Event-driven architecture-oriented spatio-temporal feature extraction method and system for multi-type fault power failure events

Abstract

The invention discloses an event-driven architecture-oriented spatial-temporal feature extraction method and system for multi-type fault power failure events, and belongs to the field of power system fault diagnosis. According to the method, historical fault data, equipment operation data and external environment data are collected through event monitoring triggering and multi-source data fusion; an XGBoost model based on probability distribution is adopted to carry out precise cleaning and complementing on the data, and a comprehensive full-scene fault sample library is constructed; and further, a time sequence convolutional network and a graph convolutional neural network are utilized to deeply mine and fuse spatio-temporal characteristics such as periodicity, high-incidence time periods, geographical distribution rules and spatial relevance of fault events from a sample library. According to the method, spatio-temporal evolution law analysis and key influence factor identification of fault events can be efficiently and intelligently realized, the accuracy and timeliness of power grid fault research and judgment are remarkably improved, a scientific basis is provided for fault positioning and emergency disposal, and construction of a novel power system is assisted.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system fault diagnosis, in particular to a spatio-temporal feature extraction method and system for multi-type fault outage events based on an event-driven architecture. BACKGROUND

[0002] As a key link of the power system directly facing terminal users, the safe and stable operation of the distribution network is of great importance. With the expansion of the power grid and the increasing demand for power supply reliability from users, it has become a core requirement for intelligent operation and maintenance of the power grid to quickly and accurately analyze fault outage events of the distribution network. The analysis of outage events aims to analyze massive operation data, extract fault features, identify fault types, and mine their occurrence rules to provide decision support for fault handling and preventive maintenance, which usually involves processing and analysis techniques for large-scale, multi-dimensional data.

[0003] In the prior art, the analysis method for outage events of the distribution network usually relies on a single data processing model or analysis dimension. Some methods focus on using time series analysis models such as autoregressive integrated moving average models or recurrent neural networks to predict the occurrence probability of outage events or identify their periodicity. Other methods use spatial statistics or clustering algorithms to analyze the distribution characteristics of outage events in geographical locations to identify high-fault areas. In the data preprocessing stage, simple statistical methods or interpolation methods are usually used to process noise and missing values in the data.

[0004] However, the existing technical solutions have obvious deficiencies in practical application. First, the time characteristics and spatial characteristics are analyzed separately, ignoring the complex dynamic coupling relationship between them, making it difficult to truly reflect the spatio-temporal evolution process of faults in the power grid. Second, the existing models have poor adaptability to multi-source, heterogeneous, and high-noise data of the distribution network, and the data cleaning effect is not good, which limits the accuracy of feature extraction. In addition, most methods use offline batch processing mode, which lacks real-time response capability to sudden fault events, making it difficult to meet the demand of modern power grids for rapid fault handling. SUMMARY

[0005] To solve the above problems, the present application provides a spatio-temporal feature extraction method and system for multi-type fault outage events based on an event-driven architecture, which uses an XGBoost model based on probability distribution to clean multi-source data cooperatively, and deeply integrates time series convolutional networks and graph convolutional neural networks to extract spatio-temporal features, enabling real-time and accurate analysis and rule mining of fault outage events.

[0006] The above objectives can be achieved by the following solutions: The method for extracting spatiotemporal features of multi-type fault outage events of an event-driven architecture comprises: listening to fault outage event signals in a power distribution network; when an event occurs, triggering a data collection process; the data collection process is to collect multi-source fault outage event data of historical fault outage event data, equipment operation data and external environment data of the power distribution network to obtain an original data set; performing XGBoost model cleaning based on probability distribution on the original data set to identify and remove noise and outliers, complete missing data, and obtain a standard data set; based on the standard data set, constructing a power distribution network full-scenario fault outage sample library; based on the fault outage sample library, using a time series convolution network algorithm to extract time features of the fault outage event data, modeling and analyzing periodicity and high-occurrence periods of the fault outage event to obtain a time feature vector; based on the fault outage sample library, using a graph convolution neural network algorithm to extract spatial features of the fault outage event data, revealing geographical distribution rules and spatial correlations of different regions, equipment and lines to obtain a spatial feature vector; dynamically weighting and fusing the time feature vector and the spatial feature vector to obtain a comprehensive spatiotemporal feature vector; based on the comprehensive spatiotemporal feature vector, identifying spatiotemporal evolution rules and key influencing factors of the fault outage event.

[0007] Optionally, the method further comprises: setting an event listener to monitor event logs and alarm signals in a power distribution network monitoring system to obtain an event trigger signal; verifying the event trigger signal according to a verification rule for distinguishing real faults from false alarms to obtain a valid event; initializing data collection parameters based on the valid event to trigger the data collection process.

[0008] Optionally, the method further comprises: optimizing the fault outage sample library, wherein: using the evolution rules and key influencing factors to feed back and update the power distribution network full-scenario fault outage sample library; based on the updated fault outage sample library, retraining the time series convolution network and the graph convolution neural network to optimize feature extraction performance; iteratively performing the feedback updating and retraining steps to realize collaborative optimization of the fault outage sample library and the model.

[0009] Optionally, the XGBoost model cleaning based on probability distribution on the original data set to identify and remove noise and outliers and complete missing data comprises: using an XGBoost model to calculate importance scores of each feature in the original data set to obtain a feature importance list; according to the feature importance list, screening features with importance scores satisfying a predetermined screening condition to form a key feature subset; performing probability distribution estimation on the key feature subset to generate a joint probability distribution model; using the joint probability distribution model to detect and remove noise and outliers in the original data set while completing missing data to output a standard data set.

[0010] Optionally, the time feature extraction of the fault power outage event data based on the fault power outage sample library and using a time sequence convolution network algorithm comprises: obtaining time sequence data from the fault power outage sample library, including power outage event sequences in day, week, and season dimensions, to obtain time sequence input; performing causal convolution and dilated convolution processing on the time sequence input using a time sequence convolution network to capture long-term dependence, to obtain preliminary time features; performing threshold mechanism adjustment and residual connection optimization on the preliminary time features to enhance feature stability, and outputting a time feature vector.

[0011] Optionally, the space feature extraction of the fault power outage event data based on the fault power outage sample library and using a graph convolution neural network algorithm comprises: obtaining power distribution network topology structure data from the fault power outage sample library, including connection relationships of stations, lines, transformers, and users, to obtain graph structure input; performing neighbor node feature aggregation on the graph structure input using a graph convolution neural network to learn spatial correlation, to obtain preliminary space features; performing normalization processing and layer-by-layer propagation on the preliminary space features to expand the receptive field, and outputting a space feature vector.

[0012] Optionally, the dynamic weighted fusion of the time feature vector and the space feature vector comprises: obtaining a dynamic weight value, the dynamic weight value being used to reflect the influence degree of the time feature vector and the space feature vector on the fault power outage event; performing weighted summation on the time feature vector and the space feature vector using the dynamic weight value to obtain fused features; performing dimension reduction and standardization processing on the fused features to output a comprehensive spatio-temporal feature vector.

[0013] Optionally, the spatio-temporal evolution rule and key influence factors of the fault power outage event are identified based on the comprehensive spatio-temporal feature vector, comprising: obtaining the cause, influence degree, and recovery characteristics of the fault power outage event; performing rule mining on the cause, influence degree, and recovery characteristics and the comprehensive spatio-temporal feature vector; based on the rule mining result, revealing the evolution rules of different types of fault power outage events in time and space, and identifying key factors affecting the event.

[0014] Based on the same inventive concept, the application also provides a space-time feature extraction system for multi-type fault power outage events, comprising: an event monitoring and collecting module, configured to monitor fault power outage event signals in a power distribution network, and trigger a collection process to collect multi-source fault power outage event data of historical fault power outage event data, device operation data and external environment data of the power distribution network when an event occurs, to obtain an original data set; a data cleaning module, configured to clean the original data set based on an XGBoost model based on probability distribution, identify and remove noise and outliers, and complete missing data, to obtain a standard data set; a sample library construction module, configured to construct a full-scenario fault power outage sample library of the power distribution network based on the standard data set; a time feature extraction module, configured to extract time features of the fault power outage event data based on the fault power outage sample library using a time series convolution network algorithm, model and analyze periodicity and high-occurrence time periods of the fault power outage event, and obtain a time feature vector; a space feature extraction module, configured to extract space features of the fault power outage event data based on the fault power outage sample library using a graph convolution neural network algorithm, and reveal geographical distribution rules and spatial correlations of different regions, devices and lines, to obtain a space feature vector; a feature fusion module, configured to dynamically weight and fuse the time feature vector and the space feature vector, to obtain a comprehensive space-time feature vector; and a rule mining module, configured to identify space-time evolution rules and key influencing factors of the fault power outage event based on the comprehensive space-time feature vector.

[0015] Compared with the prior art, the application has the following advantages: 1. The application realizes real-time response and intelligent analysis of fault power outage events in a power distribution network by constructing an event-driven data processing architecture. When a fault event occurs, the parallel collection of multi-source data can be triggered immediately, and a series of automatic data cleaning, feature extraction and rule mining processes can be performed to provide immediate support for rapid fault diagnosis and decision-making, significantly improving the timeliness and initiative of fault handling and effectively shortening the power outage response time.

[0016] 2. The application combines time series convolution networks and graph convolution neural networks to deeply fuse and extract space-time features of power outage events. The time series model is responsible for mining the time evolution rules of the fault, and the graph model is responsible for revealing the spatial conduction relationship of the fault in the power grid topology. The two models work together to capture complex coupling patterns that are ignored by traditional single-dimensional analysis methods, greatly improving the accuracy and depth of fault analysis, and making the prediction of fault causes, impact range and development trend more reliable.

[0017] 3、The application has the ability of self-evolution and continuous learning through the design of a closed-loop optimization process containing feedback update and model retraining, which can continuously learn new knowledge from new fault events and use these knowledge to optimize the data sample library and analysis model, thereby ensuring that the method can continuously adapt to changes in power grid structure and the emergence of new fault modes, maintaining the long-term effectiveness and advancement of the technical solution.

[0018] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and attained by the structure particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0020] Figure 1 is a flowchart of the event-driven architecture-oriented spatiotemporal feature extraction method of the multi-type fault outage event of the embodiment of the present application.

[0021] Figure 2 is an annual time sequence distribution graph of the fault outage event of the embodiment of the present application.

[0022] Figure 3 is a spatial distribution heat map of the fault outage event of the embodiment of the present application.

[0023] Figure 4 is a structural schematic diagram of a spatiotemporal feature extraction system of a multi-type fault outage event of the embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0025] REFERENCE Figure 1An embodiment of the present application provides a method for extracting spatio-temporal features of multi-type fault outage events of an event-driven architecture, adopts an XGBoost model based on probability distribution to clean multi-source data, and deeply fuses a time series convolution network and a graph convolution neural network to extract spatio-temporal features, so that real-time and accurate judgment and law mining of the fault outage events can be realized, and the timeliness, accuracy and initiative of fault processing are significantly improved.

[0026] The method of the embodiment specifically comprises: Listen to the fault outage event signal in the power distribution network, and trigger a data collection process when the event occurs; The data collection process is to collect multi-source fault outage event data of historical fault outage event data, equipment operation data and external environment data of the power distribution network to obtain an original data set; The original data set is cleaned by an XGBoost model based on probability distribution to identify and remove noise and outliers, and to complete missing data to obtain a standard data set; Based on the standard data set, a full-scenario fault outage sample library of the power distribution network is constructed; Based on the fault outage sample library, a time series convolution network algorithm is used to extract time features of the fault outage event data, model and analyze the periodicity and high-incidence period of the fault outage events to obtain a time feature vector; Based on the fault outage sample library, a graph convolution neural network algorithm is used to extract spatial features of the fault outage event data to reveal the geographical distribution law and spatial correlation of different regions, equipment and lines to obtain a spatial feature vector; The time feature vector and the spatial feature vector are dynamically weighted and fused to obtain a comprehensive spatio-temporal feature vector; Based on the comprehensive spatio-temporal feature vector, the spatio-temporal evolution law of the fault outage event and its key influencing factors are identified.

[0027] Specifically, by listening to the fault outage event signal in the power distribution network, the data collection process is triggered. Historical fault outage event data, equipment operation data and external environment data are collected to obtain a multi-source original data set. Based on the original data set, XGBoost model is used for accurate cleaning and completion to build a comprehensive power distribution network full-scenario fault outage sample library. This sample library is dynamic, which combines static fault history data with real-time changing equipment and environment parameters to provide high-quality benchmarks for subsequent analysis. Based on the sample library, time series convolution network and graph convolution neural network are used to deeply mine and dynamically weight and fuse the periodicity of fault outage events and the high-occurrence period and geographical distribution of different regions, equipment and lines. The seemingly independent spatio-temporal information is effectively integrated to generate a comprehensive spatio-temporal feature vector that can fully represent the fault event. Based on the comprehensive spatio-temporal feature vector, the spatio-temporal evolution law of the fault outage event and its key influencing factors are identified. These spatio-temporal features are used as decision basis to reveal the evolution patterns of different types of fault events in time and space, thereby realizing continuous, dynamic and adaptive prediction and analysis of faults and improving the operation efficiency of the power grid.

[0028] Optionally, the listening to the fault outage event signal in the power distribution network comprises: An event listener is set to monitor the event log and alarm signal in the power distribution network monitoring system to obtain an event trigger signal; Specifically, by setting an event listener, the event log and alarm signal from the power distribution network monitoring system are monitored in real time to obtain initial signals reflecting abnormal conditions of the power grid, including but not limited to device failure, voltage out-of-limit or current surge, etc., thereby obtaining an event trigger signal.

[0029] Illustratively, an event listener is set to monitor the log sent by the power distribution network monitoring center in real time. The log shows that a short circuit occurs in a line in a certain area, i.e., an event trigger signal is obtained.

[0030] According to the verification rule for distinguishing real faults from false alarms, the event trigger signal is verified to obtain an effective event; Specifically, the event trigger signal is verified according to the preset verification rule to exclude false alarms and transient interference. The verification rule can be a logic based on threshold, duration or multi-signal cooperative judgment. If the event trigger signal meets the verification rule, the event is determined to be a real fault, and an effective event is obtained.

[0031] Illustratively, suppose the verification rule is set as: when the voltage of a line is continuously lower than the safety threshold and exceeds 500 milliseconds, it is considered as an effective event that needs to be handled.

[0032] Based on the effective event, the data collection parameters are initialized to trigger the data collection process.

[0033] Specifically, based on the fault information contained in the effective event, the data collection parameters are initialized. These parameters include the time, location, type of fault occurrence, and affected device number, etc. Subsequently, these parameters are used to trigger the data collection process, starting the collection of raw data.

[0034] Illustratively, based on the effective event, the fault line number, fault occurrence time and fault type are taken as the parameters of data collection, and the data collection process is initialized to collect the relevant data of the line before and after the fault occurrence.

[0035] Optionally, the method further comprises: optimization of the fault outage sample library, wherein: Using the evolution law and key influencing factors, the power distribution network full-scenario fault outage sample library is updated by feedback; Specifically, using the evolution law and key influencing factors obtained from the analysis, these information are taken as feedback to update the labels and metadata of the corresponding fault events in the power distribution network full-scenario fault outage sample library. This ensures the continuous accuracy and effectiveness of the sample library.

[0036] Illustratively, when the algorithm identifies that the fault frequency of a certain type of line significantly increases under extreme high temperature, "high temperature weather" is added as a key influencing factor to the metadata of all related historical fault events, and the labels are updated accordingly.

[0037] Based on the updated fault outage sample library, the time series convolutional network and the graph convolutional neural network are retrained to optimize the feature extraction performance; Specifically, based on the updated fault outage sample library, the time series convolutional network and the graph convolutional neural network are retrained to optimize the feature extraction performance. This process enables the model to adapt to new fault patterns and data features, thereby improving the overall research and judgment ability.

[0038] Illustratively, using the newly fed 50 fault sample data containing the latest extreme weather influence, the time series convolutional network and the graph convolutional neural network are retrained, so that the prediction accuracy of the model for similar fault scenarios is improved by about 5%.

[0039] The feedback updating and retraining steps are iteratively performed to realize the collaborative optimization of the fault outage sample library and the model.

[0040] Specifically, by iteratively performing the above feedback updating and retraining steps, the collaborative optimization of the fault outage sample library and the model is realized. This process forms a closed-loop adaptive mechanism, enabling data and model to learn from each other and co-evolve, ensuring the long-term effectiveness and advancement of the technical solution.

[0041] Exemplarily, the sample library updating and model retraining iteration is automatically performed every other week or when the newly collected fault data accumulates more than 100, ensuring that the model always remains adaptive to the latest state of the power grid.

[0042] Optionally, the original data set is subjected to XGBoost model cleaning based on probability distribution, identifying and removing noise and outliers, and filling in missing data, including: The importance scores of each feature in the original data set are calculated using the XGBoost model to obtain a feature importance list. Specifically, the built-in feature importance analysis capability of the XGBoost model is used to train and evaluate all features in the original data set, calculate the contribution of each feature to fault analysis, and obtain a feature importance list reflecting the importance of each feature.

[0043] Exemplarily, the XGBoost model analysis result shows that the importance scores of voltage, current mutation amplitude, and meteorological conditions in the region where the fault point is located are much higher than historical maintenance records and device models, so the top three are considered as high importance features.

[0044] According to the feature importance list, features with importance scores meeting predetermined filtering conditions are screened to form a key feature subset. Specifically, according to the feature importance list, a predetermined filtering condition is set, for example, the importance score is greater than 0.80, and the features with the greatest contribution to fault analysis are screened to form a compact and efficient key feature subset, removing redundant and irrelevant features.

[0045] Exemplarily, the features of fault current, fault time, and fault type with importance scores higher than 0.80 are screened out from the original data set to form a key feature subset containing core information.

[0046] The key feature subset is subjected to probability distribution estimation to generate a joint probability distribution model. Specifically, the key feature subset is subjected to probability distribution estimation to capture the mutual relationship and distribution law among these features, thereby generating a joint probability distribution model that can describe the normal data pattern.

[0047] Exemplarily, the key feature subset containing voltage, current, and temperature is subjected to multivariate Gaussian probability distribution estimation to establish a model describing the joint distribution of these variables in normal operation state, which can be used for subsequent abnormal data detection.

[0048] The joint probability distribution model is used to detect and remove noise and outliers in the original data set, and to fill in missing data, outputting a standard data set.

[0049] Specifically, by calculating the probability density of each data point under the model, the joint probability distribution model is used to detect and remove noise and outliers with extremely low probability. At the same time, for the missing data in the data set, the joint probability distribution model is used for interpolation, and finally a high-quality standard data set is obtained.

[0050] Exemplarily, by using the joint probability distribution model, it is found that some voltage readings have a probability of less than 0.001 compared with the normal data mode, so these readings are determined as outliers and removed. At the same time, for the missing temperature data, it is calculated according to the joint distribution with voltage and current to complete the data.

[0051] Optionally, based on the fault power failure sample library, the time sequence convolution network algorithm is used to extract time features of the fault power failure event data, comprising: Obtain time sequence data from the fault power failure sample library, including power failure event sequences in day, week and season dimensions, to obtain time sequence input; Specifically, time sequence data is obtained from the fault power failure sample library, which includes power failure event sequences in day, week and season dimensions. The occurrence frequency, duration and high incidence period of the event in different time scales are fully reflected, so as to obtain time sequence input for time sequence modeling, which can fully represent the dynamic characteristics of the fault event changing with time.

[0052] Exemplarily, the number of power failures per day or the total number of high-temperature power failure events per month in a certain area in the past two years is obtained as time sequence input.

[0053] The time sequence input is processed by using the time sequence convolution network to capture long-term dependence relationship, and preliminary time features are obtained; Specifically, the time sequence input is processed by using the time sequence convolution network to capture long-term dependence relationship, and preliminary time features are obtained; , Wherein, represents the dilated convolution, is a specific time point, is an input vector, is a convolution kernel size of convolution function, It is the expansion factor that controls the magnitude of the expansion. Through these processes, preliminary temporal characteristics reflecting the laws of temporal evolution can be obtained. For example... Figure 2 As shown, the number of power outage events occurring daily throughout the year is displayed, with obvious seasonal variations and periodic peak periods. These temporal characteristics form the basis for deep analysis by temporal convolutional network algorithms.

[0054] For example, a convolution kernel of size 2 and a dilation factor are used. When calculating a specific point in time When outputting, the convolution kernel will act on the input vector. of and Data from two locations is used to capture event features that are far apart.

[0055] The initial time features are adjusted using a threshold mechanism and optimized using residual connections to enhance feature stability and output a time feature vector.

[0056] Specifically, a threshold mechanism is applied to the initial temporal features to enhance the network's modeling ability, and residual connections are introduced to optimize the network structure. Residual connections help alleviate the vanishing or exploding gradient problems during deep network training, thereby enhancing the stability of the features. Through these optimization steps, the output temporal feature vectors can be ensured to have higher robustness and reliability.

[0057] For example, after the initial time features are processed by the threshold mechanism, the processing result is added to the original initial time features through residual connections, so that the network can learn incremental changes while retaining the original feature information, thereby ensuring the stability of the features.

[0058] Optionally, based on the power outage sample library, spatial feature extraction of the power outage event data using a graph convolutional neural network algorithm includes: The power outage sample database is used to obtain distribution network topology data, including the connection relationships of stations, lines, transformers, and customers, to obtain the graph structure input; Specifically, distribution network topology data is obtained from a fault outage sample database, including the connection relationships of various stations, lines, transformers, and customers in the power grid, as well as multi-source heterogeneous information such as associated fault types, geographical coordinates, and equipment attributes. This data is then used to represent the power grid structure as a graph, providing the graph structure input for spatial analysis. The graph structure input can comprehensively reflect the spatial connectivity characteristics and multi-dimensional event information of the power grid.

[0059] For example, the connection relationships between a substation and its 10 connected lines, as well as the connection relationships between these lines and downstream transformers and users, are obtained to obtain a graph structure input describing the power grid topology.

[0060] The graph structure input is aggregated with neighbor node features using a graph convolutional neural network to learn spatial correlation and obtain preliminary spatial features. Specifically, the graph structure input is aggregated with neighbor node features using a graph convolutional neural network to learn the spatial correlation between power grid nodes. The core idea is to simplify the graph convolution operation to a first-order approximation, and capture the propagation relationship of faults in the power grid topology by aggregating the feature information of each center node and its directly connected neighbor nodes. The core operation is represented as: wherein, is the feature vector matrix of the j-th layer node, is the adjacency matrix with a self-loop, is its diagonal degree matrix, is a learnable weight matrix, is an activation function. This operation obtains preliminary spatial features reflecting local spatial correlation by weighted sum of neighbor node features of each node. As shown in , the density distribution of the fault outage event in geographical space is intuitively shown in the form of a heat map, wherein the darker the color, the higher the fault incidence in the area. These spatial distribution rules are the key to learning spatial correlation by the graph convolutional neural network algorithm. Figure 3

[0061] For example, a transformer node aggregates fault features such as average outage duration and fault frequency from adjacent line nodes and user nodes, and uses these aggregated features as its preliminary spatial feature vector.

[0062] The preliminary spatial features are normalized and propagated layer by layer to expand the receptive field and output a spatial feature vector.

[0063] Specifically, the preliminary spatial features are normalized and propagated layer by layer to avoid numerical instability and gradient problems. By stacking multiple graph convolution layers, the neighborhood information can be propagated to further nodes, thereby effectively expanding the receptive field and enabling the model to capture spatial correlation in a wider range. Finally, a spatial feature vector is output, which can comprehensively reflect the geographical distribution and spatial conduction rules of fault events.

[0064] For example, a two-layer graph convolution network is used, the first layer aggregates the features of directly connected neighbor nodes, and the second layer aggregates the features of two-hop neighbors, thereby expanding the receptive field from a single line to multiple adjacent lines, and thus capturing spatial correlation in a wider range.

[0065] ​​​Optionally, the dynamic weighting fusion of the time feature vector and the space feature vector comprises: obtaining a dynamic weight value, the dynamic weight value being used to reflect the influence degree of the time feature vector and the space feature vector on the fault outage event; Specifically, a special attention network is used to automatically learn and obtain the dynamic weight value. The attention network can adaptively assign the weight of the time feature vector and the space feature vector according to the context information such as the fault outage event type, occurrence time and location, and accurately reflect the relative influence degree of each dimension feature on the fault in different scenarios. The weight value can provide a scientific basis for subsequent feature fusion.

[0066] For example, for a fault outage event caused by extreme weather such as typhoon, since its time evolution law is more significant, such as high incidence in a specific season and period, the attention network will assign a higher weight to the time feature vector, for example, 0.7, while the weight of the space feature vector is 0.3.

[0067] using the dynamic weight value to perform weighted summation on the time feature vector and the space feature vector to obtain a fusion feature; Specifically, the time feature vector and the space feature vector are weighted and summed using the obtained dynamic weight value, so as to obtain a fusion feature that integrates the space-time information. The mathematical expression of this process is represented as: , wherein, is the fusion feature, is the time feature vector, is the space feature vector, and are the corresponding dynamic weight values, and .

[0068] For example, assuming that the time feature vector is , the space feature vector is , and the dynamic weight values are and , respectively. Then the fusion feature is calculated as: .

[0069] dimension reduction and standardization processing are performed on the fusion feature to output a comprehensive space-time feature vector.

[0070] Specifically, the fusion features are reduced in dimension to eliminate redundant information in the data and reduce the computational complexity of subsequent models. Then, the reduced features are standardized, such as Z-Score standardization, to ensure that all features have the same scale, thereby eliminating the influence of different feature units and dimensions on the performance of the model, and finally output a high-quality comprehensive spatio-temporal feature vector.

[0071] For example, first, a 200-dimensional fusion feature is reduced to 50 dimensions using the principal component analysis method. Then, the 50-dimensional features are standardized so that the mean of each feature is 0 and the variance is 1.

[0072] Optionally, based on the comprehensive spatio-temporal feature vector, the spatio-temporal evolution law of the fault outage event and its key influencing factors are identified. Obtain the cause, impact degree, and recovery characteristics of the fault outage event; Specifically, from a pre-constructed database, the cause, impact degree, and recovery characteristics of each fault outage event are obtained. These features are non-quantitative attributes of the event, for example, the cause can be "external damage" or "equipment aging", the impact degree can be divided into "mild" or "severe", and the recovery characteristics can be described as "fast recovery" or "long-term power outage". Through these attributes, it aims to provide higher-dimensional analysis insights and fully represent the nature of the event.

[0073] For example, the cause of a certain fault outage event is "external damage", the impact degree is "moderate", and the recovery characteristic is "fast".

[0074] Regularly mine the cause, impact degree, and recovery characteristics with the comprehensive spatio-temporal feature vector; Specifically, the cause, impact degree, and recovery characteristics are regularly mined with the comprehensive spatio-temporal feature vector. It aims to explore the internal relationship between the spatio-temporal characteristics of the event and the essential attributes of the event. By calculating the correlation between the two, the relationship is quantified, thereby revealing the deep connection between seemingly independent event attributes and spatio-temporal characteristics.

[0075] For example, the correlation between the "high season" time feature of a certain region and the "lightning" fault cause is calculated. When the correlation coefficient reaches 0.8 or above, it indicates that there is a strong correlation between the two, and it can be inferred that lightning faults in the region are high in certain seasons.

[0076] Based on the regular mining results, the evolution law of different types of fault outage events in time and space is revealed, and the key factors affecting the event are identified.

[0077] Specifically, based on the regularity mining result, the spatio-temporal features and event attributes with strong correlation are identified, and then the evolution rules of different types of fault outage events in time and space are revealed. By matching the cause such as "external damage" with the spatio-temporal feature such as "high incidence of specific regional in summer afternoon", the key factors affecting the event can be accurately identified, and scientific basis can be provided for fault handling and prevention strategies.

[0078] For example, the regularity mining result shows that the power outage events caused by "external damage" have strong correlation with specific "urban construction area" and "weekday daytime period", thereby revealing the spatio-temporal evolution rule of such faults and identifying "construction activity" as the key influencing factor.

[0079] Based on the same inventive concept, as shown in Figure 4 The present application also provides a spatio-temporal feature extraction system for multi-type fault outage events, which comprises: An event monitoring and collecting module is configured to monitor the fault outage event signals in the power distribution network, and when an event occurs, trigger the collection process to collect the multi-source fault outage event data of the historical fault outage event data, equipment operation data and external environment data of the power distribution network, and obtain an original data set; A data cleaning module is configured to clean the original data set based on the XGBoost model based on probability distribution, identify and remove noise and outliers, and complete missing data, to obtain a standard data set; A sample library construction module is configured to construct a full-scenario fault outage sample library of the power distribution network based on the standard data set; A time feature extraction module is configured to extract the time features of the fault outage event data based on the fault outage sample library using the time series convolution network algorithm, model and analyze the periodicity and high-incidence period of the fault outage event, and obtain a time feature vector; A spatial feature extraction module is configured to extract the spatial features of the fault outage event data based on the fault outage sample library using the graph convolution neural network algorithm, reveal the geographical distribution rule and spatial correlation of different regions, equipment and lines, and obtain a spatial feature vector; A feature fusion module is configured to dynamically weight and fuse the time feature vector and the spatial feature vector, and obtain a comprehensive spatio-temporal feature vector; A regularity mining module is configured to identify the spatio-temporal evolution rule of the fault outage event and its key influencing factors based on the comprehensive spatio-temporal feature vector.

[0080] It should be noted that the electrical connection between the various units described above does not necessarily indicate a direct connection, and the indirect connection mode can also be applied to the embodiments of the present application as long as the purpose of the present application is achieved. The above is only an exemplary embodiment of the present application, and cannot limit the scope of the present application.

[0081] That is, any equivalent changes and modifications made in accordance with the teachings of the present application are still within the scope of the present application. Other embodiments of the present application will be readily apparent to those skilled in the art upon considering the description and practice of the true principles of the disclosure. The present application is intended to cover any variations, uses, or adaptive changes of the present application that follow the general principles of the present application and include common knowledge or conventional techniques in the art that are not described in the present application.

Claims

1. A method for extracting spatiotemporal features of multiple types of power outage events for event-driven architecture, characterized in that, The method includes: Monitor power outage event signals in the power distribution network, and trigger the data acquisition process when the event occurs; The data acquisition process involves collecting multi-source fault outage event data, including historical fault outage event data of the power distribution network, equipment operation data, and external environment data, to obtain the raw dataset. The original dataset is cleaned using an XGBoost model based on probability distribution to identify and remove noise and outliers, and to fill in missing data to obtain a standard dataset. Based on the aforementioned standard dataset, a full-scenario power outage sample library for power distribution networks will be constructed. Based on the aforementioned power outage sample database, a temporal convolutional network algorithm is used to extract time features from the power outage event data, model and analyze the periodicity and high-incidence periods of the power outage events, and obtain a time feature vector. Based on the aforementioned power outage sample database, spatial features of the power outage event data are extracted using a graph convolutional neural network algorithm to reveal the geographical distribution patterns and spatial correlations of different regions, equipment, and lines, thereby obtaining spatial feature vectors. The temporal and spatial feature vectors are dynamically weighted and fused to obtain a comprehensive spatiotemporal feature vector; Based on the comprehensive spatiotemporal feature vector, the spatiotemporal evolution pattern of the power outage event and its key influencing factors are identified.

2. The spatiotemporal feature extraction method for multiple types of power outage events oriented towards an event-driven architecture as described in claim 1, characterized in that, The monitored outage event signals in the power distribution network include: Set up an event listener to monitor the event logs and alarm signals in the power distribution network monitoring system and obtain event trigger signals; The event trigger signal is verified according to the verification rules used to distinguish between real faults and false alarms to obtain valid events; Based on the valid event, initialize the data acquisition parameters and trigger the data acquisition process.

3. The spatiotemporal feature extraction method for multiple types of power outage events oriented towards an event-driven architecture according to claim 1, characterized in that, The method further includes: optimizing the fault outage sample library, wherein: The aforementioned evolutionary patterns and key influencing factors are used to update the power outage sample database for all scenarios of the distribution network. Based on the updated fault outage sample database, the temporal convolutional network and graph convolutional neural network were retrained to optimize feature extraction performance. The feedback update and retraining steps are executed iteratively to achieve collaborative optimization of the fault outage sample library and the model.

4. The spatiotemporal feature extraction method for multiple types of power outage events oriented towards an event-driven architecture according to claim 1, characterized in that, The original dataset is cleaned using an XGBoost model based on probability distribution to identify and remove noise and outliers, and to complete missing data, including: The importance score of each feature in the original dataset is calculated using the XGBoost model to obtain a list of feature importance. Based on the feature importance list, features whose importance scores meet the predetermined filtering conditions are selected to form a subset of key features; The probability distribution of the key feature subset is estimated to generate a joint probability distribution model; Using the joint probability distribution model, noise and outliers in the original dataset are detected and removed, while missing data is filled in to output a standard dataset.

5. The spatiotemporal feature extraction method for multiple types of power outage events oriented towards an event-driven architecture according to claim 1, characterized in that, Based on the aforementioned power outage sample database, the temporal feature extraction of the power outage event data using a temporal convolutional network algorithm includes: Time series data, including power outage event sequences in daily, weekly, and seasonal dimensions, are obtained from the power outage sample library to obtain time series input. The time series input is processed by a temporal convolutional network to perform causal convolution and dilated convolution to capture long-term dependencies and obtain preliminary temporal features. The initial time features are adjusted using a threshold mechanism and optimized using residual connections to enhance feature stability and output a time feature vector.

6. The spatiotemporal feature extraction method for multiple types of power outage events oriented towards an event-driven architecture according to claim 1, characterized in that, Based on the aforementioned power outage sample database, spatial feature extraction of the power outage event data using a graph convolutional neural network algorithm includes: The power outage sample database is used to obtain distribution network topology data, including the connection relationships of stations, lines, transformers, and customers, to obtain the graph structure input; The graph convolutional neural network is used to aggregate the neighbor node features of the graph structure input, learn spatial correlation, and obtain preliminary spatial features. The preliminary spatial features are normalized and propagated in layers to expand the receptive field and output a spatial feature vector.

7. The spatiotemporal feature extraction method for multiple types of power outage events oriented towards an event-driven architecture according to claim 1, characterized in that, The dynamic weighted fusion of the temporal and spatial feature vectors includes: Obtain dynamic weight values, which are used to reflect the degree of influence of the time feature vector and the spatial feature vector on the power outage event; The time feature vector and the spatial feature vector are weighted and summed using the dynamic weight values ​​to obtain the fused features; The fused features are then subjected to dimensionality reduction and standardization to output a comprehensive spatiotemporal feature vector.

8. The spatiotemporal feature extraction method for multiple types of power outage events oriented towards an event-driven architecture according to claim 1, characterized in that, Based on the comprehensive spatiotemporal feature vector, the spatiotemporal evolution pattern and key influencing factors of the power outage event are identified, including: Obtain the cause, impact, and recovery characteristics of the power outage event; The causes, degree of impact, and recovery characteristics are compared with the comprehensive spatiotemporal feature vector to extract patterns. Based on the aforementioned pattern mining results, the evolution patterns of different types of power outage events in time and space are revealed, and key factors affecting the events are identified.

9. A spatiotemporal feature extraction system for multi-type power outage events oriented towards an event-driven architecture, characterized in that, The system is used for the spatiotemporal feature extraction method of multi-type power outage events oriented towards an event-driven architecture as described in any one of claims 1-8, and the system comprises: The event listening and acquisition module is used to listen for fault outage event signals in the distribution network. When an event occurs, it triggers the acquisition process to collect multi-source fault outage event data, including historical fault outage event data, equipment operation data, and external environment data of the distribution network, to obtain the raw dataset. The data cleaning module is used to perform XGBoost model cleaning based on probability distribution on the original dataset, identify and remove noise and outliers, fill in missing data, and obtain a standard dataset. The sample library construction module is used to construct a full-scenario power outage sample library for the power distribution network based on the standard dataset. The time feature extraction module is used to extract time features from the power outage event data based on the power outage sample library using a temporal convolutional network algorithm, model and analyze the periodicity and high-incidence periods of the power outage events, and obtain time feature vectors. The spatial feature extraction module is used to extract spatial features from the power outage event data based on the power outage sample library using a graph convolutional neural network algorithm, revealing the geographical distribution patterns and spatial correlations of different regions, equipment and lines, and obtaining spatial feature vectors. The feature fusion module is used to dynamically weight and fuse the temporal feature vector and the spatial feature vector to obtain a comprehensive spatiotemporal feature vector. The pattern mining module is used to identify the spatiotemporal evolution pattern and key influencing factors of the power outage event based on the comprehensive spatiotemporal feature vector.

Images