Power facility resilience assessment system based on multi-source seismic data

Abstract

The present invention provides a power facility resilience assessment system based on multi-source seismic data, relating to the field of data processing technology. The system is used to: obtain and filter a seismic data set, eliminate seismic data whose epicenter is irrelevant to a target facility, obtain a seismic screening data set, extract inflection points of vibration amplitude changes, and divide the seismic screening data set into multiple data segments based on the inflection points, splice them in chronological order, and obtain continuous vibration data; establish a vibration transmission path based on the spatial position and connection relationship of each component in the target facility and the propagation direction and action time of the spliced ​​vibration data; calculate the vibration response value of each component in the target facility, determine the component number and vibration response value, calculate the resilience index of the target facility, and obtain a target facility resilience assessment result. The present invention can assess the resilience index of power facilities in multi-source seismic data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a power facility resilience assessment system based on multi-source seismic data. Background Art

[0002] In existing technologies, seismic assessments of power facilities are mostly based on input models of a single earthquake source. The epicenter data with the greatest impact of the main shock is selected, and a ground motion input signal is constructed and applied to the power facility model to predict the potential structural response. This method simplifies the seismic environment and is suitable for risk analysis in the context of an isolated earthquake source. However, in seismic belts where multiple earthquake sources act simultaneously or in close sequence, it may underestimate the cumulative and overlapping effects of complex vibration patterns on power facilities.

[0003] Existing technologies introduce multi-source scenario simulations, often using superimposed waveforms, selecting mainshock and aftershock combinations, or simulating multiple vibration sequences based on regional seismic activity maps. However, these methods may ignore the local response enhancement effect of vibration propagation direction and superimposed timing on weak parts of power facilities during the assessment process. For example, in long-distance transmission lines in mountainous areas, such facilities often pass through sections affected by multiple seismic sources. When multiple seismic sources act together in different directions and magnitudes, the tower foundations with smaller stiffness in certain directions in the structure may have high-amplitude responses at specific nodes due to frequency matching. However, traditional methods only evaluate the mainshock response and may ignore the resonant excitation of secondary vibrations in specific modes of the structure, resulting in the assessment results underestimating the failure risk. Summary of the Invention

[0004] The purpose of the present invention is to provide a power facility resilience assessment system based on multi-source seismic data, aiming to solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0006] A power facility resilience assessment system based on multi-source seismic data, the system comprising:

[0007] A screening module is used to obtain and screen earthquake data sets, eliminating earthquake data whose epicenters are not related to the target facility, and obtaining an earthquake screening data set;

[0008] A segmentation module is used to extract the inflection points of the vibration amplitude change based on the earthquake screening data set to obtain vibration inflection point data, and to divide the earthquake screening data set into multiple data segments based on the inflection point data to obtain segmented data sets;

[0009] The combination module is used to splice the data segments in the segmented data set in their time sequence and retain the start and end position labels of each data segment in the overall sequence to obtain continuous vibration data;

[0010] The action module is used to match the continuous vibration data with the components of the target facility according to the spatial position and connection relationship of each component in the target facility, establish the vibration transmission path according to the propagation direction and action time of the spliced ​​vibration data, and obtain the response path data;

[0011] An identification module is used to calculate the vibration response value of each component in the target facility based on the response path data, identify high-response components with multi-source superposition enhancement characteristics, and obtain a component response identification set;

[0012] The evaluation module is used to determine the component number and vibration response value of each component according to the component response identification set, calculate the resilience index of the target facility, and obtain the resilience evaluation result of the target facility.

[0013] Furthermore, the screening module includes:

[0014] A source direction analysis unit is used to calculate the spatial coordinate difference between the epicenter position and the position of the target facility component according to the epicenter position corresponding to each earthquake data in the earthquake data set, and obtain the source direction vector;

[0015] A component orientation extraction unit is used to obtain the structural orientation vector of each component in the target facility;

[0016] A direction angle calculation unit is used to calculate the component angle between the source direction vector of each seismic data and the structural orientation vector of each component to obtain a component angle set;

[0017] The direction screening judgment unit is used to judge whether the seismic data has direction correlation with the target facility based on the component angle set and the preset angle threshold. When the component angle is greater than the preset angle threshold, the seismic data is eliminated to obtain the seismic screening data set.

[0018] Furthermore, the segmentation module includes:

[0019] An amplitude sequence construction unit is used to extract the vibration amplitude of each earthquake data according to the earthquake screening data set to obtain a vibration amplitude data set;

[0020] An extreme point identification unit is used to identify local maximum points and local minimum points based on the vibration amplitude data set to obtain an extreme point sequence;

[0021] The fluctuation change rate calculation unit is used to calculate the amplitude difference and extreme value time difference between adjacent extreme value points based on the extreme value point sequence to obtain the amplitude change rate sequence;

[0022] The inflection point determination unit is used to identify the time point with significant amplitude change rate as the inflection point of vibration amplitude change according to the amplitude change rate sequence, and obtain vibration inflection point data.

[0023] Furthermore, the inflection point determination unit includes:

[0024] a rate of change difference calculation unit, configured to calculate the amplitude rate of change difference between each extreme point and the previous extreme point according to the amplitude rate of change sequence, to obtain a rate of change difference sequence;

[0025] A threshold comparison unit is used to compare the values ​​of the rate of change difference sequence with a preset difference threshold value. When the amplitude change rate difference exceeds the preset difference threshold value, the extreme point is marked as a candidate inflection point to obtain candidate inflection point data.

[0026] The minimum time distance verification unit is used to calculate the time interval between adjacent candidate inflection points based on the candidate inflection point data to obtain the inflection point time difference, and to filter out candidate inflection points whose inflection point time difference is less than a preset time difference threshold based on the inflection point time difference;

[0027] The continuous segment mean comparison unit is used to calculate the mean difference of the vibration amplitude of each fixed-length segment before and after the dividing point with each candidate inflection point as the dividing point, and compare the mean difference with the preset mean difference threshold. When the mean difference is greater than the preset mean difference threshold, the candidate inflection point is retained to obtain the vibration inflection point data.

[0028] Furthermore, the combined module includes:

[0029] A time sequence sorting unit is used to sort the segmented data sets in time sequence according to the earthquake time corresponding to each data segment, and generate a sorting index sequence;

[0030] A data segment splicing unit is used to sequentially splice the data segments according to the sorting index sequence to generate a continuous vibration sequence;

[0031] The segment boundary position marking unit is used to record the start and end positions of each data segment in the continuous vibration sequence and generate a segment boundary position information set;

[0032] The time tag adding unit is used to add the start and end position tags of each data segment boundary in the continuous vibration sequence according to the segment boundary position information set, and mark the corresponding source number to obtain continuous vibration data.

[0033] Furthermore, the action modules include:

[0034] A vibration data mapping unit is used to bind the time information of each data segment in the continuous vibration data to the source direction vector of its corresponding source number to form a direction-time comparison table;

[0035] A component sequence generation unit is used to construct a component path sequence according to the installation position and connection relationship of each component of the target facility in space, and obtain a component path data set;

[0036] The action time unit is used to calculate the action time of each component receiving continuous vibration data based on the direction time comparison table and the component path data set;

[0037] The component path construction unit is used to establish a vibration transmission path with a time tag according to the sequence of the action time of each component and obtain the response path data.

[0038] Furthermore, the action time unit includes:

[0039] A propagation starting point identification unit is used to select, in each component path sequence, the component with the smallest component angle with the earthquake source direction vector as the propagation starting point component;

[0040] A propagation distance calculation unit is used to accumulate and calculate the physical distance between the propagation starting point component and any component on the component path sequence to obtain propagation path distance data;

[0041] A propagation delay estimation unit is used to calculate the action time of components based on the vibration propagation speed and propagation path distance data of different earthquake data;

[0042] The time mapping generation unit is used to match the action time of each component with the time information of each data segment in the continuous vibration data to obtain the vibration action time point of each component.

[0043] Furthermore, the identification module includes:

[0044] A vibration response value unit is used to calculate the path response value of each component in different vibration transmission paths according to the response path data, and summarize the values ​​to obtain the vibration response value of the component;

[0045] a response threshold comparison unit, configured to compare a vibration response value of a component with a preset response threshold, and mark the component as an abnormal response component when the vibration response value is greater than the preset response threshold;

[0046] The superposition enhancement identification unit is used to determine whether the vibration response value of the abnormal response component is jointly determined by two or more path response values. When the result is yes, the abnormal response component is a high response component.

[0047] Furthermore, the vibration response value unit includes:

[0048] The path response value calculation unit is used to analyze the positive impact of the maximum vibration amplitude transmitted from the earthquake source to the component on the component based on the response path data and calculate the peak direction term; calculate the total value of the vibration amplitude transmitted from the earthquake source to the component, and correct the total vibration amplitude value by the delay factor formed by the action time to obtain the energy delay correction term; based on the peak direction term and the energy delay correction term, the comprehensive impact of different earthquake sources on the components along the vibration transmission path is calculated to obtain the path response value;

[0049] The vibration response value calculation unit is used to count the number of vibration transmission paths on which the component is located, and calculate the comprehensive impact on the component based on the path response values ​​of the component on different vibration transmission paths to obtain the vibration response value.

[0050] Furthermore, the assessment modules include:

[0051] A component state determination unit is used to determine whether the component is in a normal, critical or failed state based on the vibration response value of each component in the component response identification set, and obtain component state data;

[0052] A toughness weight allocation unit is used to allocate a toughness weight value to each component according to the structure of each component in the target facility to obtain component weight data;

[0053] The facility resilience index unit is used to calculate the resilience index of the target facility based on the component status data and component weight data, and obtain the target facility resilience assessment result.

[0054] The above solution of the present invention includes at least the following beneficial effects:

[0055] The present invention can effectively eliminate earthquake information that has no direct correlation with the spatial distribution of the target facility by performing spatial screening operations on the epicenter position of the earthquake data set, thereby retaining earthquake data that may affect the target facility. This dual calculation mechanism of the spatial difference vector and the target component structure orientation angle effectively establishes a physical spatial correspondence between the earthquake source and the facility, and solves the technical problem that the existing technology of using a large-scale geographical area to select points may introduce a large number of invalid sources. The component angle calculation and threshold screening process further improves the directional accuracy of the screening, ensuring that the input earthquake data is highly representative in the actual evaluation, and can ensure that the selected data has actual disaster-causing potential in the context of multiple sources, avoiding interference of the evaluation model by invalid or irrelevant data.

[0056] The present invention constructs a vibration amplitude sequence, identifies local extreme points and calculates the fluctuation change rate by performing vibration amplitude analysis and inflection point identification on seismic data, thereby extracting the time nodes of significant vibration changes and dividing them into multiple data segments with independent dynamic characteristics, thereby avoiding the problem of information loss in traditional whole-segment waveform processing methods, and enabling the system to capture significant response changes in multiple bands throughout the earthquake process, especially for the mixed action scenarios of foreshocks, main shocks and aftershocks. It also enhances the ability to identify abnormal change sections through strategies such as change rate threshold judgment and mean comparison, and can effectively extract vibration sequences that are highly correlated with the response of target facilities in terms of frequency, amplitude and energy distribution, which not only improves the ability to capture complex earthquake structures, but also provides a clearly structured and combinable vibration unit for subsequent splicing and path construction.

[0057] The present invention can splice multiple segmented data fragments into a complete vibration sequence according to the time sequence, and add the source number and start and end time information at the boundary of each segment to form a vibration input signal with physical origin and time label. It not only solves the alignment problem of multi-source seismic data on the time line, but also retains the physical driving path of the source superposition time sequence in the structural response. Compared with the traditional method that only uses the main shock waveform of a single source, this method provides complete, multi-segment time evolution data, which can truly reflect the superposition and interference effects of multi-source vibrations on the facility structure, especially in the complex earthquake background with strong coupling between the main shock and aftershocks. The establishment of this continuous vibration data provides a unified framework for subsequent vibration propagation path analysis, time difference estimation and component response calculation, and has good engineering applicability and data consistency.

[0058] The present invention establishes a transmission path model from source input to structural response by comprehensively considering the component layout, connection relationship and source direction of the target facility. By generating a component path sequence and combining the matching operation of direction vector and action time, the system can simulate the path of real vibration propagation along the facility components and identify the time point when each component is subjected to vibration, thereby realizing the time-space full path mapping between vibration input and structural response. Compared with the traditional method of simplifying to uniform application of vibration, this path construction mechanism more realistically simulates the nonlinear characteristics of vibration propagation caused by the structural layout inside the facility, effectively characterizes the aggregation or attenuation behavior of vibration energy, thereby improving the assessment sensitivity and identification ability of high-risk areas.

[0059] The present invention integrates the component response identification set, component structural status and structural importance to form a toughness index calculation unit based on the component as the basic unit. It not only considers the absolute strength of the component response value, but also quantifies the impact of different types of components on the entire facility through status discrimination and structural weight distribution strategy, forming a weighted facility toughness evaluation result. Compared with the existing macro method based on maximum stress or total displacement as indicators, the evaluation path of this system is more fine-grained and physically interpretable. It can retain the response contribution of local high-risk nodes in the overall status assessment of the facility, thereby forming a more realistic and operational evaluation result. It is particularly suitable for multi-level component system architecture and effectively supports layered reinforcement and structural optimization decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a flowchart of a system for evaluating the resilience of electric power facilities based on multi-source seismic data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0062] like Figure 1 As shown, an embodiment of the present invention provides a power facility resilience assessment system based on multi-source seismic data, the system comprising:

[0063] A screening module is used to obtain and screen earthquake data sets, eliminating earthquake data whose epicenters are not related to the target facility, and obtaining an earthquake screening data set;

[0064] A segmentation module is used to extract the inflection points of the vibration amplitude change based on the earthquake screening data set to obtain vibration inflection point data, and to divide the earthquake screening data set into multiple data segments based on the inflection point data to obtain segmented data sets;

[0065] The combination module is used to splice the data segments in the segmented data set in their time sequence and retain the start and end position labels of each data segment in the overall sequence to obtain continuous vibration data;

[0066] The action module is used to match the continuous vibration data with the components of the target facility according to the spatial position and connection relationship of each component in the target facility, establish the vibration transmission path according to the propagation direction and action time of the spliced ​​vibration data, and obtain the response path data;

[0067] An identification module is used to calculate the vibration response value of each component in the target facility based on the response path data, identify high-response components with multi-source superposition enhancement characteristics, and obtain a component response identification set;

[0068] The evaluation module is used to determine the component number and vibration response value of each component according to the component response identification set, calculate the resilience index of the target facility, and obtain the resilience evaluation result of the target facility.

[0069] In an embodiment of the present invention, a screening module is used to obtain and screen an earthquake data set, eliminate earthquake data whose epicenter is irrelevant to the target facility, and obtain an earthquake screening data set, effectively filtering out earthquake data with a weak spatial relationship between the epicenter and the facility, reducing redundant input, and improving the relevance and quality of subsequent processing data; a segmentation module is used to extract the inflection point of the vibration amplitude change based on the earthquake screening data set, obtain vibration inflection point data, and divide the earthquake screening data set into multiple data segments based on the inflection point data to obtain a segmented data set, and realize the hierarchical division of the vibration dynamic structure in the earthquake data by extracting significant features of the vibration amplitude, which is helpful for subsequent Local modeling of vibration data and its interference effects improves the clarity of data expression and the ability to characterize the vibration energy conversion process, and enhances the system's adaptability to complex seismic environments; the combination module is used to splice the various data segments in the segmented data set in chronological order, and retain the start and end position labels of each data segment in the overall sequence to obtain continuous vibration data. Through sequential splicing and numbering, the temporal structure of different earthquake sources is retained, so that the characteristics of the composite action of multiple earthquake sources can be fully expressed. The segment boundary position information and source labeling can be used as key inputs for subsequent propagation path and response superposition analysis, thereby improving the continuity and integrity of the vibration event chain modeling.

[0070] The action module is used to match the continuous vibration data with the components of the target facility according to the spatial position and connection relationship of each component in the target facility, establish the vibration transmission path according to the propagation direction and action time of the spliced ​​vibration data, obtain the response path data, and accurately map the continuous vibration data to the spatial network of the structural components, realizing the correspondence between the vibration propagation direction and the structural response time sequence. Through the propagation path modeling, it helps to reveal the real transmission route of the vibration energy in the facility, provide a clear cause-effect chain for the subsequent component-level response simulation, and improve the credibility of the analysis; the identification module is used to calculate the vibration response value of each component in the target facility according to the response path data, and identify high-response components with multi-source superposition enhancement characteristics. , and obtain the component response identification set, realizing the quantitative calculation of the response intensity of each component under complex vibration. Different from the traditional method of only evaluating the overall structure, identifying superimposed response components can effectively find potential high-risk nodes caused by vibration direction coupling or frequency resonance, which is helpful to carry out key reinforcement and structural optimization design in advance; the evaluation module is used to determine the component number and vibration response value of each component according to the component response identification set, calculate the resilience index of the target facility, and obtain the resilience assessment result of the target facility. The response intensity is coupled with the structural importance to form a quantitative resilience index, which can accurately reflect the function retention ability of the structure under earthquake impact, realize the risk level division from component to the whole, and improve the pertinence and efficiency of seismic strategy deployment.

[0071] In a preferred embodiment of the present invention, the screening module includes:

[0072] A source direction analysis unit is used to calculate the spatial coordinate difference between the epicenter position and the position of the target facility component according to the epicenter position corresponding to each earthquake data in the earthquake data set, and obtain the source direction vector;

[0073] A component orientation extraction unit is used to obtain the structural orientation vector of each component in the target facility;

[0074] A direction angle calculation unit is used to calculate the component angle between the source direction vector of each seismic data and the structural orientation vector of each component to obtain a component angle set;

[0075] The direction screening judgment unit is used to judge whether the seismic data has direction correlation with the target facility based on the component angle set and the preset angle threshold. When the component angle is greater than the preset angle threshold, the seismic data is eliminated to obtain the seismic screening data set.

[0076] In an embodiment of the present invention, a source direction analysis unit is used to calculate the spatial coordinate difference between the epicenter position and the position of the target facility component according to the epicenter position corresponding to each earthquake data in the earthquake data set, and obtain the source direction vector. Through the source direction vector, a quantitative judgment is made on whether the vibration propagates along the sensitive direction of the facility in the subsequent process, and a direction dimension is introduced, which effectively improves the accuracy of spatial correlation judgment; a component orientation extraction unit is used to obtain the structural orientation vector of each component in the target facility, and by extracting the orientation information of each component in the facility, an accurate basis is provided for the subsequent judgment on whether the direction of the impact of the source on the component constitutes a frontal impact; a direction angle calculation unit is used to calculate the direction of each earthquake. The component angle set is obtained by calculating the component angle between the source direction vector of the data and the structural orientation vector of each component. The angle value can be used to determine whether the vibration acts along the main force direction of the component, thereby identifying potential high-risk earthquake sources; the direction screening judgment unit is used to determine whether the earthquake data has direction correlation with the target facility based on the component angle set and the preset angle threshold. When the component angle is greater than the preset angle threshold, the earthquake data is eliminated to obtain the earthquake screening data set. An intelligent judgment mechanism combining spatial directionality is constructed, which can automatically identify which earthquake data has insufficient effect on the target facility to attract attention, and automatically exclude direction-insensitive data, thereby reducing the data scale and improving evaluation efficiency.

[0077] The direction screening judgment unit is used to judge whether the seismic data has direction correlation with the target facility based on the component angle set and the preset angle threshold. When the component angle is greater than the preset angle threshold, the seismic data is eliminated to obtain the seismic screening data set, which specifically includes:

[0078] First, obtain the component angle set constructed in the previous step. This angle set records the spatial direction angle between each seismic data and each component in the target facility. In order to determine whether the seismic data has a physical correlation with the facility components in terms of directional transmission, the system needs to set a clear preset angle threshold as a baseline for determining whether the source direction significantly affects the target component.

[0079] In practice, the preset angle threshold is typically set between 30° and 75°. For example, in the context of a typical overhead transmission line project in mountainous areas, since the tower structure is sensitive to the direction of vibration propagation, seismic waves with an angle of less than 45° are considered to have a significant effect. Therefore, the system can set the preset angle threshold to 45°. This threshold is typically determined based on structural response spectrum simulation data or historical earthquake damage analysis reports. In particular, through regression analysis of the correlation between typical earthquake source directions and facility damage distribution, an empirical critical value that effectively demarcates "correlated" and "uncorrelated" vibration directions is obtained.

[0080] During the judgment process, the system will traverse each piece of seismic data in turn and retrieve all its angle values ​​in the component angle set. When the system finds that among the angles corresponding to the seismic data, more than half of the key component angle values ​​are greater than the preset angle threshold, it is determined that the source direction lacks sufficient directional influence on the target facility, and the seismic data will be removed from the data set. On the contrary, if there are multiple components with angles within the threshold value and the source direction, it means that the source has a certain degree of directional correlation, and the system will retain the seismic data for subsequent evaluation. In the final output stage, the system collects all seismic data that have passed the judgment to generate an "earthquake screening data set", which only retains vibration inputs that have spatial directional correlation with the target facility components, providing a more accurate data basis for subsequent vibration amplitude analysis and transmission path construction. This process greatly improves the physical rationality and computational efficiency of subsequent evaluations, and significantly reduces the interference of invalid data on the toughness assessment results.

[0081] In a preferred embodiment of the present invention, the segmentation module includes:

[0082] An amplitude sequence construction unit is used to extract the vibration amplitude of each earthquake data according to the earthquake screening data set to obtain a vibration amplitude data set;

[0083] An extreme point identification unit is used to identify local maximum points and local minimum points based on the vibration amplitude data set to obtain an extreme point sequence;

[0084] The fluctuation change rate calculation unit is used to calculate the amplitude difference and extreme value time difference between adjacent extreme value points based on the extreme value point sequence to obtain the amplitude change rate sequence;

[0085] The inflection point determination unit is used to identify the time point with significant amplitude change rate as the inflection point of vibration amplitude change according to the amplitude change rate sequence, and obtain vibration inflection point data.

[0086] In an embodiment of the present invention, an amplitude sequence construction unit is used to extract the vibration amplitude of each piece of earthquake data from an earthquake screening data set to obtain a vibration amplitude data set, providing standardized and highly continuous original basic data for subsequent extreme point identification and amplitude change analysis. An extreme point identification unit is used to identify local maximum points and local minimum points from the vibration amplitude data set to obtain an extreme point sequence, which can effectively extract key points representing the characteristics of earthquake energy fluctuations from continuous vibration data and provide a basis for analyzing the stage of violent vibration changes during an earthquake. A fluctuation change rate calculation unit is used to calculate the amplitude difference and extreme time difference between adjacent extreme points based on the extreme point sequence to obtain an amplitude change rate sequence, which can quantify the energy change rate of seismic waves in a short period of time and reflect the intensity and evolution trend of the vibration waveform. An inflection point determination unit is used to identify the time point with a significant amplitude change rate as the inflection point of the vibration amplitude change based on the amplitude change rate sequence to obtain vibration inflection point data, effectively improving the reliability and physical representativeness of vibration inflection point identification, accurately identifying the nodes of violent vibration changes, and laying a solid foundation for subsequent vibration segmentation and splicing.

[0087] The fluctuation change rate calculation unit is used to calculate the amplitude difference and extreme time difference between adjacent extreme value points according to the extreme value point sequence to obtain the amplitude change rate sequence, which specifically includes:

[0088] In practice, after receiving the extreme point sequence output by the extreme point identification unit, the fluctuation change rate calculation unit first performs a linear traversal of the sequence. The extreme point sequence is an ordered set, with each extreme point containing two basic pieces of information: its corresponding time position and its corresponding vibration amplitude. The system then sequentially selects any two adjacent extreme points in the sequence to form an extreme point pair. It then reads the vibration amplitude and time values ​​of the first point, denoted as the preceding point amplitude and preceding point time, and simultaneously reads the vibration amplitude and time values ​​of the second point, denoted as the following point amplitude and following point time. The system then calculates the difference in vibration amplitude change between the two extreme points by subtracting the preceding point amplitude from the following point amplitude and taking the absolute value of the difference, which represents the magnitude of the amplitude change. The system also calculates the time interval between the two extreme points by subtracting the preceding point time from the following point time to obtain the time span. After completing these calculations, the system divides the vibration amplitude change value by the corresponding time interval to obtain the amplitude change rate corresponding to the extreme point pair, which represents the rate of change of vibration during that time period. The system uses this rate of change as the result value and stores it in the rate of change sequence in the order of calculation, ultimately forming a complete amplitude rate of change data set. Throughout this process, to avoid numerical instability caused by time intervals approaching zero, the system can set a minimum time difference threshold. When the time interval between extreme point pairs falls below this threshold, the point pair is automatically eliminated, ensuring computational stability and interpretability of the results. This sequence serves as an important reference for determining the severity of vibration during the subsequent inflection point determination process.

[0089] In a preferred embodiment of the present invention, the inflection point determination unit includes:

[0090] a rate of change difference calculation unit, configured to calculate the amplitude rate of change difference between each extreme point and the previous extreme point according to the amplitude rate of change sequence, to obtain a rate of change difference sequence;

[0091] A threshold comparison unit is used to compare the values ​​of the rate of change difference sequence with a preset difference threshold value. When the amplitude change rate difference exceeds the preset difference threshold value, the extreme point is marked as a candidate inflection point to obtain candidate inflection point data.

[0092] The minimum time distance verification unit is used to calculate the time interval between adjacent candidate inflection points based on the candidate inflection point data to obtain the inflection point time difference, and to filter out candidate inflection points whose inflection point time difference is less than a preset time difference threshold based on the inflection point time difference;

[0093] The continuous segment mean comparison unit is used to calculate the mean difference of the vibration amplitude of each fixed-length segment before and after the dividing point with each candidate inflection point as the dividing point, and compare the mean difference with the preset mean difference threshold. When the mean difference is greater than the preset mean difference threshold, the candidate inflection point is retained to obtain the vibration inflection point data.

[0094] In an embodiment of the present invention, a rate of change difference calculation unit is used to calculate the amplitude rate of change difference between each extreme point and the previous extreme point based on the amplitude rate of change sequence, and obtain a rate of change difference sequence, which can reveal the severity of the waveform change during the vibration process and identify the key points where the amplitude fluctuates violently in a short period of time; a threshold comparison unit is used to perform numerical comparison based on the rate of change difference sequence and a preset difference threshold. When the amplitude rate of change difference exceeds the preset difference threshold, the extreme point is marked as a candidate inflection point to obtain candidate inflection point data. By setting the preset difference threshold, the system can automatically distinguish which extreme points dominate the amplitude fluctuation; a minimum time distance verification unit is used to calculate the time interval between adjacent candidate inflection points based on the candidate inflection point data. , and the inflection point time difference is obtained, and the candidate inflection points whose inflection point time difference is less than the preset time difference threshold are screened out based on it. By introducing the minimum time interval verification mechanism, the problem of generating a large number of dense candidate inflection points due to short-period oscillations in the vibration waveform is avoided, thereby reducing misjudgment and redundant calculations; the continuous segment mean comparison unit is used to use each candidate inflection point as the demarcation point, calculate the mean difference of the vibration amplitude of each fixed-length segment before and after the demarcation point, and compare the mean difference with the preset mean difference threshold. When the mean difference is greater than the preset mean difference threshold, the candidate inflection point is retained to obtain the vibration inflection point data. Based on the horizontal level comparison of the vibration amplitude before and after the inflection point, it is effectively judged whether the node is accompanied by a significant turning point in the vibration mode, thereby enhancing the physical basis of inflection point identification.

[0095] The preset difference threshold identifies nodes where amplitude changes suddenly, reflecting the key points where the vibration mode switches from stable to violent or from low frequency to high frequency. This threshold is primarily determined by statistical analysis of historical seismic waveform data. For example, data from 20 earthquakes intersecting the facility's spatial structure were collected from a 110kV transmission line in a mountainous area of ​​Yunnan. The extreme amplitude values ​​were extracted, and the rate of change differences between each extreme value pair were calculated and statistically analyzed for their distribution characteristics. Empirical results show that when the peak amplitude fluctuation exceeds 20 gal / s and the rate of change difference exceeds 12 gal / s, a transition in the facility response spectrum often occurs. Therefore, a threshold of 12 gal / s can be set as the initial threshold. After the system is initially deployed, adaptive fine-tuning can be performed based on real-time input seismic waveform data.

[0096] Among them, the preset time difference threshold is to eliminate the adjacent redundant inflection points caused by local high-frequency vibrations. The setting of this threshold needs to be considered in combination with the earthquake propagation speed and the response period characteristics of the facility structure. Taking the mountainous transmission line as an example, the rock conditions in this area cause the earthquake propagation speed to be about 2.5km / s, and the first-order natural period of structural components such as transmission tower poles is usually between 0.2-0.5 seconds. Considering that the shortest interval for earthquake waves to produce obvious vibration structural response within a range of 1km is 0.1-0.2 seconds, the system can take a conservative value to avoid the risk of vibration response aliasing. Therefore, it is recommended to set it to 0.25 seconds, that is, only when the time interval between two candidate inflection points is greater than 0.25 seconds, the system considers them to be two independent response events.

[0097] Among them, the preset mean difference threshold is used to confirm that there is a significant change in the vibration energy distribution on both sides of the inflection point, in order to exclude micro-fluctuation nodes that have no practical significance. The method for determining this parameter is based on the segmented statistical mean change. In the aforementioned actual scenario, after comparing a large number of historical vibration sections, it was found that when the average amplitude change is above 8gal, it usually corresponds to key engineering points such as the actual epicenter impact enhancement, source direction conversion, or main aftershock switching. In order to enhance the sensitivity of the system, it can be set to 8gal during initial deployment, and dynamically optimized in combination with the online vibration amplitude distribution model during subsequent operation, so that the system can adapt to different earthquake magnitudes and facility response levels, taking into account both accuracy and computational cost.

[0098] In a preferred embodiment of the present invention, the combined module includes:

[0099] A time sequence sorting unit is used to sort the segmented data sets in time sequence according to the earthquake time corresponding to each data segment, and generate a sorting index sequence;

[0100] A data segment splicing unit is used to sequentially splice the data segments according to the sorting index sequence to generate a continuous vibration sequence;

[0101] The segment boundary position marking unit is used to record the start and end positions of each data segment in the continuous vibration sequence and generate a segment boundary position information set;

[0102] The time tag adding unit is used to add the start and end position tags of each data segment boundary in the continuous vibration sequence according to the segment boundary position information set, and mark the corresponding source number to obtain continuous vibration data.

[0103] In an embodiment of the present invention, a time sequence sorting unit is used to sort the segmented data sets in time sequence according to the earthquake time corresponding to each data segment, generate a sorting index sequence, ensure the continuity of the earthquake data segments on the logical time axis, and avoid the situation where the vibration splicing order is incorrect due to time sequence confusion; a data segment splicing unit is used to splice the data segments in sequence according to the sorting index sequence to generate a continuous vibration sequence, integrating the discrete vibration segments into a complete vibration input signal that is continuous in the time domain and smoothly transitions in the amplitude domain; a segment boundary position marking unit is used to record the start and end positions of each data segment in the continuous vibration sequence and generate a segment boundary position information set. After splicing, the system still retains the source information of each original segment, so that the overall data still has segment traceability under a unified structure; a time label adding unit is used to add the start and end position labels of each data segment boundary in the continuous vibration sequence according to the segment boundary position information set, and mark the corresponding earthquake source number to obtain continuous vibration data, thereby achieving a one-to-one correspondence between the earthquake data and its time and space characteristics, and ensuring that the subsequent propagation path construction module can accurately determine the vibration source and its action sequence.

[0104] The time tag adding unit is used to add the start and end position tags of each data segment boundary in the continuous vibration sequence according to the segment boundary position information set, and mark the corresponding source number to obtain continuous vibration data, which specifically includes:

[0105] The system first receives the segment boundary position information set generated by the segment boundary position marking unit. This information set usually contains the start and end point indexes of each data segment in the continuous vibration sequence. After reading this information, the system establishes a mapping relationship between the segment number and the timestamp, and constructs an index array in the data structure, for example, using the key-value pair format of "segment number → [start position, end position]". Subsequently, the system calculates the corresponding time start and end point of each data segment in the entire sequence based on the data sampling rate and original time information of the continuous vibration sequence. For example, if the start time of the original data segment is , the data sampling rate is , then the first The full sequence timestamp corresponding to the sampling points can be expressed as ,In this way, the system can restore the physical time attributes of the data segments to the ,overall sequence after splicing.

[0106] After obtaining the time tag of each data segment, the system attaches a source number to each data segment. The source number is usually calculated from the attributes of the original earthquake data such as the epicenter position, magnitude, and earthquake occurrence time, or is directly called by the earthquake data index retained by the screening module. To ensure that the data is not tampered with and to facilitate the parsing of subsequent modules, the system uses structured data embedding to attach these label information to each segment boundary point. The specific method includes: inserting a label vector before and after the sequence boundary data point. The label content includes the segment number, source number, start and end time, original source position, etc. The format can be inserted as a JSON string, or the label can be stored independently in the metadata set and associated with the continuous vibration sequence through an index.

[0107] To enhance the scalability of the system, a redundant verification mechanism is introduced into the label attachment operation. The system verifies the correctness of each segment of attached labels to ensure that its start and end positions are consistent with the segment boundary position information, the timestamp continuity is reasonable, and it is consistent with the original source data. When the label is attached, the system outputs the continuous vibration data containing the label as a structured data object such as a NumPy array with a time index or a nested dictionary structure for subsequent call in the vibration propagation path construction process. Through this labeling mechanism, the system realizes the closed-loop management of the data structure from earthquake input to response modeling, ensuring that the vibration source information remains valid in subsequent path analysis and response calculations, greatly enhancing the integrity of data processing and source traceability, and laying the foundation for the ultimate realization of high-precision resilience assessment.

[0108] In a preferred embodiment of the present invention, the action module includes:

[0109] A vibration data mapping unit is used to bind the time information of each data segment in the continuous vibration data to the source direction vector of its corresponding source number to form a direction-time comparison table;

[0110] A component sequence generation unit is used to construct a component path sequence according to the installation position and connection relationship of each component of the target facility in space, and obtain a component path data set;

[0111] The action time unit is used to calculate the action time of each component receiving continuous vibration data based on the direction time comparison table and the component path data set;

[0112] The component path construction unit is used to establish a vibration transmission path with a time tag according to the sequence of the action time of each component and obtain the response path data.

[0113] In an embodiment of the present invention, a vibration data mapping unit is used to bind the time information of each data segment in the continuous vibration data with the source direction vector of its corresponding source number to form a direction-time comparison table. By combining the vibration data with the spatial direction of its physical origin, the spatial logic of the vibration propagation path is retained, providing a reliable spatial basis for subsequent judgment of how the vibration passes through the component system; a component sequence generation unit is used to construct a component path sequence according to the installation position and connection relationship of each component of the target facility in space, obtain a component path data set, construct a propagation network map inside the power facility, and realize the conversion of physical structure information into a topological network for vibration transmission. The system is used to simulate the time sequence of vibration from the source to each component of the target facility based on the direction-time comparison table and the component path data set. The system also simulates the non-instantaneous propagation of shock waves in the real world. The component path construction unit is used to establish a vibration transmission path with time tags according to the sequence of the action times of each component, obtain response path data, and realize the reconstruction of the time-space path of the vibration propagation process. It has a traceable and visual analysis basis, enabling the system to track how the vibration energy propagates along the facility structure to each key component on a path basis.

[0114] The vibration data mapping unit is used to bind the time information of each data segment in the continuous vibration data to the source direction vector of its corresponding source number to form a direction-time comparison table, which specifically includes:

[0115] After the continuous vibration data is generated, the system performs a structured extraction operation on each segment of data. First, it reads the source number, start time, and end time of the data segment as an index item. Then, it calls the source information database constructed by the screening module, finds the epicenter coordinate position corresponding to the source according to the source number, and calculates the source direction vector through the three-dimensional coordinate difference based on the installation coordinates of the target facility component in the geographic space. Subsequently, the direction vector is bound to the time information of the vibration data segment to form a mapping relationship table with "source number" as the primary key and "time range-direction vector" as the value. The system names it the direction-time comparison table. The table can be arranged in chronological order to facilitate the directional time-series mapping of the vibration propagation path in the subsequent component path construction unit, and provides an accurate reference in space and time for the calculation of each component receiving vibration.

[0116] The component sequence generation unit is used to construct a component path sequence based on the installation position and connection relationship of each component of the target facility in space to obtain a component path dataset, which specifically includes:

[0117] The system first extracts the spatial layout information of all components from the facility design blueprint or BIM structural model, including the unique component number, three-dimensional installation coordinates and their structural connection relationships. For each component, the system identifies its physical connection path in the facility, such as the connection between poles and towers through wires, the connection between poles and substations through supporting structures, etc. Through graph traversal algorithms such as depth-first or breadth-first search, the system establishes a propagation path graph between components, and arranges the component numbers of each path in the order of connection to form a component sequence. For each path, the system saves its path number, the spatial coordinate sequence of all components on the path, and the structural connection topology to form a path-level component path dataset. This dataset will serve as the structural framework for vibration propagation modeling, providing a basic path structure for propagation direction matching and response analysis of different earthquake sources.

[0118] The component path construction unit is used to establish a vibration transmission path with time tags according to the order of the action time of each component and obtain the response path data, which specifically includes:

[0119] The system reads the precise time each component receives the vibration from the action time unit and, combined with the topological sequence in the component path dataset, sorts the components in the path according to the order of action time. For each component path, the system integrates the component number, component coordinates, corresponding action time, source source number, and propagation direction vector to form a transmission path structure that includes both time sequence and direction. Each component node in the path is assigned a time tag and source source information, forming a time-annotated propagation chain. To ensure data consistency, the system detects abnormalities in the path, such as inverted action times. If data inconsistent with the propagation logic is found, the path is automatically corrected or removed. Furthermore, the system assigns a unique response path number to each path and retains the propagation time differences between components in the response path data for use in delay correction and energy superposition analysis in subsequent response value calculations. Ultimately, response path data with complete time-direction-space characteristics is generated, enabling the recognition module to calculate path response values ​​and component composite response values.

[0120] In a preferred embodiment of the present invention, the action time unit includes:

[0121] A propagation starting point identification unit is used to select, in each component path sequence, the component with the smallest component angle with the earthquake source direction vector as the propagation starting point component;

[0122] A propagation distance calculation unit is used to accumulate and calculate the physical distance between the propagation starting point component and any component on the component path sequence to obtain propagation path distance data;

[0123] A propagation delay estimation unit is used to calculate the action time of components based on the vibration propagation speed and propagation path distance data of different earthquake data;

[0124] The time mapping generation unit is used to match the action time of each component with the time information of each data segment in the continuous vibration data to obtain the vibration action time point of each component.

[0125] In an embodiment of the present invention, a propagation starting point identification unit is used to select, in each component path sequence, a component with the smallest component angle with the source direction vector as the propagation starting point component. Based on the principle of directional consistency between the component and the source, the key node that is first exposed to the vibration wave front can be screened out, and the starting point of the vibration propagation link can be established, thereby improving the time accuracy of the subsequent propagation path construction and the logical integrity of the response timing; the propagation distance calculation unit is used to accumulate and calculate the physical distance between the propagation starting point component and any component on the component path sequence to obtain the propagation path distance data. By accurately measuring the spatial distance between components on the vibration propagation path, the system can avoid the errors caused by traditional estimation based on the average propagation length, thereby improving the vibration The system can improve the resolution and practicality of vibration time distribution; the propagation delay estimation unit is used to calculate the action time of components according to the vibration propagation speed and propagation path distance data of different seismic data, convert the physical space distance into the vibration action time point, and realize the time distribution mapping of the vibration input, so that the system can more realistically simulate the temporal advancement process of the vibration wave and enhance the ability to characterize the cumulative effect of vibration; the time mapping generation unit is used to match the action time of each component with the time information of each data segment in the continuous vibration data, obtain the vibration action time point of each component, anchor the vibration response behavior of each component to a specific time point, and provide a temporal basis for the identification of high-response components, path impact tracing and superposition statistics of structural response values.

[0126] The propagation distance calculation unit is used to accumulate the physical distance between the propagation starting point component and any component on the component path sequence to obtain the propagation path distance data, specifically including:

[0127] The system first obtains a component path dataset, which records the spatial installation coordinates of all components in the target power facility and the connection sequence between components. Taking the propagation starting component as the reference point, the system calculates the Euclidean distance between the component and any other component on the path according to the connection sequence of the components in the path. The system then accumulates the distances of all component segments in the path to obtain the cumulative path distance from the propagation starting component to any target component. All calculation results form a component path distance dataset and are bound to the component index, serving as the input for the subsequent propagation delay estimation unit. This process ensures accurate representation of the facility's spatial structure in vibration propagation simulations, avoids propagation errors caused by average distance estimation, and improves the physical accuracy of the vibration response time series.

[0128] The propagation delay estimation unit is used to calculate the action time of the component based on the vibration propagation speed and propagation path distance data of different earthquake data, specifically including:

[0129] The system first receives the aforementioned component path distance dataset, where each component corresponds to a cumulative spatial propagation distance starting from the propagation starting point. The system presets or dynamically obtains the average propagation speed of seismic waves under current geological conditions. , such as shear wave velocity or surface wave velocity, which can be determined by geological exploration data or set according to the geological structure of the facility area. On this basis, the system estimates the propagation time for each component. ,in For components The vibration action time delay, From the start point to the component The path distance, is the vibration propagation velocity under corresponding directions and geological conditions. To adapt to complex earthquake scenarios, the system can also make directional corrections to the propagation velocity based on the angle between the source direction and the component path, or distinguish different velocity models based on different source numbers. The final output is a component action delay time set, which records the estimated time difference for each component to receive continuous vibration data and is used to drive subsequent time mapping operations. This unit ensures the timing rationality of the vibration propagation model, avoids the simplistic treatment of treating all vibration signals as instantaneous transmission, and improves the dynamic realism of the response simulation.

[0130] The time mapping generation unit is used to match the action time of each component with the time information of each data segment in the continuous vibration data to obtain the vibration action time point of each component, specifically including:

[0131] The time mapping generation unit is used to match the action time of each component with the time information of each data segment in the continuous vibration data to obtain the vibration action time point of each component. During its implementation, the system first reads the component action delay time set output by the propagation delay estimation unit and combines it with the continuous vibration data constructed by the previous combination module. The continuous vibration data contains the start time of each data segment. , end time, source number and other identification information. For each component, the system searches for the start time of the data segment corresponding to the source according to the source number of the path it is in. and delay the effect time Add to the starting time to get the vibration action time point of the component . Subsequently, the system establishes a set of mapping relationship tables to record the component index, source number, start time, delay time and final action time point, forming a complete component-vibration time matching mapping. In order to support the concurrent timing management of multiple sources, the mapping table supports the situation where the same component receives the action of multiple source data segments. The system will establish an independent time record entry for each source, and in subsequent modules, jointly determine the response impact based on the vibration intensity and action time. Through the time synchronization mechanism, this unit enables the component response behavior to have a physical time axis reference, providing core timing support for response value superposition, path tracking and high response identification.

[0132] In a preferred embodiment of the present invention, the identification module includes:

[0133] A vibration response value unit is used to calculate the path response value of each component in different vibration transmission paths according to the response path data, and summarize the values ​​to obtain the vibration response value of the component;

[0134] a response threshold comparison unit, configured to compare a vibration response value of a component with a preset response threshold, and mark the component as an abnormal response component when the vibration response value is greater than the preset response threshold;

[0135] The superposition enhancement identification unit is used to determine whether the vibration response value of the abnormal response component is jointly determined by two or more path response values. When the result is yes, the abnormal response component is a high response component.

[0136] In an embodiment of the present invention, a vibration response value unit is used to calculate the path response value of each component in different vibration transmission paths based on the response path data, and summarize the values ​​to obtain the vibration response value of the component, which can provide a more realistic and multi-dimensional measurement of the vibration response of the component under the influence of multiple earthquake sources and multiple paths; a response threshold comparison unit is used to compare the vibration response value of the component with a preset response threshold. When the vibration response value is greater than the preset response threshold, the component is marked as an abnormal response component, ensuring that the response evaluation of various components has a unified measurement standard and realizing rapid screening of local high-risk areas by identifying abnormal response components; a superposition enhancement identification unit is used to determine whether the vibration response value of the abnormal response component is jointly determined by two or more path response values. When the result is yes, the abnormal response component is a high-response component, effectively identifying those structural resonance amplification phenomena caused by the joint action of multiple paths although the single path response is not extreme.

[0137] The preset response threshold value is flexibly configured based on different geographical environments, facility types, and component characteristics. In mountainous long-distance transmission line scenarios, due to the dramatic terrain changes and complex vibration propagation directions, the system sets the response threshold of the tower base anchor structure to 1.2g. This value is higher than the standard value for plains. Considering the possible frequency coupling or resonance amplification effects in mountainous areas, the identification threshold is increased by approximately 20% based on the standard seismic design. For wind-resistant wire rope devices, since they are primarily affected by lateral vibrations and are prone to sudden changes in tension or even instability during horizontal vibration propagation, the system sets its threshold to 0.6g based on simulation tests and on-site vibration test data to ensure accurate identification of potentially high-response components even in low-frequency, horizontally propagating earthquakes.

[0138] In the application environment of substations in urban core areas, the system has enhanced its recognition sensitivity for precision electrical equipment. For example, for electrical switchgear components, the response threshold is set at 0.4g. This value is determined based on laboratory seismic test data for switchgear-type electrical equipment. Relay tripping and vibration dislocation of internal structural components have been observed within the 0.3g to 0.4g range, and the maximum stable operating boundary is selected as the recognition standard. For busbar supports, due to their relatively balanced stress distribution and stable structure, the threshold is set at 0.9g. This, combined with the yield strain limit and dynamic load capacity of the supporting material, ensures that a resilient response limit is still identifiable under moderate earthquakes and above.

[0139] In photovoltaic power plants located in coastal windblown sandy areas, crossbeam components often experience fatigue due to wind erosion and uneven foundations. The system sets the vibration response recognition threshold to 0.5g. This value is derived from fatigue testing of multiple post-earthquake structures in coastal environments, and is set at the upper limit of the safety range below the fatigue limit. For cable trays, vibration can easily cause line misalignment or loosening of metal frames. Therefore, the response threshold is set to 0.3g. This lower threshold is determined based on the equipment's vibration tolerance and operational experience to improve the ability to identify abnormalities in sensitive components.

[0140] In a preferred embodiment of the present invention, the vibration response value unit includes:

[0141] The path response value calculation unit is used to analyze the positive impact of the maximum vibration amplitude transmitted from the earthquake source to the component on the component based on the response path data and calculate the peak direction term; calculate the total value of the vibration amplitude transmitted from the earthquake source to the component, and correct the total vibration amplitude value by the delay factor formed by the action time to obtain the energy delay correction term; based on the peak direction term and the energy delay correction term, the comprehensive impact of different earthquake sources on the components along the vibration transmission path is calculated to obtain the path response value;

[0142] The vibration response value calculation unit is used to count the number of vibration transmission paths on which the component is located, and calculate the comprehensive impact on the component based on the path response values ​​of the component on different vibration transmission paths to obtain the vibration response value.

[0143] In an embodiment of the present invention, a path response value calculation unit is used to analyze the positive influence of the maximum vibration amplitude transmitted from the earthquake source to the component on the component according to the response path data, calculate the peak direction term, effectively reflect the physical law of non-uniform propagation of vibration energy in the structure, and improve the spatial accuracy of the path response simulation; calculate the total value of the vibration amplitude transmitted from the earthquake source to the component, and correct the total value of the vibration amplitude by the delay factor formed by the action time to obtain the energy delay correction term, effectively make up for the shortcoming of ignoring the time effect of the vibration input, and enhance the dynamic energy distribution simulation capability under the superposition effect of the distant earthquake source; according to the peak direction term and the energy delay correction term , count the comprehensive influence of different earthquake sources on the components on the vibration transmission path, and obtain the path response value, which comprehensively considers the combined effect of vibration direction, propagation distance and source amplitude, so that the response calculation of each path is more in line with the physical reality; the vibration response value calculation unit is used to count the number of vibration transmission paths in which the component is located, and calculate the comprehensive influence on the component according to the path response value of the component on different vibration transmission paths, and obtain the vibration response value, which fully reflects the vibration enhancement effect brought by multi-path coupling, and is particularly suitable for risk analysis of components with high intersection or structural node, and significantly improves the modeling ability of complex dynamics such as resonance amplification and vibration accumulation.

[0144] The calculation formula of the path response value is:

[0145] ,

[0146] in, For components In the vibration transmission path The path response value received in is the index of the component, is the index of the vibration transmission path, is the number of different earthquake sources, is the index of the earthquake source, For the earthquake source For components The maximum value of the propagated vibration amplitude, For the earthquake source In the Always check components The amplitude of the propagated vibration, For the earthquake source and components The angle between the components, For the earthquake source To component The action time, is the coefficient.

[0147] Among them, the coefficient and The weighted weights corresponding to the vibration direction influence term and the vibration energy accumulation term in the formula should always sum to 1 to ensure that the path response value forms a dynamic balance between direction drive and energy drive. The specific value should be flexibly set according to the structural characteristics of the target facility, the seismic environment in which it is located, and the characteristics of the earthquake source.

[0148] In mountainous areas, transmission lines cross multiple earthquake source areas, and tower structures are usually arranged linearly. Directionality plays a decisive role in the response. For example, when the direction of earthquake propagation is close to the tower path, structural resonance is more likely to occur. Set to 0.8, Set to 0.2 to make the model more sensitive to directional components in order to identify high-amplitude responses caused by seismic transmission.

[0149] In the urban distribution network scenario, the structural directions of the facility components are dispersed, and vibrations may act on different nodes at multiple angles. At the same time, the accumulation of vibration energy over time due to urban underground structures and surface reflections significantly enhances the component response. Set to 0.4, When it is set to 0.6, the model focuses more on the concentrated effect of the total vibration energy on the components, which is more conducive to identifying nodes that are most affected by the cumulative impact of aftershocks.

[0150] In a regional power grid under the influence of a wide-area earthquake, especially in a situation where multiple earthquake sources are distributed but the epicenter is far away, the vibration direction is highly random and the vibration is frequent and long-lasting. Set to 0.3, It is set to 0.7 to improve the model's ability to recognize overlapping excitations from multiple sources of energy and avoid path distortion caused by direction errors.

[0151] Among them, the coefficient Appears in the denominator of the energy integral term, and its function is to calculate the propagation time Energy suppression is performed with the increase of the vibration, reflecting the trend of vibration energy attenuating over time during long-distance propagation.

[0152] In areas with dense facilities and compact structures, such as large substations and thermal power plants, the distance between components is short and the time it takes for the earthquake source to reach the target component is short, so the vibration energy attenuation is not significant. It should be set to a lower value such as 0.1-0.3 to ensure that the contribution of energy integration in a short period of time will not be underestimated by the system, which is conducive to restoring the response of local strong earthquake instantaneous impact.

[0153] In scenarios where seismic waves pass through complex terrain or have long propagation paths, such as plateaus and hilly areas, the vibration transmission may span dozens of kilometers, and due to path blocking and reflection effects, the effectiveness of the vibration energy decays rapidly over time. It can be increased to 1.0-2.0 to enhance the system's ability to model energy attenuation caused by time and avoid unreasonable amplification of remote seismic energy on response values.

[0154] In situations where the superposition effect of consecutive aftershocks or earthquake clusters needs to be analyzed, if the goal is to identify whether there is a trend of local fatigue damage caused by the superposition of distant earthquake energy, An intermediate value of 0.5-0.8 can be appropriately selected to balance the evaluation deviation between the dominance of the main shock and the accumulation of aftershocks by controlling the intensity of the energy term on a long time scale.

[0155] The calculation formula of the vibration response value is:

[0156] ,

[0157] in, For components The vibration response value affected by different earthquake sources, is the index of the vibration transmission path, For components The number of vibration transmission paths, For components In the vibration transmission path The path response value received.

[0158] In a preferred embodiment of the present invention, the evaluation module includes:

[0159] A component state determination unit is used to determine whether the component is in a normal, critical or failed state based on the vibration response value of each component in the component response identification set, and obtain component state data;

[0160] A toughness weight allocation unit is used to allocate a toughness weight value to each component according to the structure of each component in the target facility to obtain component weight data;

[0161] The facility resilience index unit is used to calculate the resilience index of the target facility based on the component status data and component weight data, and obtain the target facility resilience assessment result.

[0162] In an embodiment of the present invention, a component state identification unit is used to determine whether the component is in a normal, critical or failed state according to the vibration response value of each component in the component response identification set, obtain component state data, and establish a component-level state judgment system, so that the system can sink to the smallest structural unit for judgment when evaluating the resilience of the entire facility; a resilience weight allocation unit is used to allocate a resilience weight value to each component according to the structure of each component in the target facility, obtain component weight data, introduce a structural importance dimension, so that the resilience evaluation not only reflects the response strength of the component, but also considers its functional status in the overall system; a facility resilience index unit is used to calculate the resilience index of the target facility according to the component state data and the component weight data, obtain the target facility resilience evaluation result, map the structural state to a numerical resilience coefficient, and combine the structural functional importance to provide the system with a quantifiable, comparable and traceable safety state reference.

[0163] The component state determination unit is used to determine whether the component is in a normal, critical, or failed state based on the vibration response value of each component in the component response identification set, and obtain component state data, specifically including:

[0164] First, the component response identification set is obtained from the identification module. The identification set includes the component unique identifier, the corresponding vibration response value, and the cumulative response information of the component under different paths. The system standardizes all response values ​​to unify the response intensity under different dimensions and sources, which is convenient for subsequent comparison and analysis. On this basis, the component state judgment unit presets two response critical values: critical threshold and failure threshold ,in , both can be determined based on structural design specifications, physical properties of component materials, simulation data or measured historical data. For example, for a certain type of conductor, its yield response can be set as the critical threshold, and its breaking response value can be set as the failure threshold. The system traverses the standardized response value of each component ,when When , the component is judged to be in "normal state"; when When , it is judged as "critical state"; if In order to facilitate the integration of numerical calculation and toughness formula, the system assigns component status scores to the three states respectively. : The normal state corresponds to a score of 1, indicating full availability; the critical state score is 0.5, indicating that the function is partially attenuated but not completely failed; the failure state score is 0, indicating that the component can no longer support the facility function. After the component status is determined, the system output includes the component number and its status score. The state data set can be directly input into the subsequent resilience assessment formula as one of the key factors.

[0165] The toughness weight allocation unit is used to allocate a toughness weight value to each component according to the structure of each component in the target facility to obtain component weight data, specifically including:

[0166] First, component-level information is obtained from the structural design data of power facilities. This includes component type (such as towers, conductors, insulators, connectors, etc.), the network topological location of the node (such as whether it is a critical relay node), the component's affiliation within the structural hierarchy (such as backbone structure or auxiliary component), and substitutability assessment (such as whether there is a functional bypass after failure). Based on this data, the system establishes a weighting function for each attribute type through a rule-based model or empirical factor calculation method, converting this structural information into a quantitative score. For example, components at critical connection nodes, with poor redundancy, and directly bearing load transfer can be assigned high weights in the topological importance factor, structural hierarchy factor, and substitutability factor; while supporting segments, terminal units, or components with alternative paths have correspondingly lower weights. The system ultimately normalizes and integrates all factors according to a set ratio, calculates the toughness weight value for each component, and outputs a table corresponding to the component identification and its weight. This process not only forms a quantitative expression of the functional importance of components, but also enables subsequent resilience assessment results to have discriminatory and structural identification capabilities. Especially in power systems with complex structures, multiple nodes and multiple paths, it can accurately reflect the systemic impact of certain key components, which helps to give priority to high-weight components in post-earthquake assessments or reinforcement strategies, and improve the targeted and scientific nature of system safety operations.

[0167] The facility resilience index unit is used to calculate the resilience index of the target facility based on the component status data and component weight data, and obtain the target facility resilience assessment result, which specifically includes:

[0168] The calculation formula of the resilience index has the characteristics of rigorous structure and clear physical meaning. Its core idea is to comprehensively model the multi-dimensional factors such as vibration response, time delay, and multi-source superposition of all components in the power facility, and output a normalized index reflecting the global structural resilience level in a weighted average manner. The formula as a whole consists of two parts: the numerator is the weighted sum of the resilience performance of each component, and the denominator is the total weight normalization factor. Specifically, the basic unit of the formula is "component level", with component level as the basis. is the index unit, for the total number of components The summation process reflects the idea that the resilience index is based on the construction of overall evaluation based on local response.

[0169] The toughness performance of a component is composed of three main factors: , Indicates the importance weight of the component in the overall structure, reflecting its functional status or criticality; is the status score (normal, critical, failure) determined by the component vibration response and is state sensitive; The time decay factor is introduced to indicate the timeliness of vibration effects. That is, the earlier the component is affected by vibration, the more significant the energy transfer effect. Together, these factors form the outer evaluation structure of the "unit component residual toughness capacity."

[0170] What is more innovative is the way the internal sub-items are constructed, namely This term represents a composite expression of the intensity, directionality, and multi-source coupling effects experienced by the component in the vibration path. Specifically, It is the directional peak term in the path, which is used to describe the excitation directionality of the seismic wave transmitted to the component along the path and is highly related to the geometric arrangement of the structure; is the path response value, reflecting the energy intensity of the path, and the multiplication of the two can be understood as the effective directional excitation received by the component in the path. , the proportion of superimposed response borne by the component under the multi-source situation is introduced. If the response of a component is enhanced by the combined action of multiple seismic sources, the value of this item will increase, thereby effectively reflecting the local high response phenomenon caused by complex vibration coupling.

[0171] In summary, this formula comprehensively expresses the toughness bearing capacity of each component under multi-source vibration through the five-level coupling modeling mechanism of "importance × state × time decay × path response × source superposition", and normalizes it to the overall system index Because of the clear physical connections and spatiotemporal dependencies between the data, the final calculated resilience index is not only quantitatively comparable but also explanatory and operational. It can be used in engineering to quickly assess post-earthquake conditions, identify weaknesses, and support structural reinforcement decisions.

[0172] The calculation formula of the toughness index is:

[0173] ,

[0174] in, is the value of the resilience index of the target facility, is the index of the component of the target facility, is the total number of components of the target facility, For components The toughness weight value, For components The component status score, For the earthquake source To component The action time, For components In the vibration transmission path The peak direction term in , For components In the vibration transmission path The path response value received in is the number of different earthquake sources, is the index of the earthquake source, For components At the epicenter The percentage of responses received, is the coefficient.

[0175] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A power facility resilience assessment system based on multi-source seismic data, characterized by: The system comprises: A screening module is used to obtain and screen earthquake data sets, eliminating earthquake data whose epicenters are not related to the target facility, and obtaining an earthquake screening data set; A segmentation module is used to extract the inflection points of the vibration amplitude change based on the earthquake screening data set to obtain vibration inflection point data, and to divide the earthquake screening data set into multiple data segments based on the inflection point data to obtain segmented data sets; The combination module is used to splice the data segments in the segmented data set in their time sequence and retain the start and end position labels of each data segment in the overall sequence to obtain continuous vibration data; The action module is used to match the continuous vibration data with the components of the target facility according to the spatial position and connection relationship of each component in the target facility, establish the vibration transmission path according to the propagation direction and action time of the spliced ​​vibration data, and obtain the response path data; An identification module is used to calculate the vibration response value of each component in the target facility based on the response path data, identify high-response components with multi-source superposition enhancement characteristics, and obtain a component response identification set; An evaluation module is used to determine the component number and vibration response value of each component based on the component response identification set, calculate the resilience index of the target facility, and obtain the resilience evaluation result of the target facility; The action module includes: A vibration data mapping unit is used to bind the time information of each data segment in the continuous vibration data to the source direction vector of its corresponding source number to form a direction-time comparison table; A component sequence generation unit is used to construct a component path sequence according to the installation position and connection relationship of each component of the target facility in space, and obtain a component path data set; The action time unit is used to calculate the action time of each component receiving continuous vibration data based on the direction time comparison table and the component path data set; The component path construction unit is used to establish a vibration transmission path with a time tag according to the sequence of the action time of each component and obtain the response path data.

2. The power facility resilience assessment system based on multi-source seismic data according to claim 1, characterized in that: The screening modules include: A source direction analysis unit is used to calculate the spatial coordinate difference between the epicenter position and the position of the target facility component according to the epicenter position corresponding to each earthquake data in the earthquake data set, and obtain the source direction vector; A component orientation extraction unit is used to obtain the structural orientation vector of each component in the target facility; A direction angle calculation unit is used to calculate the component angle between the source direction vector of each seismic data and the structural orientation vector of each component to obtain a component angle set; The direction screening judgment unit is used to judge whether the seismic data has direction correlation with the target facility based on the component angle set and the preset angle threshold. When the component angle is greater than the preset angle threshold, the seismic data is eliminated to obtain the seismic screening data set.

3. The power facility resilience assessment system based on multi-source seismic data according to claim 2, characterized in that: The segmentation module includes: An amplitude sequence construction unit is used to extract the vibration amplitude of each earthquake data according to the earthquake screening data set to obtain a vibration amplitude data set; An extreme point identification unit is used to identify local maximum points and local minimum points based on the vibration amplitude data set to obtain an extreme point sequence; The fluctuation change rate calculation unit is used to calculate the amplitude difference and extreme value time difference between adjacent extreme value points based on the extreme value point sequence to obtain the amplitude change rate sequence; The inflection point determination unit is used to identify the time point with significant amplitude change rate as the inflection point of vibration amplitude change according to the amplitude change rate sequence, and obtain vibration inflection point data.

4. The power facility resilience assessment system based on multi-source seismic data according to claim 3, characterized in that: The inflection point determination unit includes: a rate of change difference calculation unit, configured to calculate the amplitude rate of change difference between each extreme point and the previous extreme point according to the amplitude rate of change sequence, to obtain a rate of change difference sequence; A threshold comparison unit is used to compare the values ​​of the rate of change difference sequence with a preset difference threshold value. When the amplitude change rate difference exceeds the preset difference threshold value, the extreme point is marked as a candidate inflection point to obtain candidate inflection point data. The minimum time distance verification unit is used to calculate the time interval between adjacent candidate inflection points based on the candidate inflection point data to obtain the inflection point time difference, and to filter out candidate inflection points whose inflection point time difference is less than a preset time difference threshold based on the inflection point time difference; The continuous segment mean comparison unit is used to calculate the mean difference of the vibration amplitude of each fixed-length segment before and after the dividing point with each candidate inflection point as the dividing point, and compare the mean difference with the preset mean difference threshold. When the mean difference is greater than the preset mean difference threshold, the candidate inflection point is retained to obtain the vibration inflection point data.

5. The power facility resilience assessment system based on multi-source seismic data according to claim 4, characterized in that: The combined modules include: A time sequence sorting unit is used to sort the segmented data sets in time sequence according to the earthquake time corresponding to each data segment, and generate a sorting index sequence; A data segment splicing unit is used to sequentially splice the data segments according to the sorting index sequence to generate a continuous vibration sequence; The segment boundary position marking unit is used to record the start and end positions of each data segment in the continuous vibration sequence and generate a segment boundary position information set; The time tag adding unit is used to add the start and end position tags of each data segment boundary in the continuous vibration sequence according to the segment boundary position information set, and mark the corresponding source number to obtain continuous vibration data.

6. The power facility resilience assessment system based on multi-source seismic data according to claim 5, characterized in that: Action time units include: A propagation starting point identification unit is used to select, in each component path sequence, the component with the smallest component angle with the earthquake source direction vector as the propagation starting point component; A propagation distance calculation unit is used to accumulate and calculate the physical distance between the propagation starting point component and any component on the component path sequence to obtain propagation path distance data; A propagation delay estimation unit is used to calculate the action time of components based on the vibration propagation speed and propagation path distance data of different earthquake data; The time mapping generation unit is used to match the action time of each component with the time information of each data segment in the continuous vibration data to obtain the vibration action time point of each component.

7. The power facility resilience assessment system based on multi-source seismic data according to claim 6, characterized in that: The recognition module includes: A vibration response value unit is used to calculate the path response value of each component in different vibration transmission paths according to the response path data, and summarize the values ​​to obtain the vibration response value of the component; a response threshold comparison unit, configured to compare a vibration response value of a component with a preset response threshold, and mark the component as an abnormal response component when the vibration response value is greater than the preset response threshold; The superposition enhancement identification unit is used to determine whether the vibration response value of the abnormal response component is jointly determined by two or more path response values. When the result is yes, the abnormal response component is a high response component.

8. The power facility resilience assessment system based on multi-source seismic data according to claim 7, characterized in that: The vibration response value units include: The path response value calculation unit is used to analyze the positive impact of the maximum vibration amplitude transmitted from the earthquake source to the component on the component based on the response path data and calculate the peak direction term; calculate the total value of the vibration amplitude transmitted from the earthquake source to the component, and correct the total vibration amplitude value by the delay factor formed by the action time to obtain the energy delay correction term; based on the peak direction term and the energy delay correction term, the comprehensive impact of different earthquake sources on the components along the vibration transmission path is calculated to obtain the path response value; The vibration response value calculation unit is used to count the number of vibration transmission paths on which the component is located, and calculate the comprehensive impact on the component based on the path response values ​​of the component on different vibration transmission paths to obtain the vibration response value.

9. The power facility resilience assessment system based on multi-source seismic data according to claim 8, characterized in that: Assessment modules include: A component state determination unit is used to determine whether the component is in a normal, critical or failed state based on the vibration response value of each component in the component response identification set, and obtain component state data; A toughness weight allocation unit is used to allocate a toughness weight value to each component according to the structure of each component in the target facility to obtain component weight data; The facility resilience index unit is used to calculate the resilience index of the target facility based on the component status data and component weight data, and obtain the target facility resilience assessment result.

Images