnerability of power grid facilities to flood disasters as described in claim 3, characterized in that: The specific judgment process for assessing the effectiveness of flood protection includes: The data on inundation depth of each historical flood event is traversed and compared with the inundation depth of each power grid facility. At the same time, the total duration of each flood is extracted and compared with the preset flood interference duration threshold. For each flood event, the power grid facility is deemed to have provided effective flood protection if all of the following conditions are met: The flooding depth is less than or equal to the disturbance flooding depth of the facility; The total duration of the flooding exceeded the preset flood disruption duration; No malfunction caused by flooding was triggered; If any of the above conditions are not met, the power grid facility is deemed ineffective in its flood protection efforts.

5. The method for dynamic assessment of the vulnerability of power grid facilities to flood disasters as described in claim 1, characterized in that: The extraction of the homogeneous flood sequence group must meet the following conditions: The standard deviation of the flood intensity influencing factor within the group is less than or equal to the preset fluctuation threshold; Events within the group are consecutive in time and the interval between them is less than or equal to the preset maximum allowed interval.

6. The method for dynamic assessment of the vulnerability of power grid facilities to flood disasters as described in claim 1, characterized in that: The determination of whether failure rate compensation is needed includes: Extract the total number of peaks and valleys from the balance failure degree variation curve, and use the ratio of the total number of peaks and valleys to the total length of the curve as the frequency of change. If the frequency of change exceeds a preset threshold, it is determined that failure rate growth compensation is required; otherwise, it is determined that failure rate growth compensation is not required.

7. The method for dynamic assessment of the vulnerability of power grid facilities to flood disasters as described in claim 1, characterized in that: The specific calculation of the failure growth assessment compensation factor is as follows: The ratio of consecutive floods is obtained by statistically analyzing the proportion of consecutive floods within the same group to the total number of floods. The mean value of the flood intensity influence factors corresponding to each historical flood within the group is used to obtain the average flood intensity influence factor; Calculate the sequence of intervals between adjacent events based on the trigger time; The flood trigger density is obtained by comparing the number of consecutive floods with a trigger interval shorter than the preset interval to the total number of historical floods. Calculate the average trigger interval of historical flood events within the group and compare it with the preset reference interval. If the average interval is greater than the reference interval, set the continuity to 0. Otherwise, calculate the ratio of the absolute value of the difference between the two to the reference interval as the flood trigger continuity. The failure assessment compensation factor is obtained by weighting and summing the ratio of consecutive floods, the average flood intensity influence factor, the flood triggering density, and the flood triggering continuity.

8. The method for dynamic assessment of the vulnerability of power grid facilities to flood disasters as described in claim 1, characterized in that: The calculation process for the abnormal coefficient of power grid facility failure growth is as follows: Arrange the historical floods within the same group in chronological order; The fault type is matched with a preset penalty coefficient, and the fault severity is corrected by the penalty coefficient to obtain the comprehensive fault severity. Differential calculations are performed on the comprehensive fault severity sequence of the same power grid facility to obtain the rate of change of comprehensive fault severity between adjacent events; For each power grid facility, calculate the maximum and average of the rates of change of all adjacent events, and determine whether the maximum is less than or equal to 0; If the maximum value is less than or equal to 0, the failure growth anomaly coefficient of the power grid facility is assigned to 0; otherwise, the maximum rate of change is normalized to obtain the failure growth anomaly coefficient. If the average value is greater than 0, the ratio of the average value to the maximum value is used as the abnormal growth correction factor, and the failure growth abnormal coefficient is corrected by the abnormal growth correction factor. Calculate the mean of the corrected failure growth anomaly coefficients for all power grid facilities, and use it as the failure growth anomaly coefficient for the entire homogeneous sequence group.

Images