Power supply quality service risk prevention and control system based on big data application

Abstract

The invention discloses a power supply quality service risk prevention and control system based on big data application. The power supply quality service risk prevention and control system comprises an Ali cloud platform, a technical medium station, a micro-service architecture, a data medium station and a front end, the Ali cloud platform is used for deploying basic resources of big data calculation; the technical platform is used for login verification; the micro-service architecture divides the application of the system into a plurality of micro-services; the data middle platform provides a unified middle platform application of data access, data cleaning, data modeling and data mining, and data sharing among multiple applications is achieved; and the front end realizes bidirectional binding of data. The system disclosed by the invention can intelligently analyze the frequent power failure risk of a transformer area and a customer, analyze the abnormal voltage quality risk of the transformer area and the customer, analyze the power supply micro-service risk, carry out risk panoramic display, and accurately guide a grassroots unit to carry out risk prevention and control.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid power supply risk prevention and control, and specifically relates to a power supply quality service risk prevention and control system based on big data application. Background Art

[0002] Power supply service risks mainly include power outage risk and voltage fluctuation risk. These risks may damage user equipment, affect the production process, cause economic losses, and affect the reputation and image of the power supply company.

[0003] Power outage risk is one of the most common problems in power supply services, which may cause damage to user equipment, production stagnation, economic losses, etc. Risk points include natural disasters, power facility failures, line maintenance, power dispatch errors, etc. Voltage fluctuation risk can cause damage to user equipment and power systems, affecting the continuity of the production process and product quality. Risk points include sudden increase in grid load, lightning, electrical equipment failure, etc.

[0004] At present, power supply service risk analysis is mainly carried out by analyzing the quality of employees, daily work processes, and the construction of mechanisms and emergency mechanisms, so as to obtain a macro-level power supply risk analysis report. However, the power supply risk analysis obtained in this way is difficult to accurately analyze the problem, the analysis cycle is long, and it is impossible to provide differentiated power supply microservices based on customer power demand. Summary of the invention

[0005] The purpose of the present invention is to provide a power supply quality service risk prevention and control system based on big data application, which can analyze the massive historical data and real-time data generated during the operation of power equipment, accurately analyze power outages and voltage problems, and accurately predict risks, thereby realizing differentiated power supply microservices for customers and completing the prevention and control of power supply quality service risks.

[0006] The present invention provides a power supply quality service risk prevention and control system based on big data application, including Alibaba Cloud platform, technical middle platform, microservice architecture, data middle platform and front end;

[0007] The Alibaba Cloud platform is used to deploy basic resources for big data computing, including distributed containers (k8s), real-time and offline computing components (datawork), and relational databases (rds);

[0008] The technical middle platform is used to verify the login of the power supply quality service risk prevention and control system based on big data application, realize the login of the whole network with a single account, and avoid repeated authentication; the technical middle platform adopts single sign-on verification, verifies the same isc account password, creates and returns a token as a unified credential for judging authority;

[0009] Nacos is used as the registration center configurator and the springcloud architecture is used to establish a distributed microservice framework; the microservice architecture divides the application of the power supply quality service risk prevention and control system based on big data application into several microservices, realizing the high cohesion and low coupling state of the application, which is convenient for subsequent system upgrades and maintenance;

[0010] The data center provides a unified data access, data cleaning, data modeling, and data mining center application to achieve data sharing among multiple applications; the data center integrates a distributed real-time offline data analysis and processing framework. Data is acquired through equipment, personnel, and data mining, and written into the RDS source layer data table in real time through Flink real-time analysis. Offline data is written into the RSD source layer data table through Spark analysis and processing;

[0011] The front end adopts the vue framework and realizes two-way binding of data through the MVVM concept.

[0012] The front end adopts the vue framework and realizes two-way binding of data through the MVVM concept, specifically:

[0013] The association between data and template is realized through the vue instance; if the data is updated in vue, the corresponding rendering content in the page is automatically updated; if the data of the variable rendered by the updated page changes in vue, the data of this variable in data will automatically change accordingly;

[0014] Use virtual DOM to avoid tedious DOM operations, improve development efficiency through component-based development, facilitate reuse, simplify debugging steps, improve overall maintainability, and facilitate collaborative development.

[0015] The microservice framework adopts the springcloud distributed microservice framework, including gateway module, business module, login module, and power supply service risk prevention and control module;

[0016] The gateway module is responsible for distributing requests to ensure the working efficiency of each node; the gateway module uses nacos as a unified configuration file to achieve dynamic increase and decrease of configuration; the nodes of the microservices are expanded and reduced according to the business scenarios, and when a node goes down, a spare node is used to replace it;

[0017] The business module classifies the data, specifically:

[0018] Data is divided into directly used data, data to be analyzed, and historical data; directly used data refers to data that has been processed by the middle platform, and is used by directly calling the middle platform interface; data to be analyzed uses the Quartz scheduled task tool to regularly aggregate data into the RDS database, and is aggregated into the business table through analysis; historical data is stored in redis for caching;

[0019] The login module verifies the token obtained from the technical center, determines its validity, stores the valid token in redis, and sets an expiration date. After the expiration date, the user needs to obtain the token again to log in.

[0020] The power supply service risk prevention and control module includes a power outage warning module, a voltage warning module, a sensitive user module and a warning risk prediction module;

[0021] The power outage warning module includes a user power outage warning module and an area power outage warning module. By analyzing the input power outage data, the corresponding power outage warning level is obtained;

[0022] The voltage warning module includes the user voltage warning module and the substation voltage warning module, which analyzes the input voltage fluctuation data to obtain the corresponding voltage warning level;

[0023] The sensitive user module collects statistics on user feedback data, selects users who have provided feedback more than a preset number of times as sensitive users, and stores them in the sensitive user table;

[0024] The early warning risk prediction module is based on the Long short-term memory (LTSM) deep learning algorithm model. By inputting the historical risk sequence, it predicts the number of power supply service risks for the next day.

[0025] The user power outage warning module includes a user power outage warning offline branch and a user power outage warning real-time branch. The user power outage warning condition table obtained by the user power outage warning offline branch is associated with the real-time condition data obtained by the user power outage warning real-time branch to finally obtain the user power outage warning level.

[0026] The user power outage warning offline branch operation process includes the following steps:

[0027] Get the user profile table for screening and obtain the screened user profile;

[0028] The filtered user profile is associated with the electric energy meter asset table to obtain the user's electric energy meter ID;

[0029] Associate the user's energy meter ID with the energy meter power outage schedule to obtain the user's daily power outage details;

[0030] Count the number of power outages per month by the daily power outage details table of the users, and obtain a monthly power outage statistics table for the users;

[0031] The user monthly power outage statistics table, warning condition configuration table and sensitive user table are associated to obtain the user power outage warning condition table, which is a daily partition table with a daily update frequency;

[0032] The specific operation process of the real-time branch of user power outage warning is as follows: the user profile table is associated with the real-time power outage event to obtain the detailed data of the user power outage on that day, and then the detailed data of the user power outage on that day is filtered to obtain the real-time conditional data that meets the conditions;

[0033] The power outage warning module includes an offline branch and a real-time branch. The power outage warning condition table obtained by the offline branch is associated with the real-time condition data obtained by the real-time branch, and the power outage warning level is finally obtained.

[0034] The offline branch operation process of the power outage warning in the substation area includes the following steps:

[0035] Obtain the area archive table for screening, and obtain the screened area archive;

[0036] The screened substation files are associated with the public substation meter relationship table and the conversion substation meter relationship table to obtain the electric energy meter ID of the public substation;

[0037] Associate the power meter ID of the public substation area with the power outage schedule of the power meter to obtain the daily power outage details of the substation area;

[0038] Count the number of power outages per month in the daily power outage details table of the substation area to obtain a monthly power outage statistics table of the substation area;

[0039] The monthly power outage statistics table and the warning condition configuration table of the substation area are associated to obtain the power outage warning condition table of the substation area. The table is a daily partition table with a daily update frequency.

[0040] The specific operation process of the real-time branch of the power outage warning in the substation area is as follows: the substation area archive table and the real-time power outage event are associated to obtain the detailed data of the power outage in the substation area on that day, and then the detailed data of the power outage in the substation area on that day is filtered to obtain the real-time conditional data that meets the conditions;

[0041] The user voltage warning module includes a user outlet voltage warning module and a user single-phase voltage push warning module. The user outlet voltage detail table obtained by the user outlet voltage warning module and the user voltage push detail table obtained by the user single-phase voltage push warning module are associated with the user voltage quality warning configuration table to filter out the detailed data that meets the conditions; then the sensitive user table is combined for processing, and the user basic information, warning level and warning description information are stored in the user voltage warning condition table; finally, the user's voltage quality warning result is generated according to the warning level and the detailed data that meets the conditions, and the user voltage quality warning result table is obtained;

[0042] The operation process of the user outlet voltage warning module includes the following steps:

[0043] The user profile table is associated with the voltage indication table to obtain the voltage indication details of the user's daily preset points;

[0044] Based on the obtained voltage indication details of the user's daily preset points and the user's outlet and voltage, the user whose voltage appears within the preset range for a preset number of consecutive voltage indication points is screened out;

[0045] The basic information, voltage value and occurrence month of the screened users are stored in the user export voltage detail table, which is a monthly partition table and the update frequency is to update the data of the current month every day;

[0046] The operation process of the user single-phase voltage push warning module includes the following steps:

[0047] The user profile table is associated with the voltage indication table to obtain the voltage indication details of the user's daily preset points;

[0048] Based on the details of the voltage indication values ​​at the preset points of the user every day, the users whose voltages at the preset number of voltage indication points continuously appear within the preset range are screened out;

[0049] The filtered user data is stored in the user voltage push details table, which is a daily partition table and the update frequency is to update the data of the current month every day;

[0050] The substation voltage warning module includes the substation contiguous user voltage warning module, the substation outlet voltage warning module and the substation single-phase voltage push warning module; the output results of the substation contiguous user voltage warning module, the substation outlet voltage warning module and the substation single-phase voltage push warning module are associated with the user warning configuration table, and the detailed data that meets the conditions is screened; then combined with the sensitive user table for processing, the substation basic information, warning level and warning description information are stored in the substation voltage warning condition table; finally, the voltage quality warning result of the substation is generated according to the warning level and the detailed data that meets the conditions, and the substation voltage quality warning result table is obtained;

[0051] The specific operation process of the voltage early warning module for users in the area is as follows:

[0052] Based on the user outlet voltage detailed table, the user information of the same substation at the same time in the user outlet voltage detailed table is obtained by statistics, and the corresponding users are associated with the substation number to obtain the number of abnormal users in each substation; finally, according to the number of abnormal users in each substation, the substations with the number of abnormal users in the same substation at the same time greater than the preset threshold or greater than the preset percentage of the total number of users in the corresponding substation are screened and stored in the substation contiguous user voltage detailed table, which is a monthly partition table, and the update frequency is to update the data of the current month every day;

[0053] The operation process of the transformer area outlet voltage warning module includes the following steps:

[0054] Link the substation archive table with the voltage indicator table to obtain the voltage indication details of the daily preset points in the substation;

[0055] For the obtained voltage indication details of the preset points in the substation every day, based on the outlet and voltage of the substation, the substation where the voltage at the preset number of consecutive voltage indication points falls within the preset range is selected;

[0056] The basic information, voltage value and occurrence month of the screened substation are stored in the substation outlet voltage details table. This table is a monthly partition table, and the update frequency is to update the data of the current month every day.

[0057] The operation process of the single-phase voltage push warning module in the substation area includes the following steps:

[0058] Link the substation archive table with the voltage indicator table to obtain the voltage indication details of the daily preset points in the substation;

[0059] Based on the voltage indication details of the preset points in the substation every day, filter out the substations where the voltage at the preset number of consecutive voltage indication points falls within the preset range;

[0060] The screened substation data is stored in the substation voltage push details table, which is a daily partition table and the update frequency is to update the data of the current month every day;

[0061] The present invention discloses a power supply quality service risk prevention and control system based on big data application, which can intelligently analyze the risks of frequent power outages in substations and customers, analyze the risks of abnormal voltage quality in substations and customers, analyze the risks of power supply microservices, display a panoramic view of risks, and accurately guide grassroots units to carry out risk prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the structure of the system of the present invention. DETAILED DESCRIPTION

[0063] The present invention provides a power supply quality service risk prevention and control system based on big data application, and its structural diagram is as follows: Figure 1As shown, it includes Alibaba Cloud platform, technical middle platform, microservice architecture, data middle platform and front end;

[0064] The Alibaba Cloud platform is used to deploy basic resources for big data computing, including distributed containers (k8s), real-time and offline computing components (datawork), and relational databases (rds);

[0065] The technical middle platform is used to verify the login of the power supply quality service risk prevention and control system based on big data application, realize the login of the whole network with a single account, and avoid repeated authentication; the technical middle platform adopts single sign-on verification, verifies the same isc account password, creates and returns a token as a unified credential for judging authority;

[0066] Nacos is used as the registration center configurator and the springcloud architecture is used to establish a distributed microservice framework; the microservice architecture divides the application of the power supply quality service risk prevention and control system based on big data application into several microservices, so as to achieve a high cohesion and low coupling state of the application, and facilitate the subsequent system upgrade and maintenance work; the several microservices are decoupled from each other, and if a single service process hangs, it will not affect other processes;

[0067] The data center provides a unified data access, data cleaning, data modeling, and data mining center application to achieve data sharing among multiple applications; the data center integrates a distributed real-time offline data analysis and processing framework. Data is acquired through equipment, personnel, and data mining, and written into the RDS source layer data table in real time through Flink real-time analysis. Offline data is written into the RSD source layer data table through Spark analysis and processing;

[0068] The front end adopts the vue framework and realizes two-way binding of data through the MVVM concept.

[0069] The front end adopts the vue framework and realizes two-way binding of data through the MVVM concept, specifically:

[0070] The association between data and template is realized through the vue instance; if the data is updated in vue, the corresponding rendering content in the page is automatically updated; if the data of the variable rendered by the updated page changes in vue, the data of this variable in data will automatically change accordingly;

[0071] Use virtual DOM to avoid tedious DOM operations, improve development efficiency through component-based development, facilitate reuse, simplify debugging steps, improve overall maintainability, and facilitate collaborative development.

[0072] The microservice framework adopts the springcloud distributed microservice framework, including gateway module, business module, login module, and power supply service risk prevention and control module;

[0073] The gateway module is responsible for distributing requests to ensure the working efficiency of each node; the gateway module uses nacos as a unified configuration file to achieve dynamic increase and decrease of configuration; the nodes of the microservices are expanded and reduced according to the business scenarios, and when a node goes down, a spare node is used to replace it;

[0074] The business module classifies the data, specifically:

[0075] Data is divided into directly used data, data to be analyzed, and historical data; directly used data refers to data that has been processed by the middle platform, and is used by directly calling the middle platform interface; the data to be analyzed uses the Quartz scheduled task tool to regularly aggregate data into the RDS database, and then aggregates it into the business table through analysis; the Quartz scheduled task tool uses scheduled tasks to schedule data aggregation and synchronize data to the database, which is equivalent to an ETL tool; historical data is stored in redis for caching;

[0076] The login module verifies the token obtained from the technical center, determines its validity, stores the valid token in redis, and sets an expiration date. After the expiration date, the user needs to obtain the token again to log in.

[0077] The power supply service risk prevention and control module includes a power outage warning module, a voltage warning module, a sensitive user module and a warning risk prediction module;

[0078] The power outage warning module includes the user power outage warning module and the substation power outage warning module. By analyzing the historical power outage data, the corresponding power outage warning level is obtained;

[0079] The voltage warning module includes the user voltage warning module and the substation voltage warning module. By analyzing the historical voltage fluctuation data, the corresponding voltage warning level is obtained;

[0080] The sensitive user module collects statistics on user feedback data, selects users who have provided feedback more than a preset number of times as sensitive users, and stores them in the sensitive user table;

[0081] The early warning risk prediction module is based on the Long short-term memory (LTSM) deep learning algorithm model. By inputting the historical risk sequence, it predicts the number of power supply service risks for the next day.

[0082] The user power outage warning module includes a user power outage warning offline branch and a user power outage warning real-time branch. The user power outage warning condition table obtained by the user power outage warning offline branch is associated with the real-time condition data obtained by the user power outage warning real-time branch to finally obtain the user power outage warning level.

[0083] The user power outage warning offline branch operation process includes the following steps:

[0084] Get the user profile table for screening and obtain the screened user profile;

[0085] The filtered user profile is associated with the electric energy meter asset table to obtain the user's electric energy meter ID;

[0086] Associate the user's energy meter ID with the energy meter power outage schedule to obtain the user's daily power outage details;

[0087] Count the number of power outages per month by the daily power outage details table of the users, and obtain a monthly power outage statistics table for the users;

[0088] The user monthly power outage statistics table, warning condition configuration table and sensitive user table are associated to obtain the user power outage warning condition table, which is a daily partition table with a daily update frequency;

[0089] The specific operation process of the real-time branch of user power outage warning is as follows: the user profile table is associated with the real-time power outage event to obtain the detailed data of the user power outage on that day, and then the detailed data of the user power outage on that day is filtered to obtain the real-time conditional data that meets the conditions;

[0090] The power outage warning module includes an offline branch and a real-time branch. The power outage warning condition table obtained by the offline branch is associated with the real-time condition data obtained by the real-time branch, and the power outage warning level is finally obtained.

[0091] The offline branch operation process of the power outage warning in the substation area includes the following steps:

[0092] Obtain the area archive table for screening, and obtain the screened area archive;

[0093] The screened substation files are associated with the public substation meter relationship table and the conversion substation meter relationship table to obtain the electric energy meter ID of the public substation;

[0094] Associate the power meter ID of the public substation area with the power outage schedule of the power meter to obtain the daily power outage details of the substation area;

[0095] Count the number of power outages per month in the daily power outage details table of the substation area to obtain a monthly power outage statistics table of the substation area;

[0096] The monthly power outage statistics table and the warning condition configuration table of the substation area are associated to obtain the power outage warning condition table of the substation area. The table is a daily partition table with a daily update frequency.

[0097] The specific operation process of the real-time branch of the power outage warning in the substation area is as follows: the substation area archive table and the real-time power outage event are associated to obtain the detailed data of the power outage in the substation area on that day, and then the detailed data of the power outage in the substation area on that day is filtered to obtain the real-time conditional data that meets the conditions;

[0098] The user voltage warning module includes a user outlet voltage warning module and a user single-phase voltage push warning module. The user outlet voltage detail table obtained by the user outlet voltage warning module and the user voltage push detail table obtained by the user single-phase voltage push warning module are associated with the user voltage quality warning configuration table to filter out the detailed data that meets the conditions; then the sensitive user table is combined for processing, and the user basic information, warning level and warning description information are stored in the user voltage warning condition table; finally, the user's voltage quality warning result is generated according to the warning level and the detailed data that meets the conditions, and the user voltage quality warning result table is obtained;

[0099] The operation process of the user outlet voltage warning module includes the following steps:

[0100] The user profile table is associated with the voltage indication table to obtain the detailed voltage indication values ​​of 96 points of the user every day;

[0101] For the voltage indication details of the user's daily preset points, based on the user's outlet and voltage, screen out users with 8 consecutive voltage indication points with voltage higher than 260V or lower than 198V;

[0102] The basic information, voltage value and occurrence month of the screened users are stored in the user export voltage detail table, which is a monthly partition table and the update frequency is to update the data of the current month every day;

[0103] The operation process of the user single-phase voltage push warning module includes the following steps:

[0104] The user profile table is associated with the voltage indication table to obtain the detailed voltage indication values ​​of 96 points of the user every day;

[0105] Based on the voltage indication details of the user's daily preset points, select users with voltage values ​​between 265V and 300V or between 110V and 180V at four consecutive voltage indication points;

[0106] The filtered user data is stored in the user voltage push details table, which is a daily partition table and the update frequency is to update the data of the current month every day;

[0107] The substation voltage warning module includes the substation contiguous user voltage warning module, the substation outlet voltage warning module and the substation single-phase voltage push warning module; the output results of the substation contiguous user voltage warning module, the substation outlet voltage warning module and the substation single-phase voltage push warning module are associated with the user warning configuration table, and the detailed data that meets the conditions is screened; then combined with the sensitive user table for processing, the substation basic information, warning level and warning description information are stored in the substation voltage warning condition table; finally, the voltage quality warning result of the substation is generated according to the warning level and the detailed data that meets the conditions, and the substation voltage quality warning result table is obtained;

[0108] The specific operation process of the voltage early warning module for users in the area is as follows:

[0109] Based on the user outlet voltage detailed table, the user information of the same substation at the same time in the user outlet voltage detailed table is obtained by statistics, and the corresponding users are associated with the substation number to obtain the number of abnormal users in each substation; finally, according to the number of abnormal users in each substation, the substations with more than 10 abnormal users at the same time in the same substation or more than 10% of the total number of users in the corresponding substation are screened and stored in the substation contiguous user voltage detailed table, which is a monthly partition table, and the update frequency is to update the data of the current month every day;

[0110] The operation process of the transformer area outlet voltage warning module includes the following steps:

[0111] The substation archive table is linked to the voltage indicator table to obtain the voltage indication details of 96 points in the substation every day;

[0112] For the voltage indication details of the preset points of the substations obtained daily, based on the outlets and voltages of the substations, the substations with 8 consecutive voltage indication points showing voltages higher than 260V or lower than 198V were selected;

[0113] The basic information, voltage value and occurrence month of the screened substation are stored in the substation outlet voltage details table. This table is a monthly partition table, and the update frequency is to update the data of the current month every day.

[0114] The operation process of the single-phase voltage push warning module in the substation area includes the following steps:

[0115] The substation archive table is linked to the voltage indicator table to obtain the voltage indication details of 96 points in the substation every day;

[0116] Based on the voltage indication details of the preset points in the substation every day, the substations with voltage values ​​between 265V and 300V or between 110V and 180V at four consecutive voltage indication points are selected;

[0117] The screened substation data is stored in the substation voltage push details table, which is a daily partition table and the update frequency is daily update of the data for the current month.

[0118] Specifically, the Java code example using the LTSM deep learning algorithm model to predict the number of power supply service risks for the next day is as follows:

[0119]

[0120]

[0121]

Claims

1. A power supply quality service risk prevention and control system based on big data application, characterized in that: Including Alibaba Cloud platform, technical middle platform, microservice architecture, data middle platform and front end; The Alibaba Cloud platform is used to deploy basic resources for big data computing, including distributed containers (k8s), real-time and offline computing components (datawork), and relational databases (rds); The technical middle platform is used to verify the login of the power supply quality service risk prevention and control system based on big data application, realize the login of the whole network with a single account, and avoid repeated authentication; the technical middle platform adopts single sign-on verification, verifies the same isc account password, creates and returns a token as a unified credential for judging authority; Nacos is used as the registration center configurator and the springcloud architecture is used to establish a distributed microservice framework; the microservice architecture divides the application of the power supply quality service risk prevention and control system based on big data application into several microservices, realizing the high cohesion and low coupling state of the application, which is convenient for subsequent system upgrades and maintenance; The data center provides a unified data access, data cleaning, data modeling, and data mining center application to achieve data sharing among multiple applications; the data center integrates a distributed real-time offline data analysis and processing framework. Data is acquired through equipment, personnel, and data mining, and written into the RDS source layer data table in real time through Flink real-time analysis. Offline data is written into the RSD source layer data table through Spark analysis and processing; The front end adopts the vue framework and realizes two-way binding of data through the MVVM concept.

2. According to the power supply quality service risk prevention and control system based on big data application according to claim 1, it is characterized in that: The front end adopts the vue framework and realizes two-way binding of data through the MVVM concept, specifically: The association between data and template is realized through the vue instance; if the data is updated in vue, the corresponding rendering content in the page is automatically updated; if the data of the variable rendered by the updated page changes in vue, the data of this variable in data will automatically change accordingly; Use virtual DOM to avoid tedious DOM operations, improve development efficiency through component-based development, facilitate reuse, simplify debugging steps, improve overall maintainability, and facilitate collaborative development.

3. According to the power supply quality service risk prevention and control system based on big data application in claim 1, it is characterized in that: The microservice framework adopts the springcloud distributed microservice framework, including gateway module, business module, login module, and power supply service risk prevention and control module; The gateway module is responsible for distributing requests and ensuring the working efficiency of each node; The business module classifies the data; The login module verifies the login based on the token obtained by the technical center; The power supply service risk prevention and control module identifies different risks based on the substation and customer meter data.

4. The power supply quality service risk prevention and control system based on big data application according to claim 3 is characterized in that: The gateway module uses nacos as a unified configuration file to achieve dynamic increase and decrease of configuration; Microservice nodes are expanded or reduced according to business scenarios, and when a node goes down, a spare node is used to replace it.

5. The power supply quality service risk prevention and control system based on big data application according to claim 3 is characterized in that: The business module classifies the data, specifically: Data is divided into directly used data, data to be analyzed and historical data; directly used data is data that has been processed by the middle platform, and is used by directly calling the middle platform interface; data to be analyzed uses the Quartz scheduled task tool to regularly summarize the data into the RDS database, and is summarized into the business table through analysis; historical data is stored in Redis for caching.

6. The power supply quality service risk prevention and control system based on big data application according to claim 3 is characterized in that: The login module will verify the token obtained from the technical center, determine the validity, then store the valid token in redis and set an expiration time. After the validity period, the user needs to obtain the token again to log in.

7. The power supply quality service risk prevention and control system based on big data application according to claim 3 is characterized in that: The power supply service risk prevention and control module includes a power outage warning module, a voltage warning module, a sensitive user module and a warning risk prediction module; The power outage warning module includes a user power outage warning module and an area power outage warning module. By analyzing the input power outage data, the corresponding power outage warning level is obtained; The voltage warning module includes the user voltage warning module and the substation voltage warning module, which analyzes the input voltage fluctuation data to obtain the corresponding voltage warning level; The sensitive user module collects statistics on user feedback data, selects users who have provided feedback more than a preset number of times as sensitive users, and stores them in the sensitive user table; The early warning risk prediction module is based on the Long short-term memory deep learning algorithm model. By inputting the historical risk sequence, it predicts the number of power supply service risks for the next day.

8. The power supply quality service risk prevention and control system based on big data application according to claim 7 is characterized in that: The user power outage warning module includes a user power outage warning offline branch and a user power outage warning real-time branch. The user power outage warning condition table obtained by the user power outage warning offline branch is associated with the real-time condition data obtained by the user power outage warning real-time branch to finally obtain the user power outage warning level. The user power outage warning offline branch operation process includes the following steps: Get the user profile table for screening and obtain the screened user profile; The filtered user profile is associated with the electric energy meter asset table to obtain the user's electric energy meter ID; Associate the user's energy meter ID with the energy meter power outage schedule to obtain the user's daily power outage details; Count the number of power outages per month by the daily power outage details table of the users, and obtain a monthly power outage statistics table for the users; The user monthly power outage statistics table, warning condition configuration table and sensitive user table are associated to obtain the user power outage warning condition table, which is a daily partition table with a daily update frequency; The specific operation process of the real-time branch of user power outage warning is as follows: the user profile table is associated with the real-time power outage event to obtain the detailed data of the user power outage on that day, and then the detailed data of the user power outage on that day is filtered to obtain the real-time conditional data that meets the conditions; The power outage warning module includes an offline branch and a real-time branch. The power outage warning condition table obtained by the offline branch is associated with the real-time condition data obtained by the real-time branch, and the power outage warning level is finally obtained. The offline branch operation process of the power outage warning in the substation area includes the following steps: Obtain the area archive table for screening, and obtain the screened area archive; The screened substation files are associated with the public substation meter relationship table and the conversion substation meter relationship table to obtain the electric energy meter ID of the public substation; Associate the power meter ID of the public substation area with the power outage schedule of the power meter to obtain the daily power outage details of the substation area; Count the number of power outages per month in the daily power outage details table of the substation area to obtain a monthly power outage statistics table of the substation area; The monthly power outage statistics table and the warning condition configuration table of the substation area are associated to obtain the power outage warning condition table of the substation area. The table is a daily partition table with a daily update frequency. The specific operation process of the real-time branch of the power outage warning for the substation is as follows: the substation archive table and the real-time power outage event are associated to obtain the detailed data of the power outage in the substation on that day, and then the detailed data of the power outage in the substation on that day is filtered to obtain the real-time conditional data that meets the conditions.

9. The power supply quality service risk prevention and control system based on big data application according to claim 8 is characterized in that: The user voltage warning module includes a user outlet voltage warning module and a user single-phase voltage push warning module. The user outlet voltage detail table obtained by the user outlet voltage warning module and the user voltage push detail table obtained by the user single-phase voltage push warning module are associated with the user voltage quality warning configuration table to filter out the detailed data that meets the conditions. Then, the sensitive user table is processed and the user basic information, warning level and warning description information are stored in the user voltage warning condition table; Finally, the user's voltage quality warning result is generated according to the warning level and the detailed data that meets the conditions, and the user's voltage quality warning result table is obtained; The operation process of the user outlet voltage warning module includes the following steps: The user profile table is associated with the voltage indication table to obtain the voltage indication details of the user's daily preset points; Based on the obtained voltage indication details of the user's daily preset points and the user's outlet and voltage, the user whose voltage appears within the preset range for a preset number of consecutive voltage indication points is screened out; The basic information, voltage value and occurrence month of the screened users are stored in the user export voltage detail table, which is a monthly partition table and the update frequency is to update the data of the current month every day; The operation process of the user single-phase voltage push warning module includes the following steps: The user profile table is associated with the voltage indication table to obtain the voltage indication details of the user's daily preset points; Based on the details of the voltage indication values ​​at the preset points of the user every day, the users whose voltages at the preset number of voltage indication points continuously appear within the preset range are screened out; The filtered user data is stored in the user voltage push details table, which is a daily partition table and the update frequency is to update the data of the current month every day; The substation voltage warning module includes the substation contiguous user voltage warning module, the substation outlet voltage warning module and the substation single-phase voltage push warning module; the output results of the substation contiguous user voltage warning module, the substation outlet voltage warning module and the substation single-phase voltage push warning module are associated with the user warning configuration table, and the detailed data that meets the conditions is screened; then combined with the sensitive user table for processing, the substation basic information, warning level and warning description information are stored in the substation voltage warning condition table; finally, the voltage quality warning result of the substation is generated according to the warning level and the detailed data that meets the conditions, and the substation voltage quality warning result table is obtained; The specific operation process of the voltage early warning module for users in the area is as follows: Based on the user outlet voltage detailed table, the user information of the same substation at the same time in the user outlet voltage detailed table is obtained by statistics, and the corresponding users are associated with the substation number to obtain the number of abnormal users in each substation; finally, according to the number of abnormal users in each substation, the substations with the number of abnormal users in the same substation at the same time greater than the preset threshold or greater than the preset percentage of the total number of users in the corresponding substation are screened and stored in the substation contiguous user voltage detailed table, which is a monthly partition table, and the update frequency is to update the data of the current month every day; The operation process of the transformer area outlet voltage warning module includes the following steps: Link the substation archive table with the voltage indicator table to obtain the voltage indication details of the daily preset points in the substation; For the obtained voltage indication details of the preset points in the substation every day, based on the outlet and voltage of the substation, the substation where the voltage at the preset number of consecutive voltage indication points falls within the preset range is selected; The basic information, voltage value and occurrence month of the screened substation are stored in the substation outlet voltage details table. This table is a monthly partition table, and the update frequency is to update the data of the current month every day. The operation process of the single-phase voltage push warning module in the substation area includes the following steps: Link the substation archive table with the voltage indicator table to obtain the voltage indication details of the daily preset points in the substation; Based on the voltage indication details of the preset points in the substation every day, filter out the substations where the voltage at the preset number of consecutive voltage indication points falls within the preset range; The screened substation data is stored in the substation voltage push details table, which is a daily partition table and the update frequency is daily update of the data for the current month.

Images