Power failure studying and judging processing method based on big data

Abstract

The invention relates to a power failure studying and judging processing method based on big data. The method comprises the following steps: carrying out data time sequence analysis on weather monitoring data by adopting an artificial intelligence technology based on deep learning, capturing local time sequence change characteristics of the weather data, and then carrying out association optimization and time sequence aggregation on the weather data time sequence change characteristics in each local time domain, so as to obtain weather data time sequence change characteristics; and mining a time sequence dynamic change mode of the weather data in a global time domain so as to carry out intelligent power failure early warning. Thus, the accuracy and timeliness of power failure early warning can be effectively improved, potential power failure risks can be found in advance, and corresponding prevention measures can be taken conveniently.

Description

Technical Field

[0001] The present application relates to the technical field of power outage analysis and judgment, and specifically, to a power outage analysis and processing method based on big data. Background Art

[0002] Electricity supply is an indispensable part of modern life. Any unexpected power outage will bring inconvenience and even serious consequences to social economy and personal life. The normal operation of the power system is affected by many factors, among which weather conditions are a very important external factor. Under severe weather conditions such as thunderstorms, typhoons, hail and other severe weather conditions, power grid facilities may suffer varying degrees of damage, resulting in large-scale power outages. Therefore, accurately predicting the impact of extreme weather on the power grid and making emergency preparations in advance are of great significance to reducing the negative impact of power outages.

[0003] Traditional power outage warning methods mainly rely on weather forecasts and historical data of the power grid, and make predictions through expert experience and some simple statistical methods. Although this method can play a warning role to a certain extent, it is difficult to predict the risk of power outages in real time and accurately because it relies on manual analysis. In addition, due to the lack of in-depth mining and analysis of weather data, its warning effect is often unsatisfactory.

[0004] Therefore, we look forward to an intelligent, big data-based method for analyzing and handling power outages. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present application provides a method for analyzing and processing power outages based on big data, the method comprising: Obtaining a time queue of weather monitoring data, wherein the weather monitoring data includes temperature value, humidity value, wind speed value and rainfall; Extracting time series features based on a local time scale from the time queue of the weather monitoring data to obtain a sequence of local time series correlation feature vectors of the weather data; Performing feature association optimization on the sequence of local time series associated feature vectors of weather data to obtain a sequence of local time series associated optimized feature vectors of weather data; Based on the feature significance time series aggregation information of the sequence of the local time series association optimization feature vector of the weather data, it is determined whether to generate a power outage warning prompt.

[0007] Optionally, the time queue of the weather monitoring data is subjected to temporal feature extraction based on a local time scale to obtain a sequence of local time series associated feature vectors of the weather data, including: using a weather data embedding matrix to embed code each weather monitoring data in the time queue of the weather monitoring data to obtain a time queue of weather monitoring embedded coding vectors; dividing the time queue of weather monitoring embedded coding vectors into a predetermined time scale and inputting the resultant into a weather local time series feature capturer based on a 1D-CNN model to obtain a sequence of local time series associated feature vectors of the weather data.

[0008] Optionally, feature association optimization is performed on the sequence of local time series correlation feature vectors of weather data to obtain a sequence of local time series correlation optimized feature vectors of weather data, including: inputting the sequence of local time series correlation feature vectors of weather data into an adaptive feature optimization module based on sequence correlation semantics to obtain the sequence of local time series correlation optimized feature vectors of weather data.

[0009] Optionally, the sequence of the local time series correlation feature vectors of weather data is input into an adaptive feature optimization module based on sequence correlation semantics to obtain a sequence of the local time series correlation optimization feature vectors of weather data, including: calculating the semantic correlation score vector between any two local time series correlation feature vectors of weather data in the sequence of local time series correlation feature vectors of weather data to obtain a sequence of local time series semantic correlation score vectors of weather data; calculating the mean vector of the sequence of local time series semantic correlation score vectors of weather data to obtain a global representation vector of endogenous correlation of the sequence; calculating the correlation optimization factor of each local time series correlation feature vector of weather data in the sequence of local time series correlation feature vectors of weather data based on the global representation vector of endogenous correlation of the sequence to obtain a sequence of correlation optimization factors; and inputting the sequence of correlation optimization factors into a module for adaptive feature optimization based on sequence correlation semantics. An activation function is used to obtain a sequence of association optimization weight factors; each association optimization weight factor in the sequence of association optimization weight factors is used as a weight to weight each local time series association feature vector of weather data in the sequence of local time series association feature vectors of weather data to obtain a sequence of local time series association optimization feature vectors of weather data.

[0010] Optionally, the semantic association score vector between any two local time series association feature vectors of weather data in the sequence of local time series association feature vectors of weather data is calculated to obtain a sequence of local time series semantic association score vectors of weather data, including: cascading any two local time series association feature vectors of weather data in the sequence of local time series association feature vectors of weather data, multiplying them by a weight coefficient matrix, and then dot-adding them with a bias vector to obtain the local time series semantic association score vector of weather data.

[0011] Optionally, based on the global representation vector of the intrinsic correlation of the sequence, the correlation optimization factor of each local time series correlation feature vector of weather data in the sequence of local time series correlation feature vectors of weather data is calculated to obtain a sequence of correlation optimization factors, including: multiplying the local time series correlation feature vector of weather data and the global representation vector of the intrinsic correlation of the sequence by different weight coefficient vectors and then performing an addition operation to obtain a semantic correlation coefficient; adding a bias parameter to the semantic correlation coefficient and then passing it through a sigmoid activation function to obtain the correlation optimization factor.

[0012] Optionally, based on the feature significance time series aggregation information of the sequence of local time series association optimization feature vectors of the weather data, determining whether to generate a power outage warning prompt includes: inputting the sequence of local time series association optimization feature vectors of the weather data into a node time series aggregation response network modulated by feature significance to obtain a weather data time series significant aggregation representation vector; inputting the weather data time series significant aggregation representation vector into a classifier-based judgment result generator to obtain a judgment result, and the judgment result is used to indicate whether to generate the power outage warning prompt.

[0013] Optionally, the sequence of the weather data local time series association optimization feature vectors is input into a node time series aggregation response network modulated by feature significance to obtain a weather data time series significant aggregation representation vector, including: calculating the feature significance description factor of each weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector; taking the last weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector as the current weather data local time series association optimization feature vector, and calculating the feature significance description factor of each weather data local time series association optimization feature vector based on the comparison of the other weather data local time series association optimization feature vectors in the sequence of the weather data local time series association optimization feature vector with the current weather data local time series association optimization feature vector. The distance span between the local temporal association optimization feature vectors of weather data is used to construct the feature significance attenuation factor of each other local temporal association optimization feature vector of weather data; the product between the feature significance attenuation factor of each other local temporal association optimization feature vector of weather data and its feature significance description factor is calculated to obtain a sequence of feature significance attenuation description factors; the sequence of feature significance attenuation description factors is input into the gated mask module to obtain a sequence of feature significance attenuation weight factors; based on the sequence of feature significance attenuation weight factors, the weighted sum of the sequence of local temporal association optimization feature vectors of weather data is calculated to obtain the temporal significant aggregation representation vector of weather data.

[0014] Optionally, the characteristic significance description factor of each local time series association optimization feature vector of weather data in the sequence of the local time series association optimization feature vector of weather data is calculated, including: calculating the expected value of the fourth power of the difference between each eigenvalue in the local time series association optimization feature vector of weather data and its characteristic mean, and dividing the expected value by the square of the characteristic variance of the local time series association optimization feature vector of weather data to obtain the characteristic significance description factor.

[0015] Optionally, based on the distance span between each other local time series association optimization feature vector of weather data in the sequence of local time series association optimization feature vector of weather data and the current local time series association optimization feature vector of weather data, a feature significance attenuation factor of each other local time series association optimization feature vector of weather data is constructed, including: calculating the difference between the maximum eigenvalue of the current local time series association optimization feature vector of weather data and the maximum eigenvalue of each other local time series association optimization feature vector of weather data in the sequence of local time series association optimization feature vector of weather data, and then dividing the difference by the number of eigenvectors separated therefrom to obtain the feature significance attenuation factor of each other local time series association optimization feature vector of weather data.

[0016] By adopting the above technical solution, the weather monitoring data is analyzed in time series by using artificial intelligence technology based on deep learning, and the local time series change characteristics of the weather data are captured. Then, by performing correlation optimization and time series aggregation on the time series change characteristics of the weather data in each local time domain, the time series dynamic change pattern of the weather data in the global time domain is mined, so as to perform intelligent power outage warning. In this way, the accuracy and timeliness of power outage warning can be effectively improved, which helps to discover potential power outage risks in advance so as to take corresponding preventive measures.

[0017] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings: Figure 1 The present invention is a flowchart of a method for analyzing and processing power outages based on big data according to an exemplary embodiment.

[0019] Figure 2 It is a block diagram of a power outage analysis and processing system based on big data according to an exemplary embodiment.

[0020] Figure 3 It is a logical model diagram of a data analysis method according to an exemplary embodiment.

[0021] Figure 4 The present invention is a flowchart of issuing warning analysis according to an exemplary embodiment.

[0022] Figure 5 The figure is a schematic diagram showing the relationship between different weather types and the number of users experiencing power outages according to an exemplary embodiment.

[0023] Figure 6 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0024] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0025] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0026] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0027] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0028] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0029] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0030] The extreme rainstorm disaster has brought severe challenges to power supply and grassroots disaster response. The disaster caused widespread power outages. Due to various reasons, the power outage information could not be reported in a timely and accurate manner, which seriously affected the accurate assessment of the power outage areas and users. As the "eyes" of power marketing, the data collection system was severely damaged by the shutdown of operator base stations and flooding of underground distribution rooms. The terminal online rate dropped to 63.5% at the lowest. Power outages in more than one-third of the distribution areas could not be reported to the system master station in time, which seriously affected the accuracy of the judgment of the power outage areas and users. In the face of the disaster, how to reasonably dispatch emergency repairs in severe weather, quickly grasp the scope of the power outage, determine the type of power failure, and accurately locate the fault location is a great test.

[0031] At the same time, due to the interruption of base station communication, the power outage information could not be reported in time, further exacerbating the problem of being unable to accurately lock the power outage area and the progress of emergency repairs. The existence of these problems highlights the urgency of the insufficient emergency response capabilities of customer service. In the face of extreme disaster weather, it is urgent to find out the problem through in-depth analysis and comprehensive review of the incident. In this way, relevant technical research can be carried out in a targeted manner, the emergency response capabilities of marketing customer services can be improved, the efficiency of disaster response can be further improved, and the workload of grassroots units can be reduced. In this disaster, the stable supply and disaster response capabilities of the power system are particularly critical. Only through technical research and innovation can power supply personnel better respond to similar disasters and ensure the safety of people's lives and property.

[0032] Therefore, it is urgent to strengthen the research and application of relevant technologies to improve the efficiency and accuracy of disaster emergency response in order to cope with the challenges of disasters that may arise in the future. This will also provide better support for grassroots units, enabling them to respond to disasters more effectively and protect people's lives and property interests.

[0033] In order to solve these problems, it is necessary to analyze and review the incidents to find out the problem of insufficient emergency response of customer service under extreme disaster weather. At present, it is urgent to carry out relevant technical research to improve the emergency response capability of marketing customer service, improve the efficiency of disaster response, and reduce the workload of grassroots units. In this disaster, the stable supply of the power system and the improvement of disaster response capabilities are crucial. Only through technological research and innovation can we better respond to similar disasters and protect the safety of people's lives and property. Therefore, strengthening the research and application of relevant technologies and improving the efficiency and accuracy of disaster emergency response are the current urgent problems to be solved. In the face of these problems, in the research and application of key digital technologies for emergency power restoration in residential areas, the technical route based on "minimized accurate acquisition + digital system calculation recommendation" is adopted to build a power outage analysis model and a power outage prediction model to further improve the ability in customer service emergency response and strengthen the prevention and response to power supply service events.

[0034] First, in the research and development of power outage assessment models, various types of data related to power outages, including the scope of power outages, duration of power outages, causes of power outages, etc., are collected and sorted out through the method of minimizing precise collection. With the help of the computing recommendation capabilities of digital systems, advanced data processing and analysis technologies are used to deeply mine and analyze these data to establish an accurate and reliable power outage assessment model. The model will be able to accurately assess the status of power outage stations and users based on real-time data, and provide support and reference for relevant decisions. Secondly, we will start to build a power outage prediction model. Through the analysis of historical power outage data and the application of prediction algorithms, we will strive to predict possible power outages in the future. The prediction model will comprehensively consider multiple factors, such as weather conditions, equipment conditions, power load, etc., to achieve effective early warning of power outage risks. At the same time, we will study the setting of risk warning thresholds, the optimization of warning methods, and the setting of load warnings to improve the accuracy and operability of warnings.

[0035] This application aims to solve various technical problems in the process of emergency power restoration in residential areas under extreme weather conditions. Through in-depth analysis and comprehensive review of events under extreme disaster weather, the shortcomings of customer service emergency response are identified, relevant technical research is carried out, and the stability and response speed of the power system in disaster response are improved.

[0036] Specifically, this application will bring the following practical significance and expected results: 1. Improve the accuracy of power outage assessment and prediction: By minimizing the precision acquisition and digital system calculation recommendation technology, an accurate and reliable power outage assessment model and power outage prediction model are established, which can evaluate and predict the status of power outage areas and users in real time. This will enable power supply companies to quickly identify affected areas when disasters occur, accurately determine the cause of power outages, take timely recovery measures, and reduce the duration and scope of power outages.

[0037] 2. Enhance disaster emergency response capabilities: Through the collection and cleaning technology of multi-source system data, combined with big data analysis methods, the accuracy of distribution transformer power outage analysis can be improved. By building risk warning and power outage prediction models, early warning of potential risks can be provided, improving the power system's ability to cope with extreme weather and protecting residents' lives and property.

[0038] 3. Reduce the workload of grassroots units: In extreme disaster weather, the traditional way of reporting power outage information is easily affected by communication interruptions, resulting in untimely information transmission and affecting the efficiency of emergency response. By studying and applying digital power outage analysis technology, power outage data can be automatically collected and processed, reducing the pressure and errors of manual reporting and improving the work efficiency of grassroots units in disasters.

[0039] 4. Promote technological innovation and application of power systems: This application aims to improve the intelligence level of power systems by researching and applying advanced data processing, machine learning and deep learning technologies. The research results can not only be applied to emergency repair and power restoration in residential areas, but also be extended to disaster response and fault handling in other power systems, promoting technological progress in the entire industry.

[0040] 5. Improve customer satisfaction and company competitiveness: Through accurate power outage analysis and prediction, we can better meet customer needs, respond to customer problems and concerns in a timely manner, and improve customer satisfaction. At the same time, we can enhance our ability to prevent and respond to power supply service incidents and consolidate our leading position and competitiveness in the market.

[0041] In short, the implementation of this application will significantly improve the emergency response capability and service quality of the power system in extreme disaster weather conditions, and provide strong technical support for ensuring the normal life and production order of residents.

[0042] This application establishes a power outage assessment and prediction model to improve the understanding and response speed of power outages, make emergency decisions and deployments faster, and minimize the impact of power outages; the prediction model can warn of possible power outages in advance, so that the company can make full preparations and responses in advance, and improve its ability to respond to emergencies such as natural disasters; based on accurate assessment results, the company can quickly deploy the required personnel, equipment and resources, improve processing efficiency, minimize the duration of power outages, and reduce the impact of power outages on customers; based on accurate assessment and prediction results, the company can better meet customers' power supply needs and service expectations, conduct targeted emergency response and resource allocation, and greatly improve the quality of power supply services and customer satisfaction. Excellent emergency service capabilities have become one of the company's core advantages in market competition; research results directly support the company's key strategic goals such as improving emergency response capabilities and optimizing resource allocation. Technological innovation and improvement of service quality will help enhance the company's competitiveness in the market and support the long-term sustainable development of power companies.

[0043] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0044] Figure 1 is a flow chart of a method for analyzing and processing power outages based on big data according to an exemplary embodiment. Figure 1 As shown, the method includes: Step S101, obtaining a time queue of weather monitoring data, wherein the weather monitoring data includes temperature value, humidity value, wind speed value and rainfall; Step S102, extracting time series features based on a local time scale from the time queue of the weather monitoring data to obtain a sequence of local time series correlation feature vectors of the weather data; Step S103, performing feature association optimization on the sequence of local time series associated feature vectors of weather data to obtain a sequence of local time series associated optimized feature vectors of weather data; Step S104: determining whether to generate a power outage warning prompt based on the feature significance time series aggregation information of the sequence of the local time series association optimization feature vector of the weather data.

[0045] In response to the above technical problems, the technical concept of this application is to use artificial intelligence technology based on deep learning to perform data time series analysis on weather monitoring data, capture the local time series change characteristics of weather data, and then through the correlation optimization and time series aggregation of the time series change characteristics of weather data in each local time domain, dig out the time series dynamic change pattern of weather data in the global time domain, so as to perform intelligent power outage warning. In this way, the accuracy and timeliness of power outage warning can be effectively improved, which helps to discover potential power outage risks in advance so as to take corresponding preventive measures.

[0046] Based on this, in the technical solution of the present application, first, a time queue of weather monitoring data is obtained, wherein the weather monitoring data includes temperature values, humidity values, wind speed values ​​and rainfall. It should be understood that the operating status of the power system is closely related to weather conditions, and temperature, humidity, wind speed and rainfall are the main factors in weather conditions that affect the safe operation and performance of power facilities. Therefore, in the technical solution of the present application, by monitoring the key meteorological parameters of the historical time period, the possible impact of weather changes on the power system can be understood, thereby realizing intelligent power outage warning prompts.

[0047] Next, in order to convert the original weather monitoring data into a data form that is easy for the machine learning model to process, the present application further uses the weather data embedding matrix to embed and encode each weather monitoring data in the time queue of the weather monitoring data, so as to map the weather monitoring data into a high-dimensional semantic space, and capture the contextual association between different meteorological parameters at the same time point, and convert the weather monitoring data at each time point into a vector representation with rich semantic information to obtain a time queue of weather monitoring embedded coding vectors, thereby providing a more effective data basis for subsequent power outage warnings. In an embodiment of the present application, the weather data embedding matrix is ​​obtained by training the Word2Vec model.

[0048] Secondly, in order to understand the dynamic change process of weather data in more detail, in the technical solution of the present application, first, the time queue of the weather monitoring embedded coding vector is time-series divided according to a predetermined time scale, and the entire time series is divided into multiple local time domains. Then, a weather local time series feature capturer based on the 1D-CNN model is used to process the weather monitoring embedded coding vector in each local time domain, so as to utilize the advantages of the 1D-CNN model in processing one-dimensional time series data, and perform a sliding window convolution operation on the time dimension to capture the time series change law of weather data in each local time domain, thereby obtaining a sequence of local time series correlation feature vectors of weather data.

[0049] In one embodiment of the present application, a time queue of the weather monitoring data is subjected to temporal feature extraction based on a local time scale to obtain a sequence of local time series associated feature vectors of the weather data, including: using a weather data embedding matrix to embed code each weather monitoring data in the time queue of the weather monitoring data to obtain a time queue of weather monitoring embedded coding vectors; dividing the time queue of the weather monitoring embedded coding vectors into a predetermined time scale and inputting the resultant into a weather local time series feature capturer based on a 1D-CNN model to obtain a sequence of local time series associated feature vectors of the weather data.

[0050] Furthermore, considering that in the global time domain, weather data in different local time domains may have different importance. Therefore, in order to further improve the accuracy of feature expression, in the technical solution of the present application, an adaptive feature optimization module based on sequence correlation semantics is introduced to process the sequence of local time series associated feature vectors of the weather data, and the feature weights are dynamically adjusted by analyzing the semantic correlation between the weather data features in each local time domain, so as to achieve a more accurate capture of the time series change pattern of the weather data. Specifically, the adaptive feature optimization module first quantifies the similarity or difference by calculating the semantic association between any two local time series associated feature vectors of weather data, so as to understand the semantic structure within the sequence and generate a set of semantic association score vectors. Then, the aggregate information of the set of semantic association score vectors is used as the overall context information, and the association optimization factor of each local time series associated feature vector of weather data is calculated to quantify the consistency or deviation between the weather data features in each local time domain and the overall semantic structure of the sequence. Furthermore, after the association optimization factor is normalized, it is used as a weight coefficient to perform weighted optimization on the original sequence of the local time series association feature vectors of the weather data, thereby suppressing features that are irrelevant to the overall semantics of the sequence and enhancing the expression of key features to obtain a sequence of local time series association optimization feature vectors of weather data.

[0051] In one embodiment of the present application, feature association optimization is performed on the sequence of local time series correlation feature vectors of weather data to obtain a sequence of local time series correlation optimized feature vectors of weather data, including: inputting the sequence of local time series correlation feature vectors of weather data into an adaptive feature optimization module based on sequence correlation semantics to obtain the sequence of local time series correlation optimized feature vectors of weather data.

[0052] Furthermore, in one embodiment of the present application, the sequence of the local time series correlation feature vectors of weather data is input into an adaptive feature optimization module based on sequence correlation semantics to obtain a sequence of the local time series correlation optimization feature vectors of weather data, including: calculating the semantic correlation score vector between any two local time series correlation feature vectors of weather data in the sequence of local time series correlation feature vectors of weather data to obtain a sequence of local time series semantic correlation score vectors of weather data; calculating the mean vector of the sequence of local time series semantic correlation score vectors of weather data to obtain a global representation vector of endogenous correlation of the sequence; calculating the correlation optimization factor of each local time series correlation feature vector of weather data in the sequence of local time series correlation feature vectors of weather data based on the global representation vector of endogenous correlation of the sequence to obtain a sequence of correlation optimization factors; and inputting the sequence of correlation optimization factors into a module. An activation function is used to obtain a sequence of association optimization weight factors; each association optimization weight factor in the sequence of association optimization weight factors is used as a weight to weight each local time series association feature vector of weather data in the sequence of local time series association feature vectors of weather data to obtain a sequence of local time series association optimization feature vectors of weather data.

[0053] Furthermore, in one embodiment of the present application, the semantic association score vector between any two local temporal association feature vectors of weather data in the sequence of local temporal association feature vectors of weather data is calculated to obtain a sequence of local temporal association score vectors of weather data, including: cascading any two local temporal association feature vectors of weather data in the sequence of local temporal association feature vectors of weather data, multiplying them by a weight coefficient matrix, and then dot-adding them with a bias vector to obtain the local temporal association score vector of weather data.

[0054] Furthermore, in one embodiment of the present application, based on the global representation vector of the intrinsic correlation of the sequence, the correlation optimization factor of each local time series correlation feature vector of weather data in the sequence of local time series correlation feature vectors of weather data is calculated to obtain a sequence of correlation optimization factors, including: multiplying the local time series correlation feature vector of weather data and the global representation vector of the intrinsic correlation of the sequence by different weight coefficient vectors and then performing addition operations to obtain semantic correlation coefficients; adding a bias parameter to the semantic correlation coefficient and then passing it through a sigmoid activation function to obtain the correlation optimization factor.

[0055] Specifically, the sequence of the local time series correlation feature vectors of the weather data is processed by the following semantic feature correlation optimization formula to obtain the sequence of the local time series correlation optimization feature vectors of the weather data, wherein the semantic feature correlation optimization formula is: in, represents a sequence of local temporal correlation feature vectors of the weather data, , , , and respectively represent the first, second, and third in the sequence of the local time series correlation feature vector of the weather data , and The local time series correlation feature vector of weather data, The value of is the number of feature vectors in the sequence of the local time series associated feature vectors of the weather data, Indicates cascade operation, is the weight coefficient matrix, is the bias vector, is the first in the sequence of the local temporal semantic association score vector of the weather data The local temporal semantic association score vector of weather data, The value of is the number of vectors in the sequence of the local temporal semantic association score vectors of the weather data, is the global representation vector of the endogenous correlation of the sequence, and Represent different weight coefficient vectors, is the bias parameter, is the sigmoid activation function, is the normalized exponential function, For the The associated optimization factor, Indicates The local time series correlation optimization feature vector of weather data.

[0056] In one embodiment of the present application, based on the feature significance time series aggregation information of the sequence of local time series association optimization feature vectors of the weather data, determining whether to generate a power outage warning prompt includes: inputting the sequence of local time series association optimization feature vectors of the weather data into a node time series aggregation response network modulated by feature significance to obtain a weather data time series significant aggregation representation vector; inputting the weather data time series significant aggregation representation vector into a classifier-based judgment result generator to obtain a judgment result, and the judgment result is used to indicate whether the power outage warning prompt is generated.

[0057] Then, in order to extract the temporal dynamic change pattern of weather data in the global time domain, the sequence of local temporal association optimization feature vectors of the weather data is further subjected to temporal information aggregation processing. Considering that in the analysis process of time series data, the influence of early weather monitoring data on the current weather state may gradually weaken over time, while recent weather monitoring data has a greater impact on the current weather state. Therefore, in the technical solution of the present application, a node temporal aggregation response network based on feature significance modulation is introduced to process the sequence of local temporal association optimization feature vectors of the weather data, and its feature significance description factor is calculated based on the mean and variance of each local temporal association optimization feature vector of the weather data to quantitatively represent its feature importance. At the same time, the feature significance is dynamically modulated according to the feature space distance span between each local time domain feature and the current local time domain feature, so that the weather information that is closer to the current local time domain in the feature space obtains a higher weighting, so as to facilitate more accurate temporal information aggregation. Then, the modulated descriptive factors are further screened using activation functions and gating mechanisms, and the sequence of the local temporal correlation optimization feature vectors of the weather data is weighted and summed using the screened descriptive factors as weights to obtain a temporally correlated weather data temporal significant aggregation representation vector. In this way, the temporal dynamic change pattern of weather data in the global time domain can be effectively extracted, providing strong data support for intelligent power outage warning.

[0058] In one embodiment of the present application, the sequence of the weather data local time series association optimization feature vectors is input into a node time series aggregation response network modulated by feature significance to obtain a weather data time series significant aggregation representation vector, including: calculating the feature significance description factor of each weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector; taking the last weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector as the current weather data local time series association optimization feature vector, and based on the other weather data local time series association optimization feature vectors in the sequence of the weather data local time series association optimization feature vectors and The distance span between the local temporal association optimization feature vectors of the current weather data is used to construct the feature significance attenuation factor of each other local temporal association optimization feature vector of the weather data; the product between the feature significance attenuation factor of each other local temporal association optimization feature vector of the weather data and its feature significance description factor is calculated to obtain a sequence of feature significance attenuation description factors; the sequence of feature significance attenuation description factors is input into the gated mask module to obtain a sequence of feature significance attenuation weight factors; based on the sequence of feature significance attenuation weight factors, the weighted sum of the sequence of the local temporal association optimization feature vector of the weather data is calculated to obtain the temporal significant aggregation representation vector of the weather data.

[0059] Furthermore, in one embodiment of the present application, the characteristic significance description factor of each weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector is calculated, including: calculating the expected value of the fourth power of the difference between each eigenvalue in the weather data local time series association optimization feature vector and its characteristic mean, and dividing the expected value by the square of the characteristic variance of the weather data local time series association optimization feature vector to obtain the characteristic significance description factor.

[0060] Furthermore, in one embodiment of the present application, based on the distance span between each other local time series association optimization feature vector of weather data in the sequence of the local time series association optimization feature vector of weather data and the current local time series association optimization feature vector of weather data, a feature significance attenuation factor of each other local time series association optimization feature vector of weather data is constructed, including: calculating the difference between the maximum eigenvalue of the current local time series association optimization feature vector of weather data and the maximum eigenvalue of each other local time series association optimization feature vector of weather data in the sequence of the local time series association optimization feature vector of weather data, and then dividing the difference by the number of eigenvectors separated therefrom to obtain the feature significance attenuation factor of each other local time series association optimization feature vector of weather data.

[0061] Specifically, the sequence of the local time series correlation optimization feature vectors of the weather data is processed by the following feature significance attenuation guided fusion formula to obtain the weather data time series significant aggregation representation vector, wherein the feature significance attenuation guided fusion formula is: in, represents a sequence of optimized feature vectors of local temporal association of the weather data, The value of is the number of feature vectors in the sequence of the local time series correlation optimization feature vector of the weather data, , , , and Respectively represent the first, second, and third in the sequence of the local time series association optimization feature vector of the weather data , The local time series correlation optimization feature vector of the current weather data, represents the maximum eigenvalue of the eigenvector, Indicates The first feature vector of the local temporal correlation optimization of weather data eigenvalues, and Respectively represent the The characteristic mean and square of the characteristic variance of the local time series correlation optimization feature vector of weather data are Indicates the The fourth-order central moment of the local time series correlation optimization feature vector of weather data, that is, The expected value of the fourth power of the difference between each eigenvalue and its eigenmean in the local time series correlation optimization eigenvector of weather data, , Indicates the The characteristic significance attenuation factor of the optimized characteristic vector of the local time series correlation of weather data, Indicates The feature significance decay factor is described by Indicates Normalized feature significance decay description factor, Indicates feature significance decay weight factor, is the preset threshold, Indicates mask processing, A representation vector representing significant aggregation of the weather data time series.

[0062] Finally, the weather data time series significant aggregation representation vector is further input into a classifier-based analysis result generator to obtain an analysis result, and the analysis result is used to indicate whether a power outage warning prompt is generated.

[0063] In a preferred example, the weather data time series significant aggregation representation vector is passed through a classifier-based judgment result generator to obtain a judgment result, which includes: Determine the maximum eigenvalue of the weather data time series significant aggregation and the minimum eigenvalue of the weather data time series significant aggregation of the weather data time series significant aggregation representation vector, and calculate the weather data time series significant aggregation mean and weather data time series significant aggregation standard deviation of the feature set of the weather data time series significant aggregation representation vector; Calculate the quotient of the significant aggregation mean of the weather data time series and the significant aggregation standard deviation of the weather data time series to obtain a significant aggregation statistical normalization value of the weather data time series; Calculate the inverse of each eigenvalue of the weather data time series significant aggregation representation vector, multiply it by the difference between the maximum eigenvalue of the weather data time series significant aggregation and the minimum eigenvalue of the weather data time series significant aggregation, and then subtract it from the statistical normalization value of the weather data time series significant aggregation to obtain an approximate vector of the weather data time series significant aggregation distribution; Calculate an exponential function with a natural constant as the base, with each eigenvalue of the weather data time series significant aggregation distribution approximation vector as an exponent, and perform point addition with the weather data time series significant aggregation statistical normalization value to obtain a weather data time series significant aggregation probability approximation vector; Calculating the base-two logarithm of the absolute value of each eigenvalue of the weather data time series significant aggregation probability approximation vector to obtain an optimized weather data time series significant aggregation representation vector; and The optimized weather data time series significant aggregation representation vector is passed through a classifier-based judgment result generator to obtain a judgment result.

[0064] Among them, the weather data time series significant aggregation representation vector is denoted as The optimization expression is: , , and are the feature sets of the weather data time series significant aggregation representation vectors The mean, standard deviation, minimum and maximum values ​​of is the logarithm to base 2, is the significant aggregation representation vector of the weather data time series, is the inverse of each eigenvalue of the weather data time series significant aggregation representation vector, It is an exponential function that calculates the exponential of each eigenvalue of a vector with a natural constant as the base. is the optimized weather data time series significant aggregation representation vector, It is added by location point. It is the point multiplication by position. It is subtracted by position point.

[0065] Here, in the preferred example, considering that the sequence of the local temporal association optimization feature vectors of the weather data respectively represents the local time domain context temporal semantic association features of the weather monitoring data, when performing a node temporal aggregation response based on feature significance modulation, the difference in the energy distribution of the temporal nodes in each local time domain will have different aggregation response weights based on the node temporal aggregation response, so that the weather data temporal significant aggregation representation vector will also have a diversified set expression distribution of aggregation response features. Therefore, it is expected to improve the balance between the regression mapping accuracy and completeness of the weather data temporal significant aggregation representation vector when performing class regression through a classifier-based judgment result generator, thereby improving the accuracy of the obtained judgment results.

[0066] Therefore, by performing random statistical normalization on the diversified feature sets of the weather data time series significant aggregation representation vector, the confidence space constructed based on the overall eigenvalue of the weather data time series significant aggregation representation vector is approximated by the standardized continuous probability density distribution relative to the response hypothesis test of each eigenvalue of the weather data time series significant aggregation representation vector, thereby establishing the target reachability of the discrete feature distribution of the weather data time series significant aggregation representation vector to a unified regression target, so as to achieve a balanced executable performance between mapping accuracy and mapping completeness in the class regression process based on the discrete feature distribution of the weather data time series significant aggregation representation vector, and improve the accuracy of the judgment results obtained by the weather data time series significant aggregation representation vector through the classifier-based judgment result generator.

[0067] In summary, the above scheme uses artificial intelligence technology based on deep learning to analyze the weather monitoring data in a time series, captures the local time series change characteristics of the weather data, and then through the correlation optimization and time series aggregation of the weather data time series change characteristics in each local time domain, digs out the time series dynamic change pattern of the weather data in the global time domain, so as to perform intelligent power outage warning. In this way, the accuracy and timeliness of power outage warning can be effectively improved, which helps to discover potential power outage risks in advance so as to take corresponding preventive measures.

[0068] Figure 2 FIG. 1 is a block diagram of a power outage analysis and processing system based on big data according to an exemplary embodiment. Figure 2 As shown, the system 200 includes: A weather monitoring data acquisition module 201 is used to acquire a time queue of weather monitoring data, wherein the weather monitoring data includes temperature value, humidity value, wind speed value and rainfall; A time series feature extraction module 202 is used to extract time series features based on a local time scale from the time queue of the weather monitoring data to obtain a sequence of local time series correlation feature vectors of the weather data; A feature association optimization module 203 is used to perform feature association optimization on the sequence of weather data local time series association feature vectors to obtain a sequence of weather data local time series association optimization feature vectors; The power outage warning prompt determination module 204 is used to determine whether to generate a power outage warning prompt based on the feature significance time series aggregation information of the sequence of the local time series association optimization feature vector of the weather data.

[0069] It should be understood that big data technology can be used to achieve efficient prediction of risk warning and system power outages and accurate analysis of carrying capacity thresholds, such as Figure 3As shown. First, the data acquisition module is responsible for collecting various types of data from the system, including real-time operation data, historical fault records, and environmental information. These data are acquired efficiently and transmitted to the data processing center in an efficient manner. Subsequently, in the data processing module, the collected raw data is preprocessed, including data cleaning, formatting, and standardization, to ensure data quality and prepare for subsequent analysis. Next, the processed data is analyzed using machine learning and data mining techniques to extract key features. It takes into account a variety of factors, such as fault characteristics, meteorological factors, historical fault data, etc., to determine the specific type and location of the fault. The entire system framework works closely together. The process of processing, analyzing, and mining large-scale data sets by applying various technologies and tools. In the power industry, the generation of massive data involves multiple aspects such as power grid monitoring, power production, and energy consumption. In this project, the use of big data analysis methods can effectively process and analyze massive data in the power industry, playing the following key roles.

[0070] Improve decision-making support capabilities. Through big data analysis, we can deeply mine and analyze power industry data from multiple dimensions and angles, provide decision makers with accurate and comprehensive information support, and make scientific and wise decisions. And monitor the operating status of the power system in real time, and promptly discover and predict potential faults or anomalies. By analyzing real-time data streams, we can predict the health status of power equipment in advance, take preventive maintenance measures, and reduce the risk of power outages.

[0071] Insight mining capability: Through the application of big data technology, we can reveal the potential patterns, trends and correlations in the power industry data, discover business problems, management loopholes and improvement opportunities, and provide insights and innovative ideas. It is the process of finding patterns, trends and correlations in data through exploratory analysis. Common data mining techniques include clustering, classification, association rule mining, etc.

[0072] Improve business process optimization capabilities, discover bottlenecks, inefficiencies and problems in business processes through analysis of power industry data, provide optimization suggestions and solutions, and achieve lean management and business process optimization. The application of big data analysis in the power industry can improve the intelligence level of the power system, improve operating efficiency, reduce energy waste, enhance system stability and security, and promote the power industry to develop in a more sustainable and intelligent direction.

[0073] Furthermore, if Figure 4As shown, the flowchart shows how to determine whether to issue a warning or identify a power outage by checking the node fault information layer by layer when the power failure warning or power failure judgment condition is triggered. The process starts when the power failure warning or power failure judgment condition is triggered. First, the system checks whether there is fault information of the bottom-level (customer) node. If it exists, it continues to call the fault information vector P^' of the bottom-level child node; if it does not exist, the process turns to the failure warning of the lower-level child node fault. Next, the system will obtain the fault information P of this node. Then check whether the fault information is zero. If P is zero, the system will obtain the fault information vector P of the child node of this layer corresponding to the parent node, and calculate the fault information P^' of the parent node. Then check whether the parent node fault information P^' is zero. If P^' is zero, the parent node is judged to be faulty; if P^' is not zero, the node is judged to be faulty. If the fault information P of this node is not zero, the node is directly judged to be faulty. Finally, the system checks whether there is root node (trunk) fault information. If it exists, the fault analysis result is output; if it does not exist, the process ends. In summary, the flowchart details the decision-making process of determining whether to issue a warning or identify a power outage by examining node fault information layer by layer, from the underlying customer nodes to the root node.

[0074] In this application, monitoring and early warning are carried out for six aspects, including power outage user changes, work order changes, call traffic changes, power outage event changes, terminal offline changes and weather changes. By monitoring indicator data and setting change thresholds, early warning reminders are issued when indicator data reaches or exceeds the threshold, and the number of power outage users is predicted, and corresponding early warning signals are issued when the early warning threshold is reached.

[0075] Monitoring and early warning of six aspects, including power outage user changes, work order changes, call traffic changes, power outage event changes, terminal offline changes and weather changes, can be achieved by using monitoring indicator data and set change thresholds. The following is an expanded description of the monitoring and early warning methods for each aspect: 1. Monitoring and early warning of abnormal changes in power outage users: Monitoring indicator data: Monitor the number, proportion or change trend of power outage users. Abnormal change threshold setting: Set the threshold of the number or proportion of power outage users. If the threshold is exceeded, it is considered that there is an abnormal change in power outage users. Early warning reminder: When the number or proportion of power outage users reaches or exceeds the set threshold, the system can send an early warning reminder to notify relevant personnel to handle and investigate.

[0076] 2. Work order change monitoring and early warning: Monitoring indicator data: Monitor the number of work orders, processing time or abnormal situations. Change threshold setting: Set the threshold of the number of work orders or processing time. If the threshold is exceeded or an abnormal situation occurs, it is considered that there is a work order change. Early warning reminder: When the number of work orders exceeds the set threshold, the processing time is delayed or an abnormal situation occurs, the system can send an early warning reminder to remind relevant personnel to handle and follow up the work order in a timely manner.

[0077] 3. Traffic change monitoring and early warning: Monitoring indicator data: monitor call volume, call quality or abnormal situations. Abnormal change threshold setting: set the threshold of call volume or call quality. If the threshold is exceeded or an abnormal situation occurs, it is considered that there is a traffic change. Early warning reminder: When the call volume exceeds the set threshold, the call quality decreases or an abnormal situation occurs, the system can send an early warning reminder to notify relevant personnel to investigate and handle it.

[0078] 4. Power outage event abnormality monitoring and early warning: Monitoring indicator data: Monitor the number, duration or abnormal conditions of power outage events. Abnormality threshold setting: Set the threshold of the number or duration of power outage events. If the threshold is exceeded or an abnormal condition occurs, it is considered that there is a power outage event abnormality. Early warning reminder: When the number of power outage events exceeds the set threshold, the duration is prolonged or an abnormal condition occurs, the system can send an early warning reminder to remind relevant personnel to take timely measures and conduct troubleshooting.

[0079] 5. Monitoring and early warning of terminal offline changes: Monitoring indicator data: Monitor the online status of terminal devices, communication anomalies or changes in the number of terminals. Abnormal threshold setting: Set the threshold for terminal offline or communication anomalies. If the threshold is exceeded or an abnormal situation occurs, it is considered that there is a terminal offline change. Early warning reminder: When the number of terminal devices offline exceeds the set threshold, communication anomalies or other abnormal situations occur, the system can send an early warning reminder to notify relevant personnel to conduct inspection and maintenance.

[0080] 6. Weather abnormality monitoring and early warning: Monitoring index data: Monitor weather data, such as temperature, wind speed, rainfall, etc. Abnormality threshold setting: Set the corresponding weather abnormality threshold according to the impact of weather on the power supply system. Early warning reminder: When weather data reaches or exceeds the set abnormal threshold, the system can send an early warning reminder to notify relevant personnel to pay attention to abnormal weather conditions that may affect the power supply system.

[0081] Figure 5 shows the relationship between different weather types (represented by weather codes) and the number of power outage users. The chart type is a box plot, which is used to show the distribution and statistical characteristics of the data. For the relationship between different weather types and the number of power outage users, the X-axis of the chart represents the weather code, ranging from 0 to 12, and different values ​​represent different weather types. The Y-axis represents the number of power outage users, ranging from 0 to 800,000, reflecting the distribution of the number of power outage users under different weather types.

[0082] Each weather code corresponds to a box plot, the upper and lower edges of the box represent the first quartile (Q1) and the third quartile (Q3) of the data, and the center line of the box represents the median of the data. The whiskers above and below the box extend to the minimum and maximum values ​​of the data (excluding outliers). The dots in the figure represent outliers, which are beyond the whisker range.

[0083] The chart shows that the distribution of the number of users with power outages varies under different weather types. For example, when the weather codes are 4 and 7, the median number of users with power outages is higher and the distribution range is wider, indicating that these weather types have a greater impact on power outages. When the weather codes are 11 and 12, the number of users with power outages is smaller and the distribution is more concentrated, indicating that these weather types have less impact on power outages.

[0084] Overall, the chart clearly shows the impact of different weather conditions on the number of users experiencing power outages, which helps to analyze and predict power outages under different weather conditions and formulate corresponding response measures.

[0085] Reference below Figure 6 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0086] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0087] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0088] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0089] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0090] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0091] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0092] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0093] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0094] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0095] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0096] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0097] In the application scenario of the power outage assessment and processing method based on big data, first, the time queue of weather monitoring data is obtained; then, the acquired time queue of weather monitoring data is input into a server deployed with a power outage assessment and processing algorithm based on big data, wherein the server can process the time queue of weather monitoring data based on the power outage assessment and processing algorithm based on big data to determine whether to generate a power outage warning prompt.

[0098] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0099] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0100] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A power outage analysis and processing method based on big data, characterized in that: include: Obtaining a time queue of weather monitoring data, wherein the weather monitoring data includes temperature value, humidity value, wind speed value and rainfall; Extracting time series features based on a local time scale from the time queue of the weather monitoring data to obtain a sequence of local time series correlation feature vectors of the weather data; Performing feature association optimization on the sequence of local time series associated feature vectors of weather data to obtain a sequence of local time series associated optimized feature vectors of weather data; Based on the feature significance time series aggregation information of the sequence of the local time series association optimization feature vector of the weather data, it is determined whether to generate a power outage warning prompt.

2. The method for analyzing and processing power outages based on big data according to claim 1 is characterized in that: Extracting time series features based on a local time scale from the time queue of the weather monitoring data to obtain a sequence of local time series correlation feature vectors of the weather data includes: Using a weather data embedding matrix, each weather monitoring data in the time queue of the weather monitoring data is embedded and encoded to obtain a time queue of weather monitoring embedded coding vectors; The time queue of the weather monitoring embedded coding vector is divided into a predetermined time scale and then input into a weather local time series feature capturer based on a 1D-CNN model to obtain a sequence of local time series associated feature vectors of the weather data.

3. The method for analyzing and processing power outages based on big data according to claim 2 is characterized in that: Performing feature association optimization on the sequence of local time series associated feature vectors of weather data to obtain a sequence of local time series associated optimized feature vectors of weather data, including: The sequence of the local time series associated feature vectors of the weather data is input into an adaptive feature optimization module based on sequence correlation semantics to obtain the sequence of the local time series associated optimized feature vectors of the weather data.

4. The method for analyzing and processing power outages based on big data according to claim 3 is characterized in that: Inputting the sequence of the local time series correlation feature vectors of the weather data into an adaptive feature optimization module based on sequence correlation semantics to obtain the sequence of the local time series correlation optimization feature vectors of the weather data, including: Calculating a semantic association score vector between any two weather data local time series association feature vectors in the sequence of weather data local time series association feature vectors to obtain a sequence of weather data local time series semantic association score vectors; Calculating the mean vector of the sequence of local temporal semantic association score vectors of the weather data to obtain a global representation vector of the sequence endogenous correlation; Based on the global representation vector of the endogenous correlation of the sequence, calculating the correlation optimization factor of each local time series correlation feature vector of weather data in the sequence of local time series correlation feature vectors of weather data to obtain a sequence of correlation optimization factors; The sequence of the associated optimization factors is input into Activation function to obtain a sequence of associated optimization weight factors; Taking each association optimization weight factor in the sequence of association optimization weight factors as a weight, each weather data local time series association feature vector in the sequence of weather data local time series association feature vectors is weighted respectively to obtain the sequence of weather data local time series association optimization feature vectors.

5. The method for analyzing and processing power outages based on big data according to claim 4 is characterized in that: Calculating the semantic association score vector between any two weather data local time series association feature vectors in the sequence of weather data local time series association feature vectors to obtain a sequence of weather data local time series semantic association score vectors, including: Any two local temporal series associated feature vectors of weather data in the sequence of local temporal series associated feature vectors of weather data are cascaded and multiplied by a weight coefficient matrix, and then dotted with a bias vector to obtain the local temporal series semantic association score vector of weather data.

6. The method for analyzing and processing power outages based on big data according to claim 5 is characterized in that: Based on the global representation vector of the sequence endogenous correlation, calculating the correlation optimization factor of each weather data local time series correlation feature vector in the sequence of the weather data local time series correlation feature vector to obtain a sequence of correlation optimization factors, including: The local time series correlation feature vector of the weather data and the global representation vector of the sequence endogenous correlation are multiplied by different weight coefficient vectors respectively and then added to obtain a semantic correlation coefficient; The semantic association coefficient is added with a bias parameter and then passed through a sigmoid activation function to obtain the association optimization factor.

7. The method for analyzing and processing power outages based on big data according to claim 6 is characterized in that: Determining whether to generate a power outage warning prompt based on the characteristic significance time series aggregation information of the sequence of the local time series association optimization feature vector of the weather data includes: Inputting the sequence of the local temporal correlation optimization feature vectors of the weather data into the node temporal aggregation response network modulated by feature significance to obtain a temporal significant aggregation representation vector of the weather data; The weather data time series significant aggregation representation vector is input into a classifier-based analysis and judgment result generator to obtain an analysis and judgment result, and the analysis and judgment result is used to indicate whether the power outage warning prompt is generated.

8. The method for analyzing and processing power outages based on big data according to claim 7 is characterized in that: Inputting the sequence of the local temporal correlation optimization feature vectors of the weather data into the node temporal aggregation response network modulated by feature significance to obtain the temporal significant aggregation representation vector of the weather data, including: Calculating a characteristic significance description factor of each weather data local time series association optimization feature vector in the sequence of weather data local time series association optimization feature vectors; Taking the last weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector as the current weather data local time series association optimization feature vector, and constructing the feature significance attenuation factors of the other weather data local time series association optimization feature vectors based on the distance span between the other weather data local time series association optimization feature vectors in the sequence of the weather data local time series association optimization feature vector and the current weather data local time series association optimization feature vector; Calculate the product of the characteristic significance attenuation factor of each other weather data local time series correlation optimization characteristic vector and its characteristic significance description factor to obtain a sequence of characteristic significance attenuation description factors; Inputting the sequence of feature significance attenuation description factors into a gated mask module to obtain a sequence of feature significance attenuation weight factors; Based on the sequence of feature significance attenuation weight factors, a weighted sum of the sequence of weather data local time series association optimization feature vectors is calculated to obtain the weather data time series significant aggregation representation vector.

9. The method for analyzing and processing power outages based on big data according to claim 8 is characterized in that: Calculating the characteristic significance description factor of each weather data local time series association optimization feature vector in the sequence of weather data local time series association optimization feature vectors, including: The expected value of the fourth power of the difference between each eigenvalue in the local time series correlation optimization feature vector of the weather data and its eigenvalue mean is calculated, and the expected value is divided by the square of the eigenvariance of the local time series correlation optimization feature vector of the weather data to obtain the feature significance description factor.

10. The method for analyzing and processing power outages based on big data according to claim 9 is characterized in that: Based on the distance span between each other weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector and the current weather data local time series association optimization feature vector, constructing the feature significance attenuation factor of each other weather data local time series association optimization feature vector, including: Calculate the difference between the maximum eigenvalue of the current weather data local time series association optimization feature vector and the maximum eigenvalue of each other weather data local time series association optimization feature vector in the sequence of the weather data local time series association optimization feature vector, and then divide the difference by the number of eigenvectors between the two to obtain the characteristic significance attenuation factor of each other weather data local time series association optimization feature vector.

Images