A dynamic grading early warning system for safety risks of a grounding grid

Abstract

This invention relates to the field of dynamic hierarchical early warning technology for grounding grids, specifically disclosing a dynamic hierarchical early warning system for grounding grid safety risks. This invention uses a raw data packet generation module to periodically collect the electrical response and environmental parameters of the grounding grid; a health status spectrum generation module to remove environmental interference and obtain an electrical health spectrum reflecting the conductor's state; a degradation trend analysis module to identify continuous degradation trends; a structural disturbance monitoring module to capture physical external force events through distributed fiber optic sensing; a spatiotemporal event correlation module to construct causal relationship markers; a composite risk diagnosis module to form a risk profile; and finally, a hierarchical early warning release module to release hierarchical dynamic risk warnings based on risk judgment rules. This achieves closed-loop management from data monitoring to operation and maintenance actions, transforming traditional passive emergency repairs into proactive intervention and preventative maintenance, optimizing resource allocation, and ensuring the safe and stable operation of the power grid.

Description

Technical Field

[0001] This invention relates to the field of dynamic hierarchical early warning technology for grounding grids, and more specifically, to a dynamic hierarchical early warning system for grounding grid safety risks. Background Technology

[0002] As a critical infrastructure ensuring the safe and stable operation of power systems and the safety of personnel, the real-time monitoring of the grounding grid's health status faces severe challenges. Because grounding conductors are buried deep underground, their operating environment is complex and variable, subjecting them to chronic degradation caused by soil corrosion and electrochemical reactions, as well as sudden damage caused by external physical forces such as third-party construction and geological subsidence. Accurately distinguishing between measurement data fluctuations caused by environmental changes and actual defects in the grounding grid itself, and simultaneously and effectively identifying these two distinct risks—chronic degradation and sudden damage—is a core issue that urgently needs to be addressed in the field of grounding grid safety operation and maintenance.

[0003] Currently, the most common grounding grid testing methods in the industry rely on periodic offline measurements. Maintenance personnel typically travel to the substation site according to annual or longer maintenance plans, disconnecting the grounding grid from the equipment and using specialized equipment such as grounding resistance testers to perform measurements. This method allows obtaining the overall grounding resistance value of the grounding grid at a specific point in time, serving as a basis for assessing its basic performance. For localized inspections, sometimes partial excavation is conducted to visually inspect the corrosion of the grounding conductor or the integrity of connection points; however, this is not a routine procedure and is only implemented when serious problems are suspected.

[0004] Traditional testing methods have significant drawbacks. Firstly, they lack early warning capabilities. The excessively long intervals between periodic tests make it difficult to detect sudden physical damage or accelerated deterioration occurring between measurements, resulting in a significant delay in identifying safety risks. Secondly, they lack diagnostic accuracy. Single resistance measurements are easily affected by environmental factors such as soil moisture and temperature, leading to distorted results. Maintenance personnel struggle to determine whether the issue stems from a genuine equipment defect or normal fluctuations caused by weather changes, easily resulting in misdiagnosis or missed diagnosis. Furthermore, traditional methods cannot pinpoint or characterize risks. Even if a resistance value is found to be substandard, it's difficult to determine the specific fault location and cause, leaving subsequent maintenance work time-consuming and labor-intensive due to a lack of clear guidance. Summary of the Invention

[0005] In view of this, in order to solve the problems mentioned in the background technology, a dynamic hierarchical early warning system for grounding grid safety risks is proposed.

[0006] The objective of this invention can be achieved through the following technical solution: This invention provides a dynamic hierarchical early warning system for grounding grid safety risks, comprising: a raw data packet generation module, which generates a time-aligned raw data packet containing grounding grid electrical response information and synchronization environmental parameters according to a preset detection period.

[0007] The health status spectrum generation module, based on time-aligned raw data packets, generates an electrical body health spectrum that reflects the state of the grounding grid conductor itself by separating environmental influences.

[0008] The degradation trend analysis module takes the electrical body health spectrum as input and identifies and quantifies a continuous degradation trend signal by comparing it with historical health status.

[0009] The structural disturbance monitoring module uses distributed optical fiber sensing cables laid along the grounding grid conductor to monitor and generate distributed structural disturbance data containing information on physical external force events.

[0010] The spatiotemporal event association module takes into account distributed structural disturbance data and associates it with the electrical entity health spectrum to construct a spatiotemporal event association marker for determining the causal relationship between physical events and electrical faults.

[0011] The composite risk diagnosis module takes into account spatiotemporal event correlation markers and combines them with continuous deterioration trend signals to diagnose and generate a composite risk causal profile.

[0012] The tiered early warning release module allows users to input a composite risk causal profile and release tiered dynamic risk warnings based on preset risk assessment rules.

[0013] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) By constructing a dynamic interference model, the present invention can effectively remove the temporary influence of environmental factors such as soil moisture and temperature on electrical measurement data, and generate an electrical health spectrum that reflects the physical state of the grounding grid conductor itself. This enables the early warning system to penetrate the fog of environmental changes and accurately identify the real performance degradation caused by conductor corrosion, connection degradation, etc., fundamentally improving the accuracy and reliability of risk identification, avoiding false alarms caused by environmental fluctuations, and making operation and maintenance decisions based on a purer and more realistic data foundation.

[0014] (2) This invention innovatively integrates two monitoring technologies, electrical response analysis and distributed fiber optic sensing, to construct a multi-dimensional risk perception system. It can not only capture slow, continuous deterioration trends caused by internal factors by analyzing the time series of the electrical system's health spectrum, but also capture sudden external structural disturbances by monitoring physical external force events in real time. This dual monitoring mechanism enables the system to comprehensively cover the two main risk types of the grounding grid, achieving simultaneous monitoring of chronic lesions and acute injuries.

[0015] (3) This invention achieves accurate diagnosis and tracing of risk causes by establishing spatiotemporal event correlation analysis. When physical disturbance events and drastic changes in electrical characteristics are highly consistent in time and space, the system can automatically determine their causal relationship and clarify that the risk is caused by external physical damage. This solves the problem that traditional methods cannot distinguish the cause of the fault, making the early warning information no longer a single alarm signal, but a diagnostic report containing a clear profile of the cause, providing direct evidence for subsequent fault investigation and responsibility determination.

[0016] (4) This invention ultimately enables the issuance of tiered, dynamic risk warnings with clearly defined action guidelines, achieving closed-loop management from data monitoring to operation and maintenance actions. Based on the diagnosed risk causes, the system automatically matches different warning levels and response strategies, providing operation and maintenance personnel with clear, specific, and actionable recommendations. This greatly improves the efficiency of emergency response and the targeted nature of maintenance work, transforming traditional passive repairs into proactive intervention and preventative maintenance, optimizing resource allocation, and ensuring the safe and stable operation of the power grid. Attached Figure Description

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

[0018] Figure 1 This is a schematic diagram of the system module structure connection of the present invention.

[0019] Figure 2 This is a flowchart illustrating the tiered early warning release decision-making process of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 This invention provides a dynamic hierarchical early warning system for grounding grid safety risks, comprising: a raw data packet generation module, a health status spectrum generation module, a degradation trend analysis module, a structural disturbance monitoring module, a spatiotemporal event correlation module, a composite risk diagnosis module, a hierarchical early warning release module, and a database.

[0022] It should be noted that the present invention also includes a database for storing the range of mutation magnitudes corresponding to the intensity of each event.

[0023] The original data packet generation module is connected to the health status spectrum generation module, the health status spectrum generation module is connected to the degradation trend analysis module, the health status spectrum generation module and the structural disturbance monitoring module are both connected to the spatiotemporal event correlation module, the spatiotemporal event correlation module and the degradation trend analysis module are both connected to the composite risk diagnosis module, the composite risk diagnosis module is connected to the graded early warning release module, and the structural disturbance monitoring module is connected to the database.

[0024] The raw data packet generation module generates a time-aligned raw data packet containing grounding grid electrical response information and synchronization environment parameters according to a preset detection period.

[0025] In one specific embodiment of the present invention, the preset detection cycle refers to the time interval at which the data acquisition task is automatically and repeatedly executed. Its function is to ensure periodic and continuous monitoring of the grounding grid status. The cycle is set based on a balance between the typical rate of change in the grounding grid status and the monitoring cost. Based on long-term observation of the grounding grids of multiple substations, setting it to 30 minutes achieves a good balance between timely detection of anomalies and avoidance of data redundancy.

[0026] In a specific embodiment of the present invention, the specific steps for generating the time-aligned original data packet containing grounding grid electrical response information and synchronization environment parameters include: injecting a set of microcurrent pulses into the grounding grid and synchronously measuring the multi-frequency composite impedance response spectrum at key nodes of the grounding grid.

[0027] It should be noted that at the start of a preset detection cycle, a set of specially designed micro-current pulses is injected into the grounding grid via a signaling device. This set of micro-current pulses is composed of multiple signals of different frequencies, covering a frequency range from low to high, ensuring comprehensive detection of the grounding grid's characteristics across different electromagnetic response intervals. The current intensity is controlled at the milliampere level to ensure that the measurement is completed without interfering with the normal operation of the power system. The injection point is typically selected at the main grounding electrode of the grounding grid or the centralized grounding terminal of the substation. Simultaneously with the injection of the micro-current pulses, voltage acquisition modules deployed at various key nodes of the grounding grid are activated, measuring and recording the voltage response amplitude and phase corresponding to each frequency component. These key nodes are pre-selected based on the grounding grid design drawings and historical fault points, such as branch points of the main line and connection points of grounding leads for important equipment. After completing the measurements of all frequencies, the resulting data set constitutes a multi-frequency composite impedance response spectrum.

[0028] It should also be noted that the core of this step is to calculate the impedance, which requires the use of Ohm's law in complex form, as shown in the following formula: ,in, Represents the grounding grid at a specific frequency The complex impedance is expressed in ohms. This represents the frequency measured at the critical node. The voltage response, a complex quantity including amplitude and phase, is measured in volts and obtained through a high-precision voltage acquisition module. Represents the frequency injected by the signal device. The current is also a complex quantity containing amplitude and phase, and its unit is ampere. This value can be known in advance according to the setting parameters of the signal device, for example, it is set to 10 milliamperes. The frequency representing the injected current and measured voltage is measured in Hertz. Its range is based on the response characteristics of the grounding grid to currents of different frequencies. Low frequencies mainly reflect the DC resistance of the conductor, while high frequencies reflect the dielectric properties of the soil and the coupling effect between conductors. Therefore, a sweep frequency range from 10 Hertz to 100 kilohertz is set.

[0029] Using environmental sensors, soil moisture, temperature, and pH data are collected in real time, synchronized with the measurement time of the multi-frequency composite impedance response spectrum, to form synchronized environmental parameters.

[0030] It should be noted that, simultaneously, environmental sensors are pre-embedded in the soil adjacent to each key voltage measurement node. These environmental sensors are triggered at the same instant the voltage measurement is initiated, collecting and uploading real-time data on soil moisture, temperature, and pH. This set of environmental data is called synchronous environmental parameters.

[0031] The multi-frequency composite impedance response spectrum is bound to the synchronization environment parameters with a unified timestamp to generate time-aligned raw data packets.

[0032] It should be noted that the multi-frequency composite impedance response spectrum and synchronization environmental parameters obtained at the same acquisition time are finally packaged and marked with a unified timestamp generated by a high-precision clock system. This timestamp is accurate to the millisecond, ensuring absolute time alignment between the electrical response information and the environmental parameters. This dataset with a unified timestamp is ultimately stored as a structured, time-aligned raw data packet, awaiting further processing.

[0033] The health status spectrum generation module generates an electrical body health spectrum that reflects the state of the grounding grid conductor itself, based on the time-aligned raw data packets and by separating environmental influences.

[0034] In a specific embodiment of the present invention, the specific steps for generating an electrical health spectrum reflecting the state of the grounding grid conductor itself include: extracting high-frequency response data that is sensitive to environmental changes from the multi-frequency composite impedance response spectrum.

[0035] It should be noted that data points above a specific frequency range are selected from the multi-frequency composite impedance response spectrum contained in the input time-aligned raw data packet. These data points constitute the high-frequency response data. The reason for choosing the high-frequency band is that when high-frequency current propagates in soil, its path and loss are more affected by the soil dielectric constant, which is directly related to soil moisture content. Therefore, high-frequency impedance exhibits the highest sensitivity to environmental changes.

[0036] It should also be noted that high-frequency response data is a subset of the multi-frequency composite impedance response spectrum, specifically referring to impedance data with frequencies above a certain threshold. Its function is to serve as input for training dynamic interference models, as this portion of data is most sensitive to environmental changes. This threshold can be set to 10 kHz, and its setting is based on statistical analysis of measured grounding grid data under several sets of different soil conditions, which found that in the frequency band above 10 kHz, the correlation coefficient between impedance values ​​and soil moisture exceeds 0.8.

[0037] A dynamic interference model is established based on the historical correspondence between high-frequency response data and synchronization environment parameters.

[0038] In a specific embodiment of the present invention, the method for establishing a dynamic interference model is as follows: using a large number of synchronization environment parameters from historical data as input variables and corresponding high-frequency band response data as output variables, a dynamic interference model that quantifies the relationship between changes in environmental parameters and changes in high-frequency impedance is established through training with a multivariate regression algorithm.

[0039] It should be noted that the system then utilizes a pre-established dynamic interference model to process the data. This model is established by analyzing a large amount of historical time-aligned raw data packets accumulated over a long period of time in a healthy state of the grounding grid. The analysis process employs a multiple regression algorithm, a method for finding the optimal mathematical relationship between multiple input variables and one output variable. It takes synchronous environmental parameters from historical data, namely soil moisture, temperature, and pH, as inputs and the corresponding high-frequency response data as outputs, thereby learning and solidifying the quantitative correspondence between changes in environmental parameters and changes in high-frequency impedance.

[0040] It should also be noted that the expression for the dynamic interference model is: , Represented by environmental factors in frequency The impedance change caused at the point is expressed in ohms, which is the output of the dynamic disturbance model. This represents a nonlinear functional relationship, namely the dynamic disturbance model itself, which is obtained through machine learning training on historical data; These represent soil moisture, temperature, and pH, which are the input environmental parameters and are the model's input variables. The frequency is represented by Hertz, and the model needs to be able to calculate the corresponding interference for different frequencies.

[0041] It should be further explained that the dynamic interference model is obtained through machine learning training on historical data in the following specific ways: 1) Model training: Training is performed using a large number of samples from historical data, each sample containing a set of synchronization environment parameters and corresponding high-frequency response data; the model parameters, i.e., the function, are adjusted through a multivariate regression algorithm. The coefficients in the model minimize the error between the predicted impedance change and the actual measured high-frequency response data. During training, the algorithm iteratively optimizes until the optimal model parameters are found, enabling the model to accurately quantify the relationship between environmental parameter changes and high-frequency impedance changes. 2) Model Validation and Evaluation: The trained model is validated using an independent validation dataset to evaluate its prediction accuracy and generalization ability. If the model performs well on the validation dataset, the model is considered successfully trained and can be used for subsequent real-time data processing.

[0042] By applying a dynamic disturbance model, the response changes caused by synchronization environment parameters are calculated and extracted from the multi-frequency composite impedance response spectrum to generate the electrical body health spectrum.

[0043] It should be noted that when processing the current data, the system inputs the synchronization environment parameters from the original time-aligned data packets into the established dynamic interference model. The model then calculates the response change at each frequency point in the entire multi-frequency composite impedance response spectrum caused by the current environmental conditions. Finally, for each impedance measurement value in the original multi-frequency composite impedance response spectrum, the corresponding frequency response change calculated by the model is subtracted. This difference calculation effectively removes the disguise of environmental factors. The new impedance spectrum obtained after this series of processing is the electrical integrity health spectrum. It no longer fluctuates drastically with short-term rain or drought, but more stably reflects the permanent physical state of the grounding grid's metal conductors, such as corrosion, breakage, or loose connections.

[0044] It should also be noted that the calculation process for generating the electrical body health spectrum is as follows: ,in, Represents frequency The impedance value of the electrical body health spectrum at the location, in ohms, is the final output of this step; Represents the frequency obtained from the time-aligned raw data packets. The impedance value of the original multi-frequency composite impedance response spectrum at the location is expressed in ohms.

[0045] It should be further explained that the electrical health spectrum is a processed impedance spectrum, which serves as a purity indicator for assessing the physical state of the grounding grid conductors themselves, eliminating interference from environmental changes. Its data structure is the same as that of the multi-frequency composite impedance response spectrum, consisting of a series of complex impedance values ​​arranged by frequency.

[0046] The degradation trend analysis module takes the electrical body health spectrum as input and identifies and quantifies a continuous degradation trend signal by comparing it with historical health status.

[0047] In a specific embodiment of the present invention, the specific steps of identifying and quantifying a continuous deterioration trend signal by comparing with historical health status include: comparing the latest electrical body health spectrum with a preset health baseline spectrum and calculating the difference.

[0048] It should be noted that, firstly, a pre-stored health baseline spectrum is retrieved. This health baseline spectrum is the electrical health spectrum of the grounding grid measured during initial commissioning or after the most recent major overhaul, confirming that it is in optimal health condition. It represents the ideal electrical fingerprint of the grounding grid. Subsequently, the system performs a point-by-point alignment comparison between the latest electrical health spectrum and the health baseline spectrum. That is, it subtracts the impedance value of the corresponding frequency in the baseline spectrum from the impedance value of each frequency in the current spectrum, thereby calculating a set of differences. This set of differences reflects the degree of deviation of the current state from the ideal state. This calculation process is repeated after each preset detection cycle, thus forming a time-series database composed of differences at continuous time points.

[0049] It should also be noted that the formula for calculating the difference is: ,in, Represents the moment ,frequency The difference at each point, in ohms; Represents the moment ,frequency The electrical body health spectrum impedance value at the location, in ohms; Represents frequency The healthy baseline spectral impedance value at the location, in ohms.

[0050] Trend analysis is performed on the differences in continuous time series to identify the deteriorating component that continues to grow in a unidirectional manner.

[0051] It should be noted that the system will next perform trend analysis on this difference time series. The core of trend analysis is to examine whether there is a continuous, unidirectional increase in the difference over a relatively long time window, such as the past three months. This analysis can be achieved by applying statistical methods such as linear regression to the time series data, i.e., fitting a straight line that can represent the long-term direction of data change. If the slope of the fitted line is consistently positive and exceeds a small threshold, it indicates the existence of a degradation component that cannot be recovered as the environment improves. This degradation component directly points to physical, cumulative damage to the grounding grid's metal conductor itself.

[0052] It should also be noted that the formula for expressing the degradation component is: ,in, Represents frequency The degradation component identified is physically represented as the rate of change of impedance over time, measured in ohms per day. This value is obtained by calculating the slope of the difference time series. It is a trend analysis function, specifically the slope of time series data can be calculated using least squares linear regression. Represents the current moment and before A continuous time series composed of the differences at each acquisition time. The setting is based on the need for a sufficiently long time to observe trends, and is usually set to cover 90 days of data points.

[0053] The degradation component is quantified into a continuous degradation trend signal.

[0054] It should be noted that after the system identifies the degradation component, it needs to be quantified into a continuous degradation trend signal, that is, converted into a specific value for continuous monitoring. The quantification can be achieved by directly using the slope of a trend-fitting straight line. This slope represents the degradation rate of the grounding grid conductor, representing the total degree of degradation, and its magnitude directly reflects the speed of degradation. This signal, as an independent indicator, is continuously updated and monitored.

[0055] The structural disturbance monitoring module uses distributed optical fiber sensing cables laid along the grounding grid conductor to monitor and generate distributed structural disturbance data containing information on physical external force events.

[0056] In a specific embodiment of the present invention, the specific steps of monitoring and generating distributed structural disturbance data containing information on physical external force events include: continuously monitoring changes in optical signals caused by stress or vibration on distributed optical fiber sensing cables using optical time-domain reflectometry.

[0057] It should be noted that the goal of this step is to utilize a special sensing technology to detect and record external events that pose a physical threat to the grounding grid in real time. In implementation, a distributed fiber optic sensing cable is laid along the entire path of the underground grounding grid conductor, ensuring it is closely integrated with the conductor or located within the same trench. An optical signal analysis device is connected to one end of the cable. This device employs optical time-domain reflectometry (OTDR), operating similarly to radar but using optical pulses. The device continuously emits laser pulses into the distributed fiber optic sensing cable. As the light pulses propagate along the fiber, the inherent inhomogeneity of the fiber's microstructure generates weak backscattered Rayleigh light, which travels back along the fiber to the analysis device. Under normal conditions, the returned scattered light signal curve is smooth and stable. When an external physical force event, such as excavation, heavy impact, or geological subsidence, occurs at a point along the grounding grid, the resulting stress or vibration is transmitted to the adjacent distributed fiber optic sensing cable, causing a minor physical deformation of the fiber at that location. This deformation changes the refractive index and scattering characteristics of the optical fiber at that point, thereby causing a change in the intensity of the scattered light signal returned from that point, i.e., a change in the optical signal.

[0058] When a sudden change in the light signal exceeding the normal threshold is detected, the occurrence time, precise physical location, and intensity of the sudden event are determined.

[0059] It should be noted that the analysis equipment achieves online monitoring of the physical status of the entire grounding grid path by continuously monitoring the returned signals. The system has an internally set normal threshold to distinguish between minor vibrations in the daily environment and potentially threatening severe disturbances. Once a sudden change in the returned optical signal exceeding the normal threshold is detected within a very short time, the system immediately identifies it as a valid physical external force event. After identifying the event, the system immediately extracts three key pieces of information. First, based on the time taken for the optical pulse to travel from emission to reception of the abrupt change signal, combined with the speed of light propagation in the optical fiber, the precise physical location of the abrupt change event can be calculated. Second, the system records the server time when the abrupt change was triggered as the event occurrence time. Finally, based on the magnitude of the optical signal abrupt change, the intensity of this physical external force event can be quantified.

[0060] It should also be noted that the intensity of this physical external force event can be quantified as follows: First, under normal conditions without physical external force interference, the optical signal is measured multiple times and the average value is taken to determine the reference amplitude of the optical signal under normal conditions. When the physical external force causes a sudden change in the optical signal, a high-precision measuring device is used to accurately measure the amplitude of the optical signal after the change. Considering the possibility of errors during the measurement process, multiple measurements can be performed and the average value taken to reduce the impact of errors. Then, the difference between the average amplitude of the optical signal after the change and the reference amplitude under normal conditions is calculated, and this difference is recorded as the change amplitude. Next, the calculated change amplitude is compared with the change amplitude intervals corresponding to each event intensity stored in the database. If the change amplitude uniquely falls within the change amplitude interval corresponding to a certain event intensity, then that event intensity is taken as the event intensity of this physical external force event; if the change amplitude is on the boundary between two event intensity intervals, the larger event intensity is taken to determine the event intensity of this physical external force event.

[0061] It should also be noted that a formula is needed to determine the precise physical location of the mutation event. The specific formula is as follows: ,in, This represents the precise physical location of the calculated mutation event, i.e., the distance from the disturbance point to the optical signal analysis device, in meters; It represents the speed of light in a vacuum and is a physical constant with a value of 299,792,458 meters per second. This represents the total time it takes for a light pulse to travel from its emission point to the disturbance point and then reflect back to the analysis device for reception. This value is measured by the high-precision timer of the analysis device. The value represents the refractive index of the fiber core of the distributed fiber optic sensing cable. This value is provided by the fiber optic manufacturer and is usually around 1.46. It is set according to the specifications of the fiber optic product. The "2" in the formula represents the round trip of the optical signal, so it needs to be divided by 2 to get the one-way distance.

[0062] In one specific embodiment of the present invention, the normal threshold is a pre-set upper limit of the amplitude of optical signal variation. Its function is to filter out environmental background noise and ensure that only significant disturbances are identified as events. The threshold is set based on 24 hours of continuous observation of the system under interference-free conditions, and 1.5 times the maximum fluctuation value of the signal is taken as the threshold.

[0063] The occurrence time, precise physical location, and event intensity are encapsulated as distributed structural perturbation data.

[0064] It should be noted that these three pieces of information—time of occurrence, precise physical location, and event intensity—are finally organized into a data record in a standard format. This record is defined as distributed structure perturbation data and stored for subsequent analysis.

[0065] The spatiotemporal event association module takes into account distributed structural disturbance data and associates it with the electrical body health spectrum to construct a spatiotemporal event association marker for determining the causal relationship between physical events and electrical faults.

[0066] In a specific embodiment of the present invention, the specific steps of constructing a spatiotemporal event association marker for determining the causal relationship between physical events and electrical faults include: using a preset geographic information mapping table to convert the precise physical location in the distributed structural disturbance data into an electrical node adjacent to it in the grounding grid topology.

[0067] The search examines whether, within the time frame of the occurrence of the distributed structural disturbance data marker and within a very short time window before and after it, the electrical entity health spectrum exhibits drastic jumps at electrical nodes at the corresponding precise physical locations.

[0068] If a physical disturbance event and a drastic change in the electrical health spectrum are consistent in time and space, then a strong causal relationship between the two is determined.

[0069] Generate a spatiotemporal event association tag that includes a strong causal relationship conclusion, event location, and time.

[0070] It should be noted that the system first extracts the event's occurrence time and precise physical location from the data. Then, the system defines a very short time window around this occurrence time, such as 5 seconds before and 5 seconds after the event. This window focuses on finding electrical responses that are closely related to the physical event in time. Simultaneously, the system uses a pre-configured geographic information mapping table to convert the precise physical location marked in the distributed structural disturbance data, such as "3200 meters along the optical cable," into the nearest electrical node in the grounding network topology, such as "critical node number 5." After completing the temporal and spatial positioning, the system retrieves all electrical health spectrum records related to this electrical node within this time window. Next, the system analyzes the retrieved electrical health spectrum sequence, primarily checking for abrupt changes. Specifically, it compares two consecutive electrical health spectra immediately before and after the event's occurrence time, calculating their impedance differences at various frequency points. If the difference at any frequency point exceeds a preset abrupt change threshold, such as an impedance value changing by more than 20% within a measurement cycle, abrupt change is considered to have occurred. If a drastic change is highly consistent with a physical disturbance event in both time and space—that is, the change occurs within a specified very short time window and at an electrical node corresponding to the physical location—then the system determines that there is a strong causal relationship between the two. Conversely, if no drastic change is detected within the time window, or if the change occurs at an unrelated electrical node, it is determined that there is no correlation. Finally, the system encapsulates this determination, along with the location and time information of the original event, into a structured data unit, namely a spatiotemporal event correlation marker, as the final output of this step.

[0071] In one specific embodiment of the present invention, the extremely short time window is a time interval centered on the occurrence time of the physical event, and its function is to limit the search range of electrical data to the direct impact period of the event. The width of this window is set based on the typical response speed of power system faults. Based on the analysis of hundreds of ground fault waveform data, setting it to 10 seconds can effectively capture most of the changes in electrical parameters directly caused by physical impacts.

[0072] The composite risk diagnosis module inputs spatiotemporal event correlation markers and combines them with continuous degradation trend signals to diagnose and form a composite risk causation profile.

[0073] In a specific embodiment of the present invention, the specific steps of diagnosing and forming a composite risk causal profile include: evaluating spatiotemporal event correlation markers; if the markers indicate a strong causal relationship, the risk causal cause is diagnosed as external physical damage.

[0074] If the spatiotemporal event correlation is marked as uncorrelated, the amplitude of the continuous degradation trend signal is further evaluated. If it exceeds the preset attention level threshold, the risk cause is diagnosed as continuous degradation of the internal conductor.

[0075] By integrating the diagnosed risk causes, location information, and severity, a composite risk cause profile is formed.

[0076] It should be noted that this step aims to comprehensively diagnose the risks currently facing the grounding grid, clarifying their root causes, locations, and nature. The process begins by receiving the spatiotemporal event correlation markers generated in the previous step, while simultaneously retrieving the latest continuous degradation trend signal. The system first evaluates the spatiotemporal event correlation markers. If the correlation conclusion field in the marker shows "strong causal correlation," it means that the system has confirmed that an external physical impact event directly caused a sudden change in the electrical performance of the grounding grid. In this case, the system directly diagnoses the risk cause as "external physical damage." Simultaneously, the event location information, such as "at 3200 meters," is extracted from the spatiotemporal event correlation markers and used as the precise location of this risk. If the conclusion of the spatiotemporal event correlation markers is "no correlation," it indicates that no external physical event has recently occurred that can explain the current electrical state. At this point, the system will proceed to evaluate the continuous degradation trend signal characterizing slow internal damage. The system will compare the current amplitude of this signal with a preset attention level threshold. This threshold is set based on equipment safety operating procedures and historical data analysis, representing the critical point where the grounding grid performance has deteriorated to the point requiring attention. If the amplitude of the continuous degradation trend signal exceeds this attention level threshold, for example, if the degradation rate exceeds 0.01 ohms / day, the system diagnoses the risk cause as "continuous internal conductor degradation." Because this degradation is slow and widespread, its location is typically marked as the entire grounding grid or a specific area with the most severe degradation. After determining the risk cause and location, the system further integrates all relevant information, including the diagnosed risk cause, such as "external physical damage" or "continuous internal conductor degradation"; the risk's location information; and quantitative indicators representing the severity, such as the amplitude of the degradation trend signal. All this information is organized into a clearly structured and easily interpretable data report, a composite risk cause profile, as the final output of this step, providing a comprehensive basis for subsequent early warning and decision-making.

[0077] It should also be noted that the composite risk causation profile is a structured diagnostic report whose function is to comprehensively depict the source, location, and severity of current risks to the grounding grid. It typically includes fields such as risk causes, risk location, and severity assessment. External physical damage is one type of risk cause, specifically referring to acute faults such as conductor breakage and connection point damage directly caused by external physical forces such as third-party construction or geological disasters. Continuous internal conductor degradation is another type of risk cause, specifically referring to chronic damage caused by internal factors such as long-term chemical corrosion, electrochemical corrosion, or material fatigue, resulting in a reduction in the cross-sectional area of ​​the grounding grid conductor and an increase in connection resistance. The attention level threshold sets a warning line for continuous degradation trend signals, its function being to distinguish between normal, minor performance fluctuations and persistent degradation trends requiring attention. This threshold is set based on industry standards for the long-term change rate of grounding resistance and statistical analysis of grounding grid aging data under several different operating conditions. This threshold is triggered when the degradation rate indicates that the grounding resistance will exceed the safe upper limit within one year.

[0078] Please see Figure 2 The graded early warning release module takes a composite risk cause profile as input and releases graded dynamic risk warnings based on preset risk judgment rules.

[0079] In a specific embodiment of the present invention, the specific steps of issuing graded dynamic risk warnings according to preset risk determination rules include: matching preset risk levels in the risk determination rules based on the risk causes diagnosed in the composite risk cause profile, wherein external physical damage is matched as an emergency level red warning, and continuous internal conductor deterioration is matched as a warning level yellow warning.

[0080] It should be noted that this step, as the endpoint of the entire early warning method, is responsible for transforming the analysis and diagnostic results of the preceding steps into action instructions with clear guidance for operations and maintenance personnel. The process is automatically triggered upon receiving the composite risk cause profile generated in the previous step. The system first reads the core field in the profile, namely "risk cause." Then, the system queries a built-in risk judgment rule base. This base predefines a hard correspondence between different risk causes and early warning levels using an "if-then" logic. For example, the rule base explicitly stipulates that if the risk cause is "external physical damage," it is automatically matched to the highest level, "emergency-level red alert," because such events are usually sudden, highly destructive, and require immediate response. If the risk cause is "continuous internal conductor degradation," it is matched to the next lower level, "attention-level yellow alert," because it represents a slowly developing process, providing a window of opportunity for planned action.

[0081] Based on the risk level, the system automatically generates early warning content that includes the risk type, precise location, quantitative indicators, and recommended measures.

[0082] It should be noted that after the risk level is determined, the system will enter the automatic generation phase of the warning content. It will extract all key information from the composite risk causation profile and assemble it according to a standardized template. The template will sequentially fill in the risk type (directly referencing the diagnosed risk cause); the precise location (referencing the location information in the profile); the quantitative indicators (referencing data in the profile used to characterize the severity of the event, such as the rate of degradation); and the recommended measures retrieved from the risk assessment rule base based on the risk level. For example, for an "emergency-level red alert," the recommended measure might be "immediately dispatch inspection personnel to the designated location for verification and prepare for emergency repairs"; for a "caution-level yellow alert," it might be "add this location to the key attention list and schedule a detailed excavation investigation in the next maintenance cycle."

[0083] The generated warning content will be published using the specified method.

[0084] It's important to note that in the final step, the system publishes the fully generated alert content using a pre-configured method. This method can be varied, such as sending an alert SMS to the mobile phone of the maintenance team leader, pushing a pop-up alert to the monitoring screen in the central control center, or automatically creating an emergency or planned maintenance work order in the work order management system. Through this series of automated operations, a closed-loop management system is achieved, encompassing the entire process from data perception to risk diagnosis and then to the issuance of precise action instructions.

[0085] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A dynamic hierarchical early warning system for grounding grid safety risks, characterized in that, include: The raw data packet generation module generates a time-aligned raw data packet containing grounding grid electrical response information and synchronization environment parameters according to a preset detection cycle. The health status spectrum generation module, based on time-aligned raw data packets, generates an electrical body health spectrum reflecting the state of the grounding grid conductor itself by separating environmental influences. The degradation trend analysis module takes the electrical body health spectrum as input and identifies and quantifies a continuous degradation trend signal by comparing it with historical health status. The structural disturbance monitoring module uses distributed optical fiber sensing cables laid along the grounding grid conductor to monitor and generate distributed structural disturbance data containing information on physical external force events. The spatiotemporal event correlation module takes into account distributed structural disturbance data and correlates it with the electrical body health spectrum to construct a spatiotemporal event correlation marker to determine the causal relationship between physical events and electrical faults. The composite risk diagnosis module takes into account spatiotemporal event correlation markers and combines them with continuous deterioration trend signals to diagnose and generate a composite risk causation profile. The tiered early warning release module allows users to input a composite risk causal profile and release tiered dynamic risk warnings based on preset risk assessment rules.

2. The dynamic hierarchical early warning system for grounding grid safety risks according to claim 1, characterized in that: The specific steps for generating the time-aligned raw data packet containing grounding grid electrical response information and synchronization environment parameters include: A set of microcurrent pulses is injected into the grounding grid, and a multi-frequency composite impedance response spectrum is formed by synchronously measuring the key nodes of the grounding grid. Using environmental sensors, soil moisture, temperature, and pH data are collected in real time, synchronized with the measurement time of the multi-frequency composite impedance response spectrum, to form synchronized environmental parameters; The multi-frequency composite impedance response spectrum is bound to the synchronization environment parameters with a unified timestamp to generate time-aligned raw data packets.

3. The dynamic hierarchical early warning system for grounding grid safety risks according to claim 2, characterized in that: The specific steps for generating the electrical health spectrum that reflects the state of the grounding grid conductor itself include: High-frequency response data sensitive to environmental changes are extracted from the multi-frequency composite impedance response spectrum. A dynamic interference model is established based on the historical correspondence between high-frequency response data and synchronization environment parameters. By applying a dynamic disturbance model, the response changes caused by synchronization environment parameters are calculated and extracted from the multi-frequency composite impedance response spectrum to generate the electrical body health spectrum.

4. The dynamic hierarchical early warning system for grounding grid safety risks according to claim 3, characterized in that: The specific method for establishing a dynamic interference model is as follows: using a large number of synchronization environment parameters from historical data as input variables and corresponding high-frequency band response data as output variables, a dynamic interference model that quantifies the relationship between changes in environmental parameters and changes in high-frequency impedance is established through training with a multivariate regression algorithm.

5. A dynamic hierarchical early warning system for grounding grid safety risks according to claim 3, characterized in that: The specific steps for identifying and quantifying a continuous deterioration trend signal by comparing with historical health status include: The latest electrical body health spectrum is compared with the preset health baseline spectrum, and the difference is calculated; Trend analysis of differences in continuous time series can identify deterioration components that exhibit continuous unidirectional growth. The degradation component is quantified into a continuous degradation trend signal.

6. The dynamic hierarchical early warning system for grounding grid safety risks according to claim 1, characterized in that: The specific steps for monitoring and generating distributed structural disturbance data containing information on physical external force events include: Using optical time-domain reflectometry, we continuously monitor changes in optical signals caused by stress or vibration on distributed optical fiber sensing cables. When a sudden change in the light signal exceeding the normal threshold is detected, the occurrence time, precise physical location, and intensity of the sudden event are determined. The occurrence time, precise physical location, and event intensity are encapsulated as distributed structural perturbation data.

7. A dynamic hierarchical early warning system for grounding grid safety risks according to claim 6, characterized in that: The specific steps for constructing a spatiotemporal event association marker to determine the causal relationship between physical events and electrical faults include: Using a pre-defined geographic information mapping table, the precise physical location in the distributed structural disturbance data is converted into the adjacent electrical node in the grounding grid topology; The search examines whether the electrical health spectrum exhibits drastic jumps at electrical nodes at precise physical locations within the time frame of the occurrence of the distributed structural disturbance data markers and within a very short time window before and after it. If a physical disturbance event and a drastic change in the electrical body's health spectrum are consistent in time and space, then a strong causal relationship is determined between the two. Generate a spatiotemporal event association tag that includes a strong causal relationship conclusion, event location, and time.

8. A dynamic hierarchical early warning system for grounding grid safety risks according to claim 7, characterized in that: The specific steps for diagnosing and generating a composite risk profile include: Assess the spatiotemporal event correlation markers. If the markers indicate a strong causal relationship, diagnose the risk cause as external physical damage. If the spatiotemporal event correlation is marked as uncorrelated, the amplitude of the continuous degradation trend signal is further evaluated. If it exceeds the preset attention level threshold, the risk cause is diagnosed as continuous degradation of the internal conductor. By integrating the diagnosed risk causes, location information, and severity, a composite risk cause profile is formed.

9. A dynamic hierarchical early warning system for grounding grid safety risks according to claim 8, characterized in that: The specific steps for issuing tiered dynamic risk warnings based on preset risk assessment rules include: Based on the risk causes diagnosed in the composite risk cause profile, the preset risk level is matched in the risk judgment rules, where external physical damage is matched as an emergency level red warning, and internal conductor continuous deterioration is matched as a warning level yellow warning. Based on the risk level, it automatically generates early warning content that includes the risk type, precise location, quantitative indicators, and recommended measures; The generated warning content will be published using the specified method.

Images