Synchronous monitoring and early warning method for substation ground current micro-power consumption

Abstract

The invention discloses a synchronous monitoring and early warning method for substation ground current micro-power consumption, and relates to the technical field of substation monitoring, and the method comprises the steps: obtaining a ground current data set; performing dynamic time warping on the ground current data set to obtain a time sequence synchronization data set; performing time sequence analysis based on the time sequence synchronization data set, and determining a plurality of ground current characteristics; traversing the time sequence synchronization data set to carry out current fluctuation learning, and generating a ground current prediction data set; and dynamically monitoring the grounding network of the transformer substation, and generating a plurality of feedback response signals to intelligently trigger the alarm of the transformer substation. The technical problems that in the prior art, ground current data of multiple ground points cannot be collected in real time, real-time abnormal fluctuation cannot be accurately captured, potential risks of ground current cannot be effectively recognized in advance, and the early warning instantaneity and accuracy are insufficient are solved, low-power-consumption continuous monitoring of the ground current is achieved, and the early warning efficiency is improved. The technical effect of improving the real-time performance and accuracy of monitoring and early warning is achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to substation monitoring, and in particular to a synchronous monitoring and early warning method for ground current micro-power consumption in substations. Background Art

[0002] With the continuous development of power systems and the improvement of the level of intelligence, substations, as key nodes of power systems, are crucial to the reliability of the entire power grid for their safe and stable operation. As an important part of the safety protection system, the substation grounding network is responsible for safely guiding the fault current or overcurrent generated by equipment and systems to the ground in the event of a fault or abnormality in the power system to protect equipment and personal safety. However, in actual operation, the grounding network may be affected by factors such as equipment aging, environmental changes, fault current, and lightning strikes, resulting in fluctuations and anomalies in ground current. Especially in large-scale power systems, the ground current characteristics of different grounding points vary greatly, and they change frequently, which may cause equipment failures or even power grid accidents, and put forward higher monitoring requirements for the operation of the grounding network. However, traditional ground current monitoring is mainly based on periodic sampling, obtaining ground current data from each grounding point to analyze current fluctuations. There is a lag in the acquisition timing, and it is impossible to synchronously monitor the data changes of multiple grounding points. It is difficult to accurately capture the real-time abnormal fluctuation characteristics, and it is impossible to effectively identify the dynamic mode and potential trend in the ground current change, so that the ground current cannot be monitored and warned in a timely and effective manner, which may cause grounding system overload, equipment damage, and even endanger personnel safety.

[0003] At present, the technologies related to synchronous monitoring and early warning of ground current in substations have the problem of being unable to collect ground current data from multiple grounding points in real time and difficult to accurately capture real-time abnormal fluctuations, which in turn leads to the inability to identify the potential risks of ground current in advance and effectively, resulting in technical problems such as insufficient real-time and accuracy of early warning. Summary of the invention

[0004] The present application provides a synchronous monitoring and early warning method for micro-power consumption of ground current in substations, thereby solving the technical problems in the prior art that ground current data of multiple grounding points cannot be collected in real time, and it is difficult to accurately capture real-time abnormal fluctuations, which leads to the inability to identify the potential risks of ground current in advance and effectively, resulting in insufficient real-time and accuracy of early warning. The application realizes low-power continuous monitoring of ground current, achieving the technical effect of improving the real-time and accuracy of monitoring and early warning.

[0005] The present application provides a synchronous monitoring and early warning method for ground current micro-power consumption in a substation, comprising: sensing multiple equipment grounding points of the substation through the grounding network of the substation to obtain a ground current data set; traversing the multiple equipment grounding points to determine the acquisition timing, dynamically time-regulating the ground current data set according to the acquisition timing to obtain a timing synchronization data set; performing time series analysis based on the timing synchronization data set, performing feature extraction according to the analysis results, and determining multiple ground current characteristics; traversing the timing synchronization data set according to the multiple ground current characteristics to perform current fluctuation learning, and generating a ground current prediction data set according to the learning results; dynamically monitoring the grounding network of the substation according to the ground current prediction data set, generating multiple feedback response signals to intelligently trigger an alarm for the substation.

[0006] In a possible implementation, multiple equipment grounding points of the substation are sensed through the grounding network of the substation to obtain a ground current data set, and the following processing is performed: the substation is covered and traversed based on the grounding network to obtain multiple collection nodes of the substation; the substation is current captured according to the multiple collection nodes, and the multiple equipment grounding points are determined according to the current capture results; multiple sensor devices are started to set the collection frequency according to the substation operation status information; the multiple equipment grounding points are traversed and sensed based on the multiple sensor devices according to the collection frequency to obtain multiple ground current data signals, each of which contains a timestamp; the multiple ground current data signals are integrated to determine the ground current data set.

[0007] In a possible implementation, the multiple device grounding points are traversed to determine the acquisition timing, the ground current data set is dynamically time-warped according to the acquisition timing to obtain a timing synchronization data set, and the following processing is performed: the multiple device grounding points are traversed to perform timing processing in combination with the timestamp of each ground current data signal to determine the acquisition timing; the acquisition timing is used as a timing reference for node selection to determine a timing benchmark; the multiple ground current data signals are amplified to obtain multiple ground current enhancement signals; the multiple ground current enhancement signals are dynamically time-warped based on the timing benchmark to determine the signal time step; the ground current data set is interpolated and padded according to the signal time step to generate multiple time windows; the consistency of the multiple time windows is verified, and the multiple time windows are associated and integrated according to the verification results to output the timing synchronization data set.

[0008] In a possible implementation, the multiple ground current enhancement signals are dynamically time-warped based on the timing benchmark to determine the signal time step, and the following processing is also performed: similarity calculation is performed on the multiple ground current enhancement signals based on the timing benchmark to obtain multiple similarity coefficients; a two-dimensional matrix is ​​constructed according to the timing benchmark, and Euclidean distance calculation is performed on the multiple ground current enhancement signals according to the multiple similarity coefficients to generate multiple distance data; the multiple distance data are mapped to the two-dimensional matrix to perform path optimization and determine the timing alignment path; the multiple ground current enhancement signals are time-aligned based on the timing alignment path to generate the signal time step.

[0009] In a possible implementation, the time-series synchronization data set is traversed according to the multiple ground current characteristics to perform current fluctuation learning, and a ground current prediction data set is generated according to the learning results. The following processing is also performed: sliding based on the multiple time windows combined with the multiple ground current characteristics to obtain multiple sliding windows; traversing and capturing according to the multiple sliding windows to obtain multiple current dynamic fluctuation data; using a long short-term memory neural network to learn the multiple current dynamic fluctuation data to generate a learning result; performing trend analysis according to the learning result combined with the time-series synchronization data set to draw a current fluctuation curve graph; training is performed based on the current fluctuation curve graph combined with the learning results to obtain training results, and prediction is performed based on the training result combined with the multiple ground current characteristics according to the time series benchmark to generate a ground current prediction data sequence; and the ground current prediction data sequence is added to the ground current prediction data set.

[0010] In a possible implementation, the grounding network of the substation is dynamically monitored according to the ground current prediction data set, and multiple feedback response signals are generated to intelligently trigger an alarm for the substation, and the following processing is also performed: multiple operating condition information is retrieved, and the grounding network is started based on the multiple operating condition information to perform operation simulation analysis to obtain multiple ground current operation simulation data; the multiple ground current operation simulation data are traversed and matched with the ground current prediction data set to generate multiple simulation-prediction data pairs; the multiple simulation-prediction data pairs are dynamically compared, and multiple feedback response signals are generated according to the comparison results; the multiple feedback response signals are traversed and combined with the multiple simulation-prediction data pairs to perform current change analysis to generate multiple abnormal labels; the multiple feedback response signals are labeled according to the multiple abnormal labels, multiple abnormal signals are determined, and intelligently trigger an alarm according to the multiple abnormal signals.

[0011] In a possible implementation, the multiple feedback response signals are traversed and combined with the multiple simulation-prediction data pairs to perform current change analysis, generate multiple abnormal labels, and further perform the following processing: perform current feedback analysis based on the multiple feedback response signals to determine multiple current change characteristics; set a current change deviation threshold according to the multiple current change characteristics, perform error calculation on the multiple simulation-prediction data pairs, and generate multiple current change error values; compare the multiple current change error values ​​with the current change deviation threshold to determine whether the multiple current change error values ​​are greater than or equal to the current change deviation threshold; if the multiple current change error values ​​are greater than or equal to the current change deviation threshold, then identify the simulation-prediction data pairs corresponding to the current change error values ​​greater than or equal to the current change deviation threshold to generate the multiple abnormal labels.

[0012] In a possible implementation, the multiple feedback response signals are identified according to the multiple abnormal labels, multiple abnormal signals are determined, and alarms are intelligently triggered according to the multiple abnormal signals. The following processing is also performed: using the multiple abnormal labels as indexes, traversing the multiple feedback response signals for matching, and determining multiple abnormal signals; performing impact assessments based on the multiple abnormal signals, and generating multiple impact levels according to the assessment results; screening the multiple abnormal signals according to the multiple impact levels to generate signal screening results; setting multiple alarm triggering conditions according to the signal screening results, and intelligently triggering alarms based on the multiple alarm triggering conditions and the multiple abnormal signals.

[0013] Through the synchronous monitoring and early warning method for ground current micro-power consumption in substations proposed in this application, a ground current data set is obtained; the ground current data set is dynamically time-warped to obtain a time-series synchronization data set; time series analysis is performed based on the time-series synchronization data set to determine multiple ground current characteristics; the time-series synchronization data set is traversed to learn current fluctuations and generate a ground current prediction data set; the grounding network of the substation is dynamically monitored to generate multiple feedback response signals to intelligently trigger alarms for the substation. This solves the technical problems in the prior art that ground current data of multiple grounding points cannot be collected in real time, and it is difficult to accurately capture real-time abnormal fluctuations, which leads to the inability to identify potential risks of ground current in advance and effectively, resulting in insufficient real-time and accuracy of early warnings. Low-power continuous monitoring of ground currents is achieved, achieving the technical effect of improving the real-time and accuracy of monitoring and early warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A flow chart of a synchronous monitoring and early warning method for ground current micro-power consumption in a substation provided in an embodiment of the present application;

[0016] Figure 2 A schematic diagram of a flow chart of obtaining a time-series synchronous data set in a synchronous monitoring and early warning method for ground current micro-power consumption in a substation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The embodiment of the present application provides a synchronous monitoring and early warning method for substation ground current micro-power consumption, such as Figure 1 As shown, the method includes:

[0021] Step S100, sensing multiple equipment grounding points of the substation through the grounding network of the substation to obtain a ground current data set.

[0022] Preferably, the grounding network usually covers multiple key equipment of the substation, including transformers, switch cabinets, busbars, control cabinets, etc. These devices have their own grounding points, which are responsible for directing excess current to the ground in the event of a fault or abnormal situation to ensure the safety of the equipment. Specifically, current transformers, Hall effect sensors or micro-power consumption measurement devices are installed at the grounding point of each key equipment to monitor the current size and changes at each equipment grounding point in real time. These current sensors are connected to the substation grounding network, and by detecting the current flow in the grounding system, the ground current data of the grounding point are collected, including the current value of each grounding point and its changes at different times, such as instantaneous current value, periodic current fluctuations and abnormal current peaks, etc., to form a ground current data set containing multiple time points and multiple grounding points. The ground current data set includes the ground current record of each equipment grounding point, that is, it contains the ground current distribution characteristics of the substation equipment under normal and abnormal conditions.

[0023] In a possible implementation, step S100 further includes step S110, covering and traversing the substation based on the grounding network to obtain multiple collection nodes of the substation; step S120, capturing current of the substation according to the multiple collection nodes, and determining the multiple equipment grounding points according to the current capture results; step S130, starting multiple sensor devices to set the collection frequency according to the substation operation status information; step S140, traversing and sensing the multiple equipment grounding points according to the collection frequency based on the multiple sensor devices to obtain multiple ground current data signals, each of which includes a timestamp; step S150, integrating the multiple ground current data signals to determine the ground current data set.

[0024] Preferably, the grounding network of the substation is traversed to determine each key position in the grounding network, that is, multiple acquisition nodes are obtained, which are usually located at key grounding points of the grounding system, including transformer grounding points, busbar grounding points, switch cabinet grounding points, etc., and sensing devices (such as current transformers or Hall sensors) are used at each acquisition node to capture current signals and record current data in real time. The actual position and connection status of each device grounding point are further confirmed through the current capture results; then, according to the operating status information of the substation (such as load conditions, equipment operating frequency, etc.), the sensing devices at each acquisition node are automatically started and the acquisition frequency is set. For example, the sampling frequency is increased in extreme conditions such as thunderstorms, high loads or equipment abnormalities to ensure that all key current fluctuation information is captured; the sampling frequency is lowered in normal conditions to save energy (for example, once per second); based on the set collection frequency of each collection node, the ground current signal of each grounding point is collected using a sensor device. Each signal contains a timestamp to record the specific moment of sampling; finally, the ground current data signal of each collection node is integrated, including timestamp alignment, signal denoising and data correction, to form a ground current data set, providing real-time and accurate monitoring data for substation ground current monitoring and fault diagnosis.

[0025] Step S200, traverse the multiple device grounding points to determine the acquisition timing, dynamically time-regulate the ground current dataset according to the acquisition timing, and obtain a timing synchronized dataset.

[0026] Preferably, there are multiple equipment grounding points (such as transformer grounding points, busbar grounding points, switch cabinet grounding points, etc.) in the substation, and each grounding point has an independent ground current sensor. Specifically, traverse multiple equipment grounding points in the substation to determine the collection sequence, that is, assign a collection time sequence to each grounding point, that is, set a polling mechanism. This collection sequence can be optimized according to the physical location of the sensor, signal transmission delay or other factors, so that the collection process is both efficient and meets the time synchronization requirements of the system. Then, the acquired ground current data set is dynamically time-warped according to the collection sequence. Specifically, by dynamically adjusting the time axis of each time series, the ground current data of different sampling times can be aligned under the same time reference, that is, adjusting the time label of the sampling point, so that the data of each grounding point is synchronized in time, wherein the dynamic time warping (Dynamic Time DTW is used to align time series data, which is particularly suitable for processing the time deviation between different data collection points. Since the sampling times of different grounding points are slightly different, directly comparing their time series data may result in time misalignment. Finally, a time series synchronization data set is generated, which contains the ground current values ​​of each substation grounding point at the same time. It is used to analyze the ground current relationship between devices, identify synchronous fluctuation patterns, and improve the accuracy of anomaly detection.

[0027] In one possible implementation, Figure 2 As shown, step S200 further includes step S210, traversing the multiple device grounding points and combining the timestamp of each ground current data signal for timing processing to determine the acquisition timing; step S220, using the acquisition timing as a timing reference for node selection to determine the timing benchmark; step S230, amplifying the multiple ground current data signals to obtain multiple ground current enhancement signals; step S240, dynamically time-warping the multiple ground current enhancement signals based on the timing benchmark to determine the signal time step; step S250, interpolating and filling the ground current data set according to the signal time step to generate multiple time windows; step S260, verifying the consistency of the multiple time windows, correlating and integrating the multiple time windows according to the verification results, and outputting the timing synchronization data set.

[0028] Preferably, the grounding point of each device in the substation is traversed to obtain the ground current data signal of each grounding point, each signal has a timestamp, and the obtained ground current data signals are sorted according to the timestamp to generate an acquisition timing, which reflects the temporal sequence of the ground current data, and then used as a timing reference, and then a key node or a group of nodes is selected, such as selecting the average value of the sampling time of each device or selecting the time node where the data first appears as a timing reference, which is used to compare and synchronize the acquisition timing of other grounding points, and the acquisition timing of the reference node is used as the timing reference, as the standard time axis for subsequent dynamic time regularization, to ensure that all data points can be aligned in a unified time frame; then the multiple ground current data signals are amplified, the amplitude of the signal is enhanced to improve the observability of the signal, and multiple ground current enhanced signals are obtained to ensure that the signal amplitude and frequency are at a unified order of magnitude, so as to facilitate subsequent timing alignment and analysis.

[0029] Preferably, when the signal time step is determined, dynamic time warping (DTW) is used to time align the enhanced signals, that is, by adjusting the signal time step, the enhanced signals of each grounding point can be aligned to the same timing reference, and then the signal time step is determined. This signal time step represents the time interval of each time step, ensuring that all data points are on the same time scale; in each enhanced signal, there may be some missing data points. According to the determined signal time step, the ground current data set is interpolated to fill the missing data points to ensure that each signal sequence has a continuous time series, and the interpolated and filled data are divided into multiple time windows according to a fixed time interval, respectively representing different time periods of the ground current data; finally, a consistency check is performed on the generated multiple time windows to verify the rationality and consistency of the data in each window to ensure that there is no deviation or erroneous data, and then according to the results of the consistency verification, the data of each time window is associated and integrated to ensure that the data between the time windows are consistent in time and characteristics, thereby outputting a timing synchronization data set, integrating the ground current data signals of all equipment grounding points, and having a unified time step, time reference and complete time window.

[0030] In a possible implementation, step S240 further includes step S241, performing similarity calculation on the multiple ground current enhancement signals based on the timing benchmark to obtain multiple similarity coefficients; step S242, constructing a two-dimensional matrix according to the timing benchmark, performing Euclidean distance calculation on the multiple ground current enhancement signals according to the multiple similarity coefficients, and generating multiple distance data; step S243, mapping the multiple distance data to the two-dimensional matrix for path optimization and determining a timing alignment path; step S244, time-aligning the multiple ground current enhancement signals based on the timing alignment path to generate the signal time step.

[0031] Preferably, similarity calculation and Euclidean distance calculation are used to determine the best alignment path for each signal to achieve time synchronization processing. Specifically, based on the set timing benchmark, the similarity of multiple ground current enhancement signals is calculated, such as using cosine similarity, Pearson correlation coefficient or dynamic time warping similarity, to quantify the similarity between each pair of signals. The similarity calculation result of each pair of signals is called a similarity coefficient. The higher the similarity coefficient, the more similar the change patterns of the two signals are. A two-dimensional matrix is ​​constructed according to the timing benchmark, in which each element represents the similarity of a pair of ground current enhancement signals at a specific time point. According to the similarity coefficient, the Euclidean distance calculation is performed for each time point of each pair of ground current enhancement signals to quantify the degree of difference between the two signals at a certain time point. The smaller the distance, the more similar the two signals are at a certain time point. The higher the similarity at the time point, the more distance data are generated; then the multiple distance data are mapped into a two-dimensional matrix, and dynamic programming or dynamic time warping (DTW) is used to find an optimal path in the matrix so that the cumulative distance along the path is minimized, and this is used as the timing alignment path, which represents the best time alignment of multiple signals and ensures that the signals are synchronized on the time axis; finally, based on the determined timing alignment path, multiple ground current enhancement signals are adjusted in time to align them on the time axis, including adjusting the time step of each signal so that they are consistent at every moment, and finally a signal time step is generated, which reflects the alignment interval at each time point, reduces the error caused by the acquisition timing difference, and provides a high-precision timing benchmark for timing analysis and anomaly detection.

[0032] Step S300 , performing time series analysis based on the time series synchronization data set, performing feature extraction according to the analysis result, and determining a plurality of ground current features.

[0033] Preferably, the ground current data set after time series synchronization is used to identify and extract important features reflecting the characteristics of the ground current through time series analysis technology. Specifically, the long-term change trend of the ground current is determined through time series analysis to reflect the overall change of the grounding system over a period of time (for example, a gradually increasing current may indicate an increase in grounding resistance or equipment aging). The periodic components that may exist in the ground current signal are detected using methods such as autocorrelation and periodogram to obtain periodic features, identify mutation points in the ground current signal, such as instantaneous current peaks or abnormal fluctuations, which may indicate instantaneous faults or abnormal conditions (such as short circuits, changes in ground resistance, etc.), and check whether the ground current data remains stable over time. If the signal shows significant non-stationarity, it is further analyzed. Analysis is performed to identify potential sources of anomalies; then, based on time series analysis, multiple features are extracted from the ground current data, which contain the core feature information of the ground current signal and can be used to monitor the status of the grounding network in real time. For example, the mean and standard deviation are used to describe the central trend and fluctuation degree of the ground current, and the maximum and minimum values ​​and their spacing in the ground current are extracted to reflect the extreme current fluctuations, which is helpful to identify sudden anomalies. The intensity of the periodic component is quantified by Fourier transform and other methods. The periodic intensity reflects the repetitive pattern of the ground current, which may be different under normal and abnormal conditions. The autocorrelation coefficient is used to measure the correlation of the ground current at different time lags, calculate the time rate of change of the ground current signal, and identify the frequency and degree of fluctuations or mutations.

[0034] Step S400 , traversing the time-series synchronization data set according to the multiple ground current characteristics to perform current fluctuation learning, and generating a ground current prediction data set according to the learning result.

[0035] Preferably, the ground current characteristics in the time series synchronization data set are analyzed and modeled to learn the fluctuation pattern and change law of the ground current, so as to predict the change trend of the ground current in the future. Specifically, the time series synchronization data set is traversed according to multiple ground current characteristics, that is, the ground current time series data of each substation equipment grounding point is analyzed, the historical fluctuation of the ground current at each grounding point is captured, and its change pattern is discovered, and then the current fluctuation learning is performed using a machine learning model or a deep learning model. For example, a fluctuation prediction model is constructed based on a time series model, such as ARIMA, LSTM (long short-term memory network), smoothing filtering or regression analysis, etc., aiming to extract the inherent law of ground current fluctuation by analyzing the change pattern of the ground current characteristics of each grounding point over time, capture its time series characteristics, and predict the ground current characteristic values ​​at future times according to the fluctuation pattern in the historical data. For example, the LSTM model can be used to predict the change trend of the ground current in the next few minutes or hours, and the predicted ground current characteristic values ​​are combined to form a ground current prediction data set, which includes the predicted ground current value and fluctuation of each grounding point at a future time as the possible distribution of the ground current in the future.

[0036] In a possible implementation, step S400 further includes step S410, sliding based on the multiple time windows in combination with the multiple ground current characteristics to obtain multiple sliding windows; step S420, traversing and capturing according to the multiple sliding windows to obtain multiple current dynamic fluctuation data; step S430, using a long short-term memory neural network to learn the multiple current dynamic fluctuation data to generate a learning result; step S440, performing trend analysis according to the learning result in combination with the time series synchronization data set, and drawing a current fluctuation curve graph; step S450, training based on the current fluctuation curve graph in combination with the learning result to obtain a training result, and predicting based on the training result in combination with the multiple ground current characteristics according to the time series benchmark to generate a ground current prediction data sequence; step S460, adding the ground current prediction data sequence to the ground current prediction data set.

[0037] Preferably, the geocurrent data is divided into multiple time windows according to the time axis, each window contains the geocurrent data within a certain time period, and the sliding window technology is used to slide on the time axis with a fixed step size to generate multiple sliding windows containing different time periods, and the dynamic fluctuation data of the geocurrent is traversed and captured in the multiple sliding windows, and the geocurrent characteristics (such as peak value, change rate, periodicity, etc.) are extracted, wherein the dynamic fluctuation data reflects the instantaneous change of the geocurrent in each sliding window; then a prediction model is constructed based on a long short-term memory neural network (LSTM), and the captured current dynamic fluctuation data is input into the prediction model for training, and the inherent law and time-dependent characteristics of the geocurrent fluctuation are learned, and the learning results are output, including the model's understanding and prediction ability of the future fluctuation trend of the geocurrent, wherein the long short-term memory neural network is a deep learning model suitable for time series analysis, which can capture Long-term and short-term dependencies; combining the learning results and the time-series synchronization data set, the ground current data is trend analyzed. Specifically, by analyzing the changing trend of the ground current over time, its main modes, such as the upward trend, periodic fluctuations, etc., are identified, and the trend analysis results are displayed in the form of a curve graph, and a fluctuation curve graph of the ground current over time, that is, a current fluctuation curve graph, is drawn; then based on the drawn current fluctuation curve graph and the learning results, the prediction model is trained so that the model can better capture the fluctuation characteristics of the ground current, and improve the prediction accuracy of the model and the sensitivity to abnormal fluctuations; finally, combining the training results and multiple ground current characteristics, and according to the time series benchmark, the fluctuation of the future ground current is predicted, and a prediction data sequence is generated, which reflects the ground current fluctuation of each grounding point in the future, and the generated ground current prediction data sequence is added to the ground current prediction data set.

[0038] Step S500: dynamically monitor the grounding network of the substation according to the ground current prediction data set, and generate multiple feedback response signals to intelligently trigger an alarm for the substation.

[0039] Preferably, the future ground current trend provided by the prediction data set is used to dynamically monitor the operating status of the grounding network in real time, generate feedback signals for abnormal situations, and trigger intelligent alarms for timely response. Specifically, the expected ground current value in the ground current prediction data set is compared with the actually collected ground current data to detect whether an abnormal situation occurs. A dynamic threshold is set based on the prediction data. If the actual data exceeds the prediction range or reaches a certain threshold, it may indicate that the grounding network has an abnormality or potential fault. Dynamic monitoring refers to automatically adjusting the focus and frequency of monitoring according to changes in the prediction data. For example, when the prediction data indicates that the ground current fluctuation at a certain grounding point is intensifying, the system can increase the monitoring of that point. frequency to capture more detailed real-time data; and then generate multiple feedback response signals, including warning signals, risk warning signals, abnormal alarm signals, etc. When the deviation between the actual data and the predicted data reaches the set threshold, the corresponding feedback signal will be generated and divided into different levels (such as early warning, warning, emergency alarm, etc.), and the real-time feedback will be automatically provided through the dispatching sensor and the intelligent trigger alarm will be performed, that is, the intelligent alarm can be divided into different types and priorities. For example, for slight deviations or fluctuations, a low-level alarm will be triggered to remind the operation and maintenance personnel to pay attention; when significant deviations or high-risk fluctuations are detected, a high-level alarm will be triggered to indicate that there may be a risk of failure and that it needs to be checked and handled immediately. Intelligent alarms can be graded according to the priorities and triggering conditions of different response signals, and accurate alarms can be issued to optimize the response speed and processing efficiency of the operation and maintenance personnel, thereby improving the safety and stability of the substation grounding network.

[0040] In a possible implementation, step S500 further includes step S510, retrieving multiple operating condition information, starting the grounding network to perform operation simulation analysis based on the multiple operating condition information, and obtaining multiple ground current operation simulation data; step S520, traversing and matching the multiple ground current operation simulation data with the ground current prediction data set to generate multiple simulation-prediction data pairs; step S530, dynamically comparing the multiple simulation-prediction data pairs, and generating multiple feedback response signals based on the comparison results; step S540, traversing the multiple feedback response signals and combining the multiple simulation-prediction data pairs to perform current change analysis to generate multiple abnormal labels; step S550, identifying the multiple feedback response signals according to the multiple abnormal labels, determining multiple abnormal signals, and intelligently triggering alarms according to the multiple abnormal signals.

[0041] Preferably, abnormal ground current conditions are identified and alarms are triggered by comparison and analysis of simulation and prediction. Specifically, operating condition information (various conditions of the substation under different operating conditions, such as load changes, equipment operation, ambient temperature, etc.) is retrieved, and simulation analysis of the grounding network is started based on the operating condition information, that is, by simulating the behavior of the grounding network under different operating conditions, the ground current distribution under different conditions is obtained, and multiple ground current operation simulation data are generated, reflecting the theoretical ground current characteristics under various operating conditions. The simulation data is matched with the ground current prediction data set one by one, and the correspondence between each simulation data point and the prediction data point is found to form a simulation-prediction data pair. Each simulation-prediction data pair contains simulation data and prediction data at the same time and the same grounding point, which are used to compare the difference between the actual operating state and the predicted state.

[0042] Preferably, multiple simulation-prediction data pairs are dynamically compared, such as comparing current size, fluctuation amplitude, change trend, etc., to generate feedback response signals, wherein all data pairs correspond to a signal, and no matter the size of the difference in the data pairs, abnormal or normal, there will be a signal for feedback; then all feedback response signals are traversed, combined with the corresponding simulation-prediction data pairs, the actual changes in ground current are analyzed, including trend analysis, abnormal amplitude assessment, periodic change identification, etc., to identify which changes are beyond the normal range, and generate multiple abnormal labels based on the current change analysis results to identify specific abnormal types and characteristics (for example, excessive fluctuations, sudden increases, continuous exceeding of standards, etc.); finally, the feedback signal is identified, such as the cause of the abnormality, the location of occurrence, the degree of abnormality, etc., to distinguish different types of abnormalities, and then multiple abnormal signals are formed to reflect the abnormal state in the grounding network, and intelligently trigger alarms based on the abnormal signals.

[0043] In a possible implementation, step S540 further includes step S541, performing current feedback analysis based on the multiple feedback response signals to determine multiple current change characteristics; step S542, setting a current change deviation threshold according to the multiple current change characteristics, performing error calculation on the multiple simulation-prediction data pairs, and generating multiple current change error values; step S543, comparing the multiple current change error values ​​with the current change deviation threshold to determine whether the multiple current change error values ​​are greater than or equal to the current change deviation threshold; step S544, if the multiple current change error values ​​are greater than or equal to the current change deviation threshold, then marking the simulation-prediction data pairs corresponding to the current change error values ​​greater than or equal to the current change deviation threshold to generate the multiple abnormal labels.

[0044] Preferably, the actual feedback of the ground current is analyzed according to multiple feedback response signals to identify the change mode and trend of the current, and determine multiple current change characteristics, which may include the average change rate of the current, the fluctuation amplitude, the abnormal peak value, the periodic fluctuation, etc., reflecting the change law of the ground current under different conditions. Based on the determined current change characteristics, multiple corresponding deviation thresholds are set as a benchmark for judging the abnormal current change, reflecting the ground current fluctuation range under normal conditions, wherein the deviation threshold is used for subsequent error calculation and comparison, and the stricter the deviation threshold, the higher the sensitivity to abnormal changes; the actual current data of each pair of simulation-prediction data pairs is compared with the predicted current data, and the error between them is calculated, which is called the current change error value, and each An error value represents the degree of deviation between a simulation-prediction data pair. The larger the error value, the more significant the difference between the simulation and prediction data, and there may be anomalies. Then the generated current change error value is compared with the set current change deviation threshold one by one, that is, whether the error value exceeds the deviation threshold to identify abnormal changes. If the current change error value is greater than or equal to the current change deviation threshold, it means that the error value has reached or exceeded the allowable fluctuation range, which may indicate an abnormal current situation. For the simulation-prediction data pair whose error value exceeds the deviation threshold, it is identified as abnormal data. According to the identification result, multiple abnormal labels are generated for the current change error value that exceeds the deviation threshold, such as "mild abnormality", "moderate abnormality", "serious abnormality", etc.

[0045] In a possible implementation, step S550 further includes step S551, taking the multiple abnormal labels as indexes, traversing the multiple feedback response signals for matching, and determining multiple abnormal signals; step S552, performing impact assessment based on the multiple abnormal signals, and generating multiple impact levels according to the assessment results; step S553, screening the multiple abnormal signals according to the multiple impact levels, and generating signal screening results; step S554, setting multiple alarm trigger conditions according to the signal screening results, and intelligently triggering alarms based on the multiple alarm trigger conditions and the multiple abnormal signals.

[0046] Preferably, all feedback response signals are traversed, matched according to the abnormal labels, and each feedback signal is identified as containing abnormal conditions, and all feedback signals matching the abnormal labels are identified as abnormal signals, and multiple abnormal signals are determined, representing that the substation may have abnormal conditions, and then each abnormal signal is impact evaluated to quantify its potential impact on the grounding network and the system as a whole, such as evaluating the current fluctuation amplitude, duration, frequency of abnormal occurrence, etc. According to the impact evaluation results, each abnormal signal is assigned an impact level, such as divided into multiple levels (such as low, medium, and high) to indicate the severity and urgency of the abnormal signal. The higher the impact level, the greater the potential threat of the abnormal signal to the system. ; All abnormal signals are screened according to the impact level, and abnormal signals with different impact degrees are separated to generate signal screening results. For example, serious abnormal signals will be classified into the highest level group, while minor abnormal signals will be classified into lower level groups. Finally, according to the screening results of different impact levels, corresponding alarm trigger conditions are set for each group of abnormal signals, among which the alarm trigger conditions include response level, alarm method, processing priority, etc.; According to different trigger conditions, alarm signals of different urgency are issued, and intelligent alarm decisions are automatically made, so as to more accurately identify and respond to abnormalities of different levels, reduce false alarms and missed alarms, and timely and efficiently identify and handle potential threats, thereby improving the safety and stability of substations.

[0047] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A synchronous monitoring and early warning method for substation ground current micro-power consumption, characterized in that: The method comprises: The ground current data set is obtained by sensing multiple equipment grounding points of the substation through the substation grounding network; Traversing the plurality of equipment grounding points to determine a collection timing, and dynamically time-warping the ground current data set according to the collection timing to obtain a time-series synchronized data set; Performing time series analysis based on the time series synchronization data set, extracting features according to the analysis results, and determining multiple geocurrent features; Traversing the time-series synchronization data set according to the multiple ground current characteristics to perform current fluctuation learning, and generating a ground current prediction data set according to the learning results; The grounding network of the substation is dynamically monitored according to the ground current prediction data set, and multiple feedback response signals are generated to intelligently trigger alarms for the substation.

2. A synchronous monitoring and early warning method for substation ground current micro-power consumption as claimed in claim 1, characterized in that: The grounding network of the substation is used to sense multiple grounding points of the substation equipment to obtain a ground current data set. The method includes: Based on the grounding network, the substation is covered and traversed to obtain multiple collection nodes of the substation; Capturing current of the substation according to the multiple acquisition nodes, and determining the multiple equipment grounding points according to the current capturing results; Start multiple sensor devices to set the collection frequency according to the substation operation status information; Based on the multiple sensing devices, traversal sensing is performed on the multiple device grounding points according to the acquisition frequency to obtain multiple ground current data signals, each of which includes a timestamp; The plurality of ground current data signals are integrated to determine the ground current data set.

3. A synchronous monitoring and early warning method for substation ground current micro-power consumption as claimed in claim 2, characterized in that: Traversing the plurality of equipment grounding points to determine a collection timing, dynamically time-warping the ground current data set according to the collection timing, and obtaining a time-series synchronized data set, the method comprising: Traversing the plurality of equipment grounding points and combining the timestamp of each ground current data signal to perform timing processing to determine the acquisition timing; Using the acquisition timing as a timing reference to select nodes and determine a timing benchmark; Amplifying the plurality of ground current data signals to obtain a plurality of ground current enhancement signals; Performing dynamic time warping on the plurality of ground current enhancement signals based on the timing reference to determine a signal time step; Interpolating and completing the ground current data set according to the signal time step to generate multiple time windows; The consistency of the multiple time windows is verified, the multiple time windows are associated and integrated according to the verification result, and the time series synchronization data set is output.

4. A synchronous monitoring and early warning method for substation ground current micro-power consumption as claimed in claim 3, characterized in that: Dynamically time-warping the plurality of ground current enhancement signals based on the timing reference to determine a signal time step, the method comprising: Performing similarity calculation on the plurality of ground current enhancement signals based on the timing reference to obtain a plurality of similarity coefficients; Constructing a two-dimensional matrix according to the timing reference, performing Euclidean distance calculation on the multiple ground current enhancement signals according to the multiple similarity coefficients, and generating multiple distance data; Mapping the multiple distance data to the two-dimensional matrix to perform path optimization and determine a timing alignment path; The plurality of ground current enhancement signals are time-aligned based on the timing alignment path to generate the signal time step.

5. A synchronous monitoring and early warning method for substation ground current micro-power consumption as claimed in claim 3, characterized in that: The method includes: traversing the time-series synchronization data set according to the multiple ground current characteristics to perform current fluctuation learning, and generating a ground current prediction data set according to the learning result. Sliding the multiple time windows in combination with the multiple geocurrent characteristics to obtain multiple sliding windows; Perform traversal and capture according to the multiple sliding windows to obtain multiple current dynamic fluctuation data; Using a long short-term memory neural network to learn the multiple current dynamic fluctuation data to generate a learning result; Perform trend analysis according to the learning result and the time-series synchronization data set, and draw a current fluctuation curve graph; Perform training based on the current fluctuation curve diagram in combination with the learning result to obtain a training result, and perform prediction based on the training result in combination with the multiple ground current characteristics according to the timing reference to generate a ground current prediction data sequence; The geocurrent prediction data sequence is added to the geocurrent prediction data set.

6. A synchronous monitoring and early warning method for substation ground current micro-power consumption as claimed in claim 1, characterized in that: Dynamically monitoring the grounding network of the substation according to the ground current prediction data set, generating multiple feedback response signals to intelligently trigger an alarm for the substation, the method comprising: Retrieving multiple operating condition information, starting the grounding network to perform operation simulation analysis based on the multiple operating condition information, and obtaining multiple ground current operation simulation data; Performing traversal matching on the multiple geocurrent operation simulation data and the geocurrent prediction data set to generate multiple simulation-prediction data pairs; dynamically comparing the plurality of simulation-prediction data pairs, and generating a plurality of feedback response signals according to the comparison results; Traversing the plurality of feedback response signals and combining the plurality of simulation-prediction data pairs to perform current change analysis and generate a plurality of abnormal labels; The multiple feedback response signals are identified according to the multiple abnormal tags, multiple abnormal signals are determined, and alarms are intelligently triggered according to the multiple abnormal signals.

7. A synchronous monitoring and early warning method for substation ground current micro-power consumption as claimed in claim 6, characterized in that: The method of traversing the plurality of feedback response signals and combining the plurality of simulation-prediction data pairs to perform current change analysis and generate a plurality of abnormal labels comprises: Performing current feedback analysis based on the multiple feedback response signals to determine multiple current change characteristics; Setting a current change deviation threshold according to the multiple current change characteristics, performing error calculation on the multiple simulation-prediction data pairs, and generating multiple current change error values; Comparing the multiple current change error values ​​with the current change deviation threshold, and determining whether the multiple current change error values ​​are greater than or equal to the current change deviation threshold; If the multiple current change error values ​​are greater than or equal to the current change deviation threshold, the simulation-prediction data pairs corresponding to the current change error values ​​greater than or equal to the current change deviation threshold are marked to generate the multiple abnormal labels.

8. A synchronous monitoring and early warning method for substation ground current micro-power consumption as claimed in claim 6, characterized in that: The multiple feedback response signals are identified according to the multiple abnormal tags, multiple abnormal signals are determined, and an alarm is intelligently triggered according to the multiple abnormal signals. The method includes: Using the multiple abnormal labels as indexes, traversing the multiple feedback response signals for matching, and determining multiple abnormal signals; Performing an impact assessment based on the multiple abnormal signals, and generating multiple impact levels according to the assessment results; Screening the multiple abnormal signals according to the multiple impact levels to generate signal screening results; A plurality of alarm triggering conditions are set according to the signal screening results, and an alarm is intelligently triggered based on the plurality of alarm triggering conditions and the plurality of abnormal signals.

Images