Intelligent generation and submission regulation and control management method and system based on power grid big data

Abstract

The invention provides an intelligent generation and submission regulation and control management method and system based on power grid big data, and the system comprises a data layer which is used for obtaining power grid data and building a power grid operation data information pool; the business layer is used for monitoring the power grid operation data obtained by the data layer and intelligently assisting in generating submitted data; the service layer is used for notifying the submitted data generated by the aid of the service layer; the intelligent generation and submission regulation and control management method comprises the following steps: acquiring power grid data by using a data layer and establishing a power grid operation data information pool; the service layer is utilized to intelligently assist in generating submitted data according to the power grid operation data obtained by the data layer; and monitoring and notifying submitted data generated by the service layer in an auxiliary manner by using the service layer. When a fault occurs, fault information can be obtained in time, fault report content can be generated in time, system notification and short message sending can be carried out, and the problems that multi-source data integration is difficult and the data report process is tedious are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power dispatching, and in particular to an intelligent generation and reporting control management method and system. Background Art

[0002] As the construction of new power systems continues to advance, multiple dispatch objects continue to emerge. The massive data collected by the dispatching department increases the real-time control burden of dispatchers. While the power grid operation faces problems such as large power grid security, new energy consumption, power balance, and power grid peak regulation, the power grid operation perception, regulation control, fault analysis, and business collaboration also face many challenges, which put forward higher requirements for power grid dispatching management. The current dispatching mode has the problems of insufficient data integration of equipment operation status, difficulty in adapting to traditional dispatching control auxiliary decision-making, and insufficient mining of various massive data. Data reporting requires dispatchers to search and analyze data in relevant power grid operation systems and fill in the reported data information, which takes up a lot of dispatchers' time and energy and also has problems in timeliness.

[0003] Although smart grids have made great progress and require a cloud-based smart grid monitoring and management system, the existing control and management methods still have problems such as difficulty in integrating multi-source data, difficulty in adapting to traditional dispatching and control decision-making assistance, and cumbersome data reporting processes. Summary of the invention

[0004] In view of the technical problems of difficulty in integrating multi-source data and cumbersome data reporting process in existing methods, the present invention proposes a method and system for intelligent generation and reporting of control management based on power grid big data. By integrating control operation data resources, forming data linkage, and establishing data intelligent reporting function, when a fault occurs, fault information is obtained in time, fault reporting content is generated, system notification and text message sending are carried out, which effectively solves the technical problems of difficulty in integrating multi-source data and cumbersome data reporting process.

[0005] In order to achieve the above object, the technical solution of the present invention is achieved as follows:

[0006] An intelligent generation and reporting control management system based on power grid big data, characterized by comprising:

[0007] The data layer is used to obtain power grid data and establish a power grid operation data information pool;

[0008] The business layer is used to monitor the power grid operation data obtained by the data layer and intelligently assist in generating and reporting data;

[0009] The service layer is used to transmit the reporting data generated with the assistance of the business layer.

[0010] The data layer includes a data acquisition module, a data preprocessing module, a data integration module and a data storage module connected in sequence;

[0011] The data acquisition module is used to collect various raw data from various sources of the power grid;

[0012] The data preprocessing module is used to preprocess the various collected raw data;

[0013] The data integration module is used to integrate the pre-processed raw data into multi-source heterogeneous data to obtain unified and standardized power grid operation data;

[0014] The data storage module is used to store original power grid data, structured data and cache data.

[0015] The multiple sources include SCADA, EMS, DMS, OMS, and D5000 systems; the multiple original data include real-time collected power grid operation data, power grid energy management data, distribution system data, operation management system data, and dispatching automation system data.

[0016] The data acquisition module adopts a universal API interface and uses message queue technology to receive data streams from different systems;

[0017] The data preprocessing module includes a format conversion unit, a de-manipulation processing unit and a cluster analysis unit;

[0018] A format conversion unit, used to convert a variety of raw data in different formats into a unified standard format;

[0019] A denoising processing unit, used for denoising a variety of raw data in a unified standard format by using a denoising processing method based on wavelet transform;

[0020] A clustering analysis unit, used for classifying and dividing the noise-free power grid operation data obtained by the denoising processing unit by using a clustering algorithm;

[0021] The data integration module uses the ETL method to organize the data processed by the cluster analysis unit, and uses a distributed computing method to associate and fuse the organized data to achieve multi-source heterogeneous data integration;

[0022] The data storage module includes a distributed file system, a distributed column database and a distributed cache database;

[0023] Distributed file system is used to store raw power grid data;

[0024] The distributed column database is used to store the structured data, which includes pre-processed data, data after integration of multi-source heterogeneous data, reporting data, notification reminder data and SMS sending data;

[0025] The distributed cache database is used to store cache data.

[0026] The business layer includes an intelligent monitoring module, a template definition module and a fault information generation module;

[0027] The intelligent monitoring module is used to monitor the power grid operation data, obtain fault monitoring data and store it in the distributed cache database; the intelligent monitoring module detects abnormal behavior and potential faults in the power grid through a machine learning algorithm, and the machine learning algorithm adopts Isolation Forest or One-Class SVM;

[0028] The template definition module is used to define different types of fault templates, including fault type, fault characteristics, fault parameters and fault handling actions; fault types include 220kV, 110kV tripping and county-wide large-scale power outages;

[0029] The fault information generation module is used to generate reporting data based on the fault template and fault monitoring data;

[0030] The fault information generation module includes a data analysis module and an information integration module; the data analysis module is used to analyze and process the fault monitoring data and remove useless information; the information integration module is used to combine the analysis and processing results with the fault template to synthesize the reporting data, and the information integration module uses a method based on a Gaussian mixture model to generate the reporting data.

[0031] The method based on Gaussian mixture model is as follows: a set of power grid data obtained after analysis by the data analysis module is X, where each sample x n Contains multiple features, such as voltage, current, power, etc. Each sample also contains the label of fault type and fault feature;

[0032] 1. First build a Gaussian mixture model:

[0033] 2.

[0034] Where K is the number of Gaussian distributions, π k is the weight of the kth Gaussian distribution, satisfying ρ() is the probability density function of the kth Gaussian distribution, with mean μ k , the covariance matrix is ​​σ k ;

[0035] 3. Initialize weight π k , mean value π k, covariance matrix σ k ;

[0036] 4.E step: Calculate the posterior probability that each sample belongs to each Gaussian distribution:

[0037] 5.

[0038] 6. Among them, γ(z nk ) is the posterior probability that n samples belong to the kth Gaussian distribution;

[0039] 7.M step: Update weight π k , mean value π k , covariance matrix σ k ;

[0040] Repeat steps E and M until the parameters converge or the predetermined number of iterations is reached to obtain a trained model;

[0041] According to the weight π corresponding to the trained model k Randomly select a Gaussian distribution; generate a sample from the selected Gaussian distribution; and assign corresponding fault types and fault feature labels to the generated sample according to the characteristics of the sample.

[0042] The service layer includes a notification reminder module and a text message sending module;

[0043] The notification and reminder module is used to send real-time notifications of anomalies and failures to relevant personnel through system pop-up windows based on the reported data;

[0044] The SMS sending module is used to send emergency notifications of abnormalities and failures to key personnel via SMS based on the reported data.

[0045] A method for intelligent generation, reporting and control management based on power grid big data, comprising the steps of:

[0046] S1: Use the data layer to obtain power grid data and establish a power grid operation data information pool;

[0047] S2: Use the business layer to intelligently assist in generating reporting data based on the power grid operation data obtained from the data layer;

[0048] S3: Use the service layer to monitor and notify the reporting data assisted by the business layer.

[0049] Step S1 is specifically as follows: a data acquisition module is used to connect with a data source using a general API interface to obtain a variety of raw data; a preprocessing module is used to clean, convert and organize a variety of raw data to obtain a variety of preprocessed raw data; a data integration module is used to integrate the preprocessed raw data from multiple sources and heterogeneous data to obtain unified and standardized power grid operation data, and store the data in a distributed column database to obtain a power grid operation data information pool;

[0050] Step S2 specifically includes using the intelligent monitoring module to monitor the power grid operation data and record the fault monitoring data; using the template definition module to define the fault type, fault characteristics, and fault parameters; using the data analysis module to analyze and process the fault monitoring data and remove useless information; using the information integration module to combine the analysis and processing results with the fault template to synthesize the reporting data;

[0051] Step S3 specifically uses the notification reminder module to send real-time notifications of abnormalities and failures to relevant personnel through a system pop-up window based on the reported data; and uses the SMS sending module to send emergency notifications of abnormalities and failures to key personnel through SMS based on the reported data.

[0052] The beneficial effects of the present invention include: real-time and accurate data-assisted generation and transmission can ensure that dispatchers obtain key information at the first time, so as to make decisions quickly; the automatic data generation and transmission process reduces manual participation and errors, reduces labor costs, saves manpower and material resources, and reduces the risk of power grid operation; the indirect benefits after the promotion of the results include: promoting business innovation of control and management through the auxiliary generation and transmission of control and management data, strengthening internal collaboration and communication, and providing positive feedback for the intelligent development and improvement of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 This is a structural diagram of the intelligent generation and reporting control management system of the present invention.

[0055] Figure 2 The data preprocessing module of the present invention.

[0056] Figure 3 It is a structural diagram of the data storage module of the present invention.

[0057] Figure 4 This is a structural diagram of the fault information generation module of the present invention.

[0058] Figure 5 This is a flow chart of the intelligent generation and reporting regulation and management method of the present invention.

[0059] Figure 6 This is the reporting data of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] Example 1

[0062] like Figure 1 As shown, an intelligent generation and reporting control management system based on power grid big data includes a data layer, a business layer and a service layer;

[0063] The data layer includes a data acquisition module, a data preprocessing module, a data integration module and a data storage module connected in sequence. The data layer is used to obtain power grid data and establish a power grid operation data information pool to facilitate data extraction and circulation, facilitate fault analysis and judgment, and realize timely and comprehensive perception of the power grid operation status by the dispatcher. It also considers the feasibility and plans of reducing workload and shortening working hours from the perspective of work flow.

[0064] The data acquisition module is used to collect various raw data from various sources of the power grid, including power grid operation and equipment detection or monitoring data; data sources, such as SCADA (supervisory control and data acquisition system), EMS (energy management system), DMS (distribution management system), OMS (operation management system), D5000 system, etc. The various raw data include real-time collected power grid operation data, power grid energy management data, distribution system data, operation management system data and dispatching automation system data.

[0065] The data acquisition module adopts a universal API interface and supports common power industry communication protocols, such as IEC60870-5-104, IEC 61850, Modbus TCP, etc. It uses Apache Kafka or RabbitMQ message queue technology to receive data streams from different systems, ensure data transmission between different microservices, decouple producers and consumers, and ensure the reliability of data transmission. For batch data, it supports FTP / SFTP and other methods for file transfer to ensure the efficiency and reliability of transmission.

[0066] Data preprocessing modules, such as Figure 2 As shown, it includes a format conversion unit, a de-operation processing unit and a cluster analysis unit, and a data preprocessing module, which is used to preprocess the collected various raw data to improve the data quality;

[0067] Specific steps:

[0068] Format conversion unit: converts various raw data in different formats into a unified standard format (such as JSON, CSV, Parquet). Timestamp standardization: unifies the timestamp format and uses UTC time.

[0069] De-noising unit: used for denoising various raw data in a unified standard format by using a de-noising method based on wavelet transform, specifically: performing discrete wavelet transform on various raw data in a unified standard format, extracting wavelet coefficients at different levels (including approximate coefficient A and detail coefficient D); selecting a threshold by using a heuristic threshold method or a SURE threshold method, performing soft threshold processing on each detail coefficient, removing coefficients less than the threshold, obtaining detail coefficient Dˊ after threshold processing, and not processing approximate coefficient A; retaining approximate coefficient A and detail coefficient Dˊ after threshold processing; performing wavelet reconstruction by inverse discrete wavelet transform, and obtaining noise-free power grid operation data;

[0070] Cluster analysis unit: used to classify and divide the noise-free power grid operation data obtained by the denoising processing unit using a clustering algorithm for better management and analysis:

[0071] The K-means clustering algorithm is used, specifically:

[0072] Input: M noise-free power grid operation data and the number of clusters m;

[0073] Initialization: randomly select m noise-free power grid operation data from M noise-free power grid operation data as the initial clustering centers;

[0074] Allocation: For each noise-free power grid operation data, the Euclidean distance to each cluster center is calculated, and the points closest to the same center point are grouped into one class;

[0075] Update: Recalculate the mean of all objects in each class and update the class center.

[0076] Iteration: Repeat the assignment and update steps until the categories of the sample points before and after clustering are the same, that is, the clustering has become stable.

[0077] Output: Output the class with the smallest variance, and use the cluster center of this class as the valid data after filtering out redundant data.

[0078] The data integration module is used to integrate multi-source heterogeneous data of various pre-processed raw data to obtain unified and standardized power grid operation data.

[0079] Use ETL tools such as Talend, Informatica, Apache Nifi, etc., and use data extraction, conversion and loading methods to organize the data processed by the cluster analysis unit, so as to configure batch and incremental data synchronization tasks to ensure timely data updates. Relying on the change data capture function of the ETL tool, incremental data can be processed. Data conversion and data acquisition are loosely coupled, and asynchronous processing can be achieved; adopt a distributed computing-based method, use a distributed computing framework (such as Apache Spark, Flink) to process the sorted data in parallel, associate and fuse the sorted data, and form a comprehensive data view, thereby improving the availability, consistency and analysis capabilities of the data; specifically: divide the sorted data into multiple shards, each shard is processed on a different computing node; independently process the shard data on each computing node, summarize the processing results of each node, and generate the final fusion result. .

[0080] Data storage modules, such as Figure 3 As shown, it is used to store original power grid data, structured data and cached data; the data set storage module includes a distributed file system, which is used to store original power grid data, supports the writing and reading of large-scale data, and is suitable for batch processing and data analysis; the distributed column database is used to store structured data, such as pre-processed data, data after integration of multi-source heterogeneous data, and the reported data, notification reminder data and SMS sending data described later, supports high-concurrency read and write operations, is suitable for real-time or near real-time data access, efficiently processes sparse data, only stores non-null values, and saves storage space; the distributed cache database is used to store cached data, provides extremely low read and write latency, improves the response speed of the system, and the data is mainly stored in the memory, and the reading speed is very fast;

[0081] The power grid operation data information pool includes power grid operation data collected in real time from the SCADA system, such as real-time electrical parameters (voltage, current, power, etc.), switch status, and protection device status;

[0082] Energy management data of the power grid from the EMS system: power grid topology, generation and load forecasting, economic dispatch, market data, state estimation;

[0083] Data from the distribution system of the DMS system: distribution network topology, load management, fault location, power outage management, and distributed energy data;

[0084] Data from the operation management system of the OMS system, including power outage events, emergency repair tasks, customer complaints, and resource management;

[0085] Data from the dispatching automation system of the D5000 system, including real-time data, historical data, dispatching instructions, alarm information, and report generation.

[0086] The business layer includes an intelligent monitoring module, a template definition module, and a fault information generation module. The business layer is used to monitor the power grid operation data obtained by the data layer and intelligently assist in generating and reporting data.

[0087] The intelligent monitoring module is used to detect abnormal behaviors and potential faults in the power grid operation data information through machine learning algorithms (such as Isolation Forest, One-Class SVM), and issue an alarm in time when an abnormal situation is detected; record fault monitoring data, including timestamp, location, related parameters, etc., and store it in a distributed cache database to provide a basis for subsequent analysis.

[0088] The template definition module is used to define different types of fault templates, including fault type, fault characteristics, fault parameters and fault handling actions. The template definition is based on the IEC 61850 standard and is defined based on reporting tools, such as Tableau, Power BI and Crystal Reports. Fault types include 220kV, 110kV tripping and large-scale power outages in counties.

[0089] Fault information generation module, used to generate reporting data based on fault templates and fault monitoring data. The generated data reflects the actual power grid operation characteristics;

[0090] The fault information generation module includes a data analysis module and an information integration module. Figure 4 As shown; the data analysis module is used to analyze and process the fault monitoring data and remove useless information; the information integration module is used to combine the analysis and processing results with the fault template to synthesize the reporting data. Or Gaussian mixture model (GMM) technology

[0091] The information integration module generates fault data using a method based on a Gaussian mixture model (GMM). ARIMA, LSTM or Prophet time series models are used to generate fault time series data to simulate the change process before and after the fault occurs. The generated fault data is fused with the real data to form a comprehensive data set to ensure that the generated data is consistent with the real data in distribution. Feature selection techniques (such as PCA and LASSO) are used to extract key features from the generated fault time series data, such as voltage and current waveforms before and after the fault occurs, and protection action time. Hadoop, Spark and other big data processing frameworks are used to fill the extracted key features into the corresponding fields according to the defined fault templates, and the filled fault templates are combined into complete reporting data to ensure that the data format meets the standard requirements, such as JSON, XML or CSV format, and is stored in a distributed column database.

[0092] The method for generating fault data using a Gaussian mixture model (GMM)-based method is:

[0093] A set of power grid data obtained after analysis by the data analysis module is X, where each sample x n It contains multiple features, such as voltage, current, power, etc. Each sample also contains the label of fault type and fault feature.

[0094] First, construct a Gaussian mixture model (GMM), as follows:

[0095]

[0096] Where K is the number of Gaussian distributions, that is, the number of mixed components, which refers to the fault type, and π k is the weight of the kth Gaussian distribution, satisfying ρ() is the probability density function of the kth Gaussian distribution, with mean μ k The covariance matrix is ​​σ k ;

[0097] Initialize weight π k , mean μ k , covariance matrix σ k ;

[0098] (Step E) Calculate the posterior probability that each sample belongs to each Gaussian distribution:

[0099]

[0100] Among them, x n is the nth sample; γ(z nk ) is the posterior probability that n samples belong to the kth Gaussian distribution;

[0101] (M step) Update weight πk , mean μ k , covariance matrix σ k ;

[0102]

[0103] Where N is the total number of samples;

[0104] Repeat the E and M steps until the parameters converge or the predetermined number of iterations is reached, 100 times.

[0105] After the above model training is completed, select Gaussian distribution: according to the weight π k Randomly select a Gaussian distribution μ k ρ(x|μ k ,σ k );

[0106] Generate samples: From the selected Gaussian distribution π k ρ(x|μ k ,σ k ) to generate a sample.

[0107] Add fault labels: According to the characteristics of the generated samples, assign corresponding fault types and fault feature labels to them.

[0108] Through the above methods, the fault response speed, accuracy and resource utilization efficiency can be greatly improved while improving data quality.

[0109] The method based on Gaussian mixture model (GMM) is used to generate fault data, which can effectively capture different fault types: there may be multiple fault types in the power grid data, such as 220kV tripping, 110kV tripping and large-scale power outages in counties. GMM can capture these different types of faults, and each Gaussian distribution can represent a specific fault mode. Different operating states can be identified: the power grid may have different characteristic distributions under different operating states (such as normal operation, light load, peak load, etc.). GMM can identify and model these different operating states. There may be noise in the power grid data. GMM shows good robustness when processing noisy data, and can adapt to different degrees of noise by adjusting the number of mixed components.

[0110] The service layer includes a notification and reminder module and a text message sending module, which are used to interact with users. The service layer is used to transmit the reporting data generated by the business layer;

[0111] Notification module, real-time notification; when the system detects an abnormal situation or failure, it sends notifications to relevant personnel through multiple channels (such as email, system message, mobile application push). According to the severity and type of the event, the notification is classified so that users can quickly identify and respond. All notification events are recorded, including timestamp, event description, recipient and other information and stored in a distributed column database.

[0112] SMS sending module, used for emergency notification: when a serious failure occurs or an immediate response is required, emergency notifications are sent to key personnel via SMS; support batch sending of SMS to ensure that all relevant personnel can receive the information in a timely manner; track the sending status of SMS (such as sent, delivered, and failed to send) and record relevant information; record the logs of all SMS sending, including sending time, recipient, and content, and store them in a distributed column database.

[0113] The above notification methods are both real-time and timely, ensuring that the reported data is accurate, complete and notified to the dispatch personnel in a timely manner. The data reporting function improves the information reporting process, improves the quality and efficiency of information reporting, provides a basis for dispatching decisions, and improves the safety and reliability of the power grid. The reported data sent by notification reminders and SMS messages is as follows: Figure 6 shown.

[0114] Example 2

[0115] like Figure 5 As shown, a method for intelligent generation and reporting control management based on power grid big data includes the following steps:

[0116] S1: Use the data layer to obtain power grid data and establish a power grid operation data information pool;

[0117] Specifically, the data acquisition module uses a general API interface to connect with the data source to obtain a variety of raw data; the preprocessing module is used to clean, convert and organize the various raw data to obtain a variety of preprocessed raw data; the data integration module is used to integrate the preprocessed raw data into multi-source heterogeneous data to obtain unified and standardized power grid operation data, and store them in a distributed column database to obtain a power grid operation data information pool;

[0118] S2: Use the business layer to intelligently assist in generating reporting data based on the power grid operation data obtained from the data layer;

[0119] Specifically, the intelligent monitoring module is used to monitor the power grid operation data and record the fault monitoring data; the template definition module is used to define the fault type, fault characteristics, and fault parameters; the data analysis module is used to analyze and process the fault monitoring data and remove useless information; the information integration module is used to combine the analysis and processing results with the fault template to synthesize the reporting data;

[0120] S3: Use the service layer to monitor and notify the reporting data assisted by the business layer;

[0121] Specifically, the notification reminder module is used to send real-time notifications of abnormalities and failures to relevant personnel through system pop-ups based on the reported data; the SMS sending module is used to send emergency notifications of abnormalities and failures to key personnel through SMS based on the reported data.

[0122] The present invention establishes a multi-source heterogeneous data fusion and analysis system through multi-source heterogeneous data integration and big data acquisition and analysis technology, quickly perceives primary information, which usually refers to the original data directly collected from sensors or devices, such as electrical parameters (voltage, current, power, etc.) and switch states in the SCADA system; intelligently extracts secondary signals, which refer to information obtained after processing and analyzing primary information, such as fault detection; retrieves relevant data such as D5000 and OMS in real time, realizes automatic identification of faults based on necessary information and rules for fault identification, automatically generates fault notification content, reminds the dispatcher of the current fault situation of the power grid, issues notification reminders, and strengthens reporting of fault power outage information.

[0123] The other structures and principles are the same as those in Example 1.

[0124] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent generation and reporting control management system based on power grid big data, characterized in that: include: The data layer is used to obtain power grid data and establish a power grid operation data information pool; The business layer is used to monitor the power grid operation data obtained by the data layer and intelligently assist in generating and reporting data; The service layer is used to transmit the reporting data generated with the assistance of the business layer.

2. The intelligent generation and reporting control management system based on power grid big data according to claim 1 is characterized in that: The data layer includes a data acquisition module, a data preprocessing module, a data integration module and a data storage module connected in sequence; The data acquisition module is used to collect various raw data from various sources of the power grid; The data preprocessing module is used to preprocess the various collected raw data; The data integration module is used to integrate the pre-processed raw data into multi-source heterogeneous data to obtain unified and standardized power grid operation data; The data storage module is used to store original power grid data, structured data and cache data.

3. The intelligent generation and reporting control management system based on power grid big data according to claim 2 is characterized in that: The multiple sources include SCADA, EMS, DMS, OMS, and D5000 systems; the multiple original data include real-time collected power grid operation data, power grid energy management data, distribution system data, operation management system data, and dispatching automation system data.

4. The intelligent generation and reporting control management system based on power grid big data according to claim 2 or 3 is characterized in that: The data acquisition module adopts a universal API interface and uses message queue technology to receive data streams from different systems; The data preprocessing module includes a format conversion unit, a de-manipulation processing unit and a cluster analysis unit; A format conversion unit, used to convert a variety of raw data in different formats into a unified standard format; A denoising processing unit, used for denoising a variety of raw data in a unified standard format by using a denoising processing method based on wavelet transform; A clustering analysis unit, used for classifying and dividing the noise-free power grid operation data obtained by the denoising processing unit by using a clustering algorithm; The data integration module uses the ETL method to organize the data processed by the cluster analysis unit, and uses a distributed computing method to associate and fuse the organized data to achieve multi-source heterogeneous data integration.

5. The intelligent generation and reporting control management system based on power grid big data according to claim 4 is characterized in that: The data storage module includes a distributed file system, a distributed column database and a distributed cache database; Distributed file system is used to store raw power grid data; The distributed column database is used to store the structured data, which includes pre-processed data, data after integration of multi-source heterogeneous data, reporting data, notification reminder data and SMS sending data; The distributed cache database is used to store cache data.

6. The intelligent generation and reporting control management system based on power grid big data according to claim 5 is characterized in that: The business layer includes an intelligent monitoring module, a template definition module and a fault information generation module; The intelligent monitoring module is used to monitor the power grid operation data, obtain fault monitoring data and store it in the distributed cache database; The intelligent monitoring module detects abnormal behaviors and potential faults in the power grid through a machine learning algorithm, and the machine learning algorithm adopts an Isolation Forest or a One-Class SVM; The template definition module is used to define different types of fault templates, including fault type, fault characteristics, fault parameters and fault handling actions; fault types include 220kV, 110kV tripping and county-wide large-scale power outages; The fault information generation module is used to generate reporting data based on the fault template and fault monitoring data; The fault information generation module includes a data analysis module and an information integration module; the data analysis module is used to analyze and process the fault monitoring data and remove useless information; the information integration module is used to combine the analysis and processing results with the fault template to synthesize the reporting data, and the information integration module uses a method based on a Gaussian mixture model to generate the reporting data.

7. The intelligent generation and reporting control management system based on power grid big data according to claim 6 is characterized in that: The method based on Gaussian mixture model is as follows: a set of power grid data obtained after analysis by the data analysis module is X, where each sample x n Contains multiple features, such as voltage, current, power, etc. Each sample also contains the label of fault type and fault feature; First, construct a Gaussian mixture model: Where K is the number of Gaussian distributions, π k is the weight of the kth Gaussian distribution, satisfying p() is the probability density function of the kth Gaussian distribution, with mean μ k , the covariance matrix is ​​σ k ; Initialize weight π k , mean μ k , covariance matrix σ k ; Step E: Calculate the posterior probability that each sample belongs to each Gaussian distribution: Among them, γ(z nk ) is the posterior probability that n samples belong to the kth Gaussian distribution; M step: Update weight π k , mean μ k , covariance matrix σ k ; Repeat steps E and M until the parameters converge or the predetermined number of iterations is reached to obtain a trained model; According to the weight π corresponding to the trained model k Randomly select a Gaussian distribution; generate a sample from the selected Gaussian distribution; and assign corresponding fault types and fault feature labels to the generated sample according to the characteristics of the generated sample.

8. The intelligent generation and reporting control management system based on power grid big data according to claim 7 is characterized in that: The service layer includes a notification reminder module and a text message sending module; The notification reminder module is used to send real-time notifications of abnormalities and faults to relevant personnel through system pop-ups based on the reported data; the SMS sending module is used to send emergency notifications of abnormalities and faults to key personnel through SMS based on the reported data.

9. A method for intelligent generation, reporting and control management based on power grid big data, characterized in that: The intelligent generation and reporting control management system based on power grid big data as described in claim 8 comprises the following steps: S1: Use the data layer to obtain power grid data and establish a power grid operation data information pool; S2: Use the business layer to intelligently assist in generating reporting data based on the power grid operation data obtained from the data layer; S3: Use the service layer to monitor and notify the reporting data assisted by the business layer.

10. The intelligent generation and reporting control management method based on power grid big data according to claim 9 is characterized in that: Step S1 is specifically as follows: a data acquisition module is used to connect with a data source using a general API interface to obtain a variety of raw data; a preprocessing module is used to clean, convert and organize a variety of raw data to obtain a variety of preprocessed raw data; a data integration module is used to integrate the preprocessed raw data from multiple sources and heterogeneous data to obtain unified and standardized power grid operation data, and store the data in a distributed column database to obtain a power grid operation data information pool; Step S2 specifically includes using the intelligent monitoring module to monitor the power grid operation data and record the fault monitoring data; using the template definition module to define the fault type, fault characteristics, and fault parameters; using the data analysis module to analyze and process the fault monitoring data and remove useless information; using the information integration module to combine the analysis and processing results with the fault template to synthesize the reporting data; Step S3 specifically uses the notification reminder module to send real-time notifications of abnormalities and failures to relevant personnel through a system pop-up window based on the reported data; and uses the SMS sending module to send emergency notifications of abnormalities and failures to key personnel through SMS based on the reported data.

Images