Cloud and edge integrated time series data processing system and method

Abstract

The invention belongs to the technical field of computer application, and particularly relates to a cloud and edge integrated time sequence data processing system. Deploying a power plant side time sequence database in each power plant, and collecting and storing time sequence data of each power plant in a local time sequence database of the power plant; a center side time sequence database is deployed in the data center, and the power plant side time sequence database is uploaded and accessed to the center side time sequence database according to needs; and deploying a time sequence data processing system, and performing unified management on the power plant side time sequence data and the center side time sequence data. Efficient management and service of the time series data of the nuclear power industry internet platform are achieved, the time series data management efficiency is improved, the data service quality is improved, and the data use threshold is reduced.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of computer application, and particularly relates to a cloud-edge integrated time series data processing system and method. BACKGROUND

[0002] Measurement data is data for perceiving the operation state of a nuclear power plant, promoting the digitalization and intelligentization management of the nuclear power plant, and a large amount of measurement data is generated by the nuclear power plant in the operation stage. The measurement data mainly in the form of online measurement data is also called time series data, and effective processing and analysis of the time series data have important significance for the daily operation of the nuclear power plant. At present, the China Nuclear Power Internet Platform DHP has been built by the China Nuclear Power Operation Technology Co., Ltd. as one of the basic data platforms of digital nuclear power, and the DHP platform needs to access, store and process time series data from different nuclear power plants and provide unified time series data services to applications on the basis. Since the DHP platform is a total and partial structure, it is deployed in the data center and each nuclear power base. In order to collect and converge data, data collection tools, time series databases, time series data service software and the like need to be deployed in each nuclear power plant, in addition, other supporting tools such as operation and maintenance management tools, configuration management tools, data analysis tools and the like are also needed for the convenience of management and maintenance. The deployment and daily operation of these dispersed components and tools need to consume a large amount of manpower, material resources and financial resources. In order to effectively support the access, storage, processing and service requirements of the massive time series data of the DHP platform, it is necessary to build a cloud-edge integrated time series data processing system based on the basic environment and data resources in the DHP platform, realize unified management of nuclear power time series data, improve the processing and distribution efficiency of time series data resources, improve the data service quality and reduce the data use threshold. SUMMARY

[0003] The application aims to provide a cloud-edge integrated time series data processing system and method to realize efficient management and service of time series data of a nuclear power internet platform, improve the time series data management efficiency, improve the data service quality and reduce the data use threshold.

[0004] To achieve the above-mentioned purpose, the technical scheme adopted by the application is as follows:

[0005] A cloud-edge integrated time series data processing system, a power plant side time series database is deployed in each nuclear power plant, time series data of each nuclear power plant is collected and stored in the local time series database of the power plant, a center side time series database is deployed in the data center, the power plant side time series database is uploaded and accessed in the center side time series database as needed, and a time series data processing system is deployed to realize unified management of the power plant side time series data and the center side time series data.

[0006] The time series data processing system is deployed in a server of the center side or any power plant side, and manages and accesses the time series database instances of the center side and each power plant side.

[0007] The timing data processing system is compatible with single machine version, double active version and distributed version timing database. The stored data is mounted to multiple SSD solid state disks. The underlying hardware storage supports planning and design using RAID method.

[0008] The timing database on the power plant side supports the storage of power plant DCS system, PI system, sensor data, inspection data, chemical data and calculation point data.

[0009] The timing database on the power plant side parses the received data and places different types of data in different storage areas. For structured data, it is converted into timing data storage format and then stored.

[0010] The timing database on the power plant side registers the timing data interface on demand in the industrial service open platform. Through the industrial service open platform, the timing data service is opened, supporting access authentication, concurrent access control and data access range control of timing data.

[0011] The timing database on the power plant side uses one or more channels for data upload and access according to the data size of each nuclear power plant.

[0012] The timing database on the power plant side accesses data according to power plants, units and measurement points.

[0013] The timing database on the power plant side controls the granularity of authorized data according to the needs of different users. For administrator users, it grants global data rights. For power plant users, it grants database rights. For applications, it grants entity device rights. For test personnel, it grants measurement point rights.

[0014] The timing database on the center side collects and accesses timing data of each power plant on demand, supports management of all power plant full measurement point metadata, and supports data permission granularity control at four levels of global, database, entity and measurement point.

[0015] The timing data in the timing database on the power plant side and the timing database on the center side serve as each other's copy. When one party loses data or other data is inconsistent, the other party's data is used for completion and repair. Multiple timing databases on the power plant side and the timing database on the center side synchronize real-time data and historical data at the measurement point level and the storage file level through synchronization settings.

[0016] The timing data interface is deployed in a way that one interface corresponds to one microservice, or multiple interfaces are placed in one microservice for deployment.

[0017] The time series data interface includes: obtaining attribute information of a single or multiple measuring points; obtaining index information of a single or multiple high-frequency measuring points within a time period; obtaining the latest value of the actual value of the latest timestamp of a single or multiple measuring points; obtaining the interpolation value of a single or multiple measuring points at a specified time point; obtaining all actual values of a single or multiple measuring points within a specified time period; obtaining the interpolation value of a single or multiple measuring points within a specified time period and at a specified time interval; obtaining the aggregation query result of a single or multiple measuring points within a specified time period; writing point value to a time series database; batch writing point value to a time series database; obtaining the original value of high-frequency data at a time point; measuring point registration.

[0018] Determine whether the power plant side time series data needs to be accessed to the center side time series database. If yes, upload the data through the MQTT protocol.

[0019] When using the MQTT protocol for data upload, determine the data size that needs to be uploaded. If all measuring point data of a certain power plant needs to be uploaded to the center side, multiple MQTT connections need to be configured, and the data of each unit is uploaded through an MQTT connection.

[0020] Register the time series data interface: fill in the gateway address, service address, request method, interface name information of the data API, complete the measuring point registration, and open the registered data interface to the outside through the time series data processing system. Access authentication configuration is performed, online verification is performed in the JWT+Token mode, and if the verification is passed, the system data can be accessed. Through the industrial service opening platform, the interface is added in the platform, the service name, interface address, interface description, request type, maximum number of supported connections, and maximum number of supported requests within a specified time period are configured, the audit personnel perform audit, and after the audit is passed, the platform generates a Token bound to the user and the interface, and specifies the effective start time and the effective end time of the Token. When the user uses the interface, the corresponding Token is added in the request header to access the system data.

[0021] Data access: power plant applications access data in the time series data processing system through a unified time series data service. Data requests are forwarded to the data gateway, which checks whether the current system exceeds the allowed number of concurrent accesses per second according to the configuration parameters. If it exceeds, an error code and corresponding prompt information are returned to the user. If it is within the allowed range, the data gateway parses the request message and performs access authentication according to the Token information in the message to determine whether the user in the request is authorized to access. If not authorized, an error code and corresponding prompt information are returned to the user. If authorized, the data request is forwarded to the data proxy server on the center side. The data proxy server determines whether the request IP is a center side IP address. If not, it determines which power plant the request comes from. If it is a center side IP address, go to X. If the request IP comes from a power plant that applies a time series data processing system, forward the request to the data proxy server of the power plant and access its local time series data interface. If the request IP does not belong to any power plant that applies a time series data processing system, go to X. The power plant local time series data interface service determines whether the accessed measurement point name is correct and the data range is reasonable. If the measurement point name is incorrect or the data range is unreasonable, an error code and corresponding prompt information are returned to the user. If the above information is correct, query the data from the power plant side time series database system and return it to the user. Access the center side time series data interface. The center side time series data interface service determines whether the accessed measurement point name is correct and the data range is reasonable. If the measurement point name is incorrect or the data range is unreasonable, an error code and corresponding prompt information are returned to the user. If the above information is correct, query the data from the center side time series database system and return it to the user.

[0022] A cloud-edge integrated time series data processing method, comprising the following steps:

[0023] Step one: deploy a power plant side time series database in each power plant, collect and store the time series data of each power plant in the local time series database of the power plant;

[0024] Step two: deploy a center side time series database in the data center, upload and access the power plant side time series database to the center side time series database as needed;

[0025] Step three: deploy a time series data processing system to uniformly manage the power plant side time series data and the center side time series data.

[0026] The time series data processing system provides connection management, time series data storage, configuration management, data management, operation and maintenance management, data analysis, and data statistics functions. Users can manage and analyze the time series data of one or more power plants according to actual needs.

[0027] The connection management steps in the time series data processing system are as follows: adding a new connection: the user sets the database instance connection name, host or IP address, port, username, password information, and then tests the connection. If the connection is successful, the connection is confirmed to be correct, and the configuration information is saved. Otherwise, check whether the port between the time series data processing system and the database server to be connected is open to ensure that the link is in normal communication; continue to add new connections or delete invalid connections, configure the display information of all connections, support display of one or more of the following: display name, group number, host, port, measurement point number, username, database, creator, and creation time information, search for a specified connection by connection name, view connection information, switch connections according to the data to be accessed, and switch the current connection to a specified database instance.

[0028] The configuration management steps in the time series data processing system are as follows: system management: support displaying the nuclear power time series data model in a tree diagram; user management: new and delete users of the time series database instance, set the user's permissions, support selecting and setting the data range, support setting 5 permissions of creating a storage group, creating a time series, inserting data, querying data, and deleting a time series or data, and support configuring the resource permissions related to users, roles, UDFs, and triggers; role management: support adding, deleting, and binding roles with users; group management: create a time series data group, and manage one unit measurement point data or data of a certain business field in each group; data query: the user newly creates a query, inputs a SQL-like statement in the console to query time series data, supports selecting measurement points and selecting common functions, and supports aggregation functions, mathematical functions, string functions, selection functions, trend calculation functions, and time functions.

[0029] The data management steps in the time series data processing system are as follows: general service: select a unit under a certain power plant, select a time range, a storage path, equipment, a measurement point, a sorting method, and a query method to query the measurement point data, and support arranging the queried data in order or reverse order according to time, and support historical query, data query, and aggregation query; equipment service: query measurement point attribute information according to equipment, and the queried measurement point attribute information includes power plant, unit, equipment name, equipment code, measurement point ID, measurement point description, engineering unit, measurement point type group, measurement point value type, measurement point level, sensor type, measurement point source type, maximum range, minimum range, measurement type, whether a calculation point, and import time; query measurement point attribute information according to the system, and the queried measurement point attribute information includes power plant, unit, system name, system code, measurement point ID, measurement point description, engineering unit, measurement point type group, measurement point value type, measurement point level, sensor type, measurement point source type, maximum range, minimum range, measurement type, whether a calculation point, and import time.

[0030] The operation and maintenance management steps in the time series data processing system are as follows: server monitoring: the monitoring content includes CPU usage, memory usage, memory overflow OOM, disk IO per second, disk data read / write situation, system time synchronization status; time series database instance monitoring: the monitoring content includes the number of groups, the amount of data written per second, the amount of data read per second, memory usage, the number of concurrent access per second, the time consumed for each query, the current number of idle connections of the database connection pool, high time-consuming query and writing, the current user accessing the system; log audit: all operation information is recorded, and the operation information is supported to be searched according to the access IP, operation type, operator, unit, operation time.

[0031] The data analysis steps in the time series data processing system are as follows: basic data entry: the basic data related to nuclear power is entered, edited and displayed, including nuclear power unit data and safety index data; mapping group management: the nuclear power time series data index is classified and managed, the index classification is created, edited and deleted, and the corresponding index is added in each index classification; mapping configuration management: the corresponding measuring point is added in each index, the measuring point meeting the requirements is searched in the system, the corresponding index is selected for the measuring point, the mapping configuration is completed, and the user views the information corresponding to each index; monitoring data management: the nuclear power monitoring data is configured through the mapping configuration management function, including increasing the power system monitoring group, extracting and monitoring the electric power, thermal power and generator voltage index of the power system of each power plant and each unit, and extracting and monitoring the coolant average temperature, coolant flow and coolant pressure index of each power plant and each unit through adding the cooling system monitoring group; historical trend analysis: the power plant and unit are selected, the required measuring point is queried and selected, the time range is selected, the historical trend curve graph of the selected measuring point is drawn, the selected measuring point is batch-labeled, the historical trend curves are compared according to the label information, and the historical trend analysis supports the comprehensive trend comparison of all measuring point data under different power plants and different units.

[0032] The data analysis steps in the time series data processing system are as follows: system data statistics: the measuring point data under the nuclear power process system are statistically analyzed, including the total system quantity, the total measuring point quantity and the measuring point quantity under each system; equipment data statistics: the nuclear power equipment measuring point data are statistically analyzed, including the total equipment quantity, the total measuring point quantity and the measuring point quantity under each equipment; system equipment statistics: the association information of the system and the equipment is statistically analyzed, including the total system quantity, the total equipment quantity, the total measuring point quantity, the selected current system name, the equipment quantity under the current system and the information list of all equipment under the current system; statistical report export: after the system data statistics, the equipment data statistics or the power production statistical information of the power plant unit are configured, the information is exported as a report.

[0033] The beneficial effects achieved by the present application are:

[0034] The present application can effectively manage the time sequence database instances and time sequence data of multiple power plant sides and center sides, while meeting the requirements of security and reliability. Through the time sequence data processing system of the present application, when a large amount of data from multiple power plants is transmitted simultaneously, parallel transmission and consumption are performed through multiple channels, avoiding the occurrence of data loss or serious data delay caused by single data transmission link and consumption program; at the same time, the unified time sequence data service provided by the present application can be registered and accessed in the industrial service open platform, and can perform concurrent access control and data access range check, to ensure the security and stability of data access; in addition, the unified time sequence data service can obtain data from the time sequence database system closest to the user according to the actual request of the user, to reduce the data acquisition time and cost; the present application classifies and manages the time sequence database instances of each power plant and the data from different power plants and different units according to business attributes through connection management, configuration management, data management, operation and maintenance management, data analysis and other functions, to improve the management efficiency and the data retrieval efficiency; the present application can be compatible with multiple single machine version, dual active version and distributed version time sequence databases, and the data can be mounted to multiple SSD solid state disks, and the hardware storage support can be planned and designed in a RAID mode, to further improve the storage and access efficiency of time sequence data and improve the security of data. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 It is a cloud edge integrated time sequence data processing system workflow diagram;

[0036] Figure 2 It is a cloud edge integrated time sequence data processing system function architecture diagram;

[0037] Figure 3 It is a cloud edge integrated time sequence data processing system data access flowchart. DETAILED DESCRIPTION

[0038] The present application will be described in detail below in combination with the drawings and specific embodiments.

[0039] For example, Figure 1—3 shows a cloud-edge integrated time series data processing system: a power plant side (edge end) time series database is deployed in each power plant, and the time series data of each power plant is collected and stored in the local time series database of the power plant; a center side (cloud end) time series database is deployed in the data center, and the power plant side time series database is uploaded and accessed in the center side time series database as needed; a cloud-edge integrated time series data processing system is deployed to uniformly manage the power plant side time series data and the center side time series data; the cloud-edge integrated time series data processing system provides time series data physical storage, connection management, cloud-edge integrated time series data storage, configuration management, data management, operation and maintenance management, data analysis, data statistics and other functions, and users can manage and analyze the time series data of a certain power plant or multiple power plants according to actual needs.

[0040] The power plant side time series database not only supports the access and storage of power plant DCS systems, PI systems and sensor data, but also supports the storage of inspection data, chemical data and calculation point data. The specific steps are as follows: (a) parsing the received data and placing different types of data in different storage areas; (b) for structured data, converting it into a time series data storage format before storing; (c) developing a power plant side time series data access service to support applications to obtain corresponding data according to actual needs; step (c) includes multiple forms of data services, each of which can meet specific data needs. The specific steps are as follows: registering a time series data interface in the industrial service open platform as needed; opening the time series data service through the industrial service open platform to support time series data access authentication, concurrent access control and data access range control. The center side time series database collects and accesses the time series data of each power plant as needed, and supports managing the full measurement point metadata of all power plants and supports data permission granularity control at four levels of global, database, entity and measurement point.

[0041] According to the data size of each nuclear power plant, one or more channels are used for data uploading and access in each power plant. Each power plant data can be accessed as needed, and data can be accessed according to power plant, unit, measurement point, etc. According to the needs of different users, the authorized data is controlled in terms of granularity, and the global data permission can be granted to the administrator user, the database permission can be granted to the power plant user, the entity device permission can be granted to the application, and the measurement point permission can be granted to the tester, etc.

[0042] The cloud-edge integrated time series data processing system can be deployed in the server of the center side or any power plant side, and only one set needs to be deployed to manage and access the time series database instances of the center side and each power plant side. The storage and access of all cloud and edge data can be logically regarded as accessing a set of cloud-edge integrated time series data processing system. The cloud-edge integrated time series data processing system supports vertical resource expansion and horizontal distributed expansion, and can support the management and analysis of tens of millions of measuring points and more than 1 trillion time series data.

[0043] The time series data in the power plant side time series database and the center side time series database can be used as each other's copy. When data loss or other data inconsistency occurs in one party, the data of the other party can be used for completion and repair. In addition, the power plant side time series database and the center side time series database can be synchronized at the measuring point level and the storage file level through synchronization settings. The cloud-edge integrated time series data processing system can be compatible with single machine version, dual-active version and distributed version time series database. The stored data can be mounted to multiple SSD solid state disks, and the underlying hardware storage supports RAID planning and design.

[0044] A specific application example is given below: deploy time series database on the center side and each power plant side of the DHP platform. The deployed time series database can be any form of single machine version, dual-active version and distributed version. Deploy the collection program in each power plant to collect data according to the business needs of the power plant side. In this embodiment, the data of each power plant includes PI system data, DCS system data and sensor data, in addition to inspection data and chemical data collected from the business system. These data need to be collected and stored in a standardized data format. Determine whether it is PI system data, DCS system data or sensor data. If so, process it into the format required by the standardized time series data storage and store it in the power plant side time series database. If not, it means that the data is from the business system, such as inspection data and chemical data. These structured data from the business system need to be converted into time series data storage format and then stored in the power plant side time series database. Taking inspection data as an example, the data conversion process is as follows: the equipment readings in the inspection record are updated once a day in the time series data processing system, which is low-frequency time series data. The following fields of data are saved in the form of [point_name, timestamp, value] in the time series data processing system:

[0045]

[0046] Inspection time series record:

[0047]

[0048] The time series data processing system includes the following steps: adding a new connection; setting database instance connection name, host or IP address, port, username, password, and other information, and then testing the connection; if the connection is successful, confirming that the connection is correct, saving the configuration information, otherwise, checking whether the port between the system and the database server to be connected is open to ensure that the link communication is normal; continuing to add a new connection or deleting an invalid connection; configuring display information of all connections, supporting one or more of the following information: display name, group number, host, port, number of measuring points, username, database, creator, and creation time; searching for a specified connection through a connection name to view connection information; and switching the current connection to a specified database instance according to the data to be accessed.

[0049]

[0050] A plurality of time series data service interfaces are deployed to provide applications, which can be deployed in a manner that one interface corresponds to one microservice, or multiple interfaces are placed in one microservice for deployment, and mainly include the following data service interfaces: obtaining attribute information of a specified single or multiple measuring points; obtaining index information of a single or multiple high-frequency measuring points in a time period; obtaining the latest actual value of a recent timestamp of a specified single or multiple measuring points; obtaining interpolation of a specified single or multiple measuring points at a specified time point; obtaining all actual values of a specified single or multiple measuring points in a specified time period; obtaining interpolation of a specified single or multiple measuring points in a specified time period and a specified time interval; obtaining an aggregation query result of a specified single or multiple measuring points in a specified time period; writing point value to a time series database; batch writing point value to a time series database; obtaining an original value of high-frequency data at a time point; and measuring point registration.

[0051] According to the configuration, it is determined whether the power plant side time series data needs to be accessed to the center side time series database, if yes, the data is uploaded through the MQTT protocol. When using the MQTT protocol to upload data, it is necessary to determine the scale of data to be uploaded. If all measuring point data of a power plant need to be uploaded to the center side, multiple MQTT connections need to be configured. Generally, the data of each unit is uploaded through one MQTT connection. The time series data uploaded to the center side is consumed to the center side time series database. The cloud-edge integrated time series data processing system manages the center side and power plant side time series database instances and all time series data.

[0052] The connection management steps in the cloud-edge integrated time series data processing system are as follows: adding a new connection. The user sets the database instance connection name, host or IP address, port, username, password, and other information, and then tests the connection. If the connection is successful, the connection is confirmed to be correct, the configuration information is saved, otherwise, it is checked whether the port between the system and the database server to be connected is open to ensure that the link communication is normal. Continue to add a new connection or delete an invalid connection. Configure the display information of all connections, support one or more of the following information: display name, group number, host, port, number of measuring points, username, database, creator, and creation time. Search for a specified connection through a connection name to view connection information. According to the data to be accessed, the connection is switched to a specified database instance.

[0053] The configuration management work steps in the cloud-edge integrated time series data processing system are as follows: system management: tree diagram display of nuclear power time series data model is supported; user management: new creation and deletion of users of the time series database instance are performed, and the rights of the users are set, and four levels of data right granularity settings of global, database, entity and measuring point are supported. Selection and setting of data range are supported, and five right settings of creating a storage group, creating a time series, inserting data, querying data and deleting a time series or data are supported, and the data granularity, data range and selection right together determine the data management right of the user. In terms of resource management right, the resource right configuration related to the user, the role, the UDF and the trigger is supported. Role management: the addition and deletion of system roles and the binding between the system roles and the users are supported. Group management: a time series data group is created, and one unit measuring point data or data of a certain business field is managed in each group. The group name is named in the form of power plant code_unit code, such as No. 1 group managing the data of No. 1 unit and No. 20 group managing the inspection data. The group information can also be edited and deleted. Data query: the user newly creates a query, and a SQL-like statement is input in the console to query the time series data, the selection of the measuring point and the selection of the commonly used function are supported, and the commonly used function supports the aggregation function, the mathematical function, the string function, the selection function, the trend calculation function and the time function. The query operation can be paused in the query process, and the query statement can also be saved for later use.

[0054] The data management work steps in the cloud-edge integrated time series data processing system are as follows:

[0055] General service: a certain unit under a certain power plant is selected, a time range, a storage path, an equipment, a measuring point, a sorting mode and a query mode are selected for query, and the measuring point data can be queried. The queried data can be exported in the CSV format. The queried data supports sequential and reverse sequential arrangement according to time, and the query mode supports historical query, data query and aggregation query.

[0056] Equipment service: the measuring point attribute information is queried according to the equipment. The equipment code is input, and the corresponding equipment belonging measuring point attribute information is queried. The queried measuring point attribute information includes power plant, unit, equipment name, equipment code, measuring point ID, measuring point description, engineering unit, measuring point type group, measuring point value type, measuring point level, sensor type, measuring point source type, maximum range, minimum range, measurement type, whether to calculate point and import time; the measuring point attribute is queried according to the system. The system code is input, and the corresponding system belonging measuring point attribute information is queried. The queried measuring point attribute information includes power plant, unit, system name, system code, measuring point ID, measuring point description, engineering unit, measuring point type group, measuring point value type, measuring point level, sensor type, measuring point source type, maximum range, minimum range, measurement type, whether to calculate point and import time.

[0057] Data interface registration and opening: fill in the gateway address, service address, request method, interface name and other information of the data API, and complete the registration of the measurement point. The registered data interface can be opened to the outside through the time series data processing system. First, access authentication configuration is performed in the system, then online verification is performed using the JWT+Token method, and if the verification is passed, the system data can be accessed. In addition, it can also be opened to the outside through the industrial service opening platform. First, add an interface in the platform, configure the service name, interface address, interface description, request type, maximum number of connections supported, maximum number of requests supported within a specified time period and other information. Then the reviewer audits it, and after the audit is passed, the platform will generate a Token bound to the user and the interface, and specify the effective start time and end time of the Token. Finally, when the user uses the interface, add the corresponding Token in the request header, and the system data can be accessed. In order to ensure data security and system security, the range of measurement points and the time range that the user is allowed to access can also be controlled in more detail.

[0058] Data access: users access the required data on demand through the unified time series data service. The cloud-edge integrated data access process is as follows: the power plant application accesses the data in the cloud-edge integrated time series data processing system through the unified time series data service. The data request is forwarded to the data gateway. The data gateway checks whether the current data request in the system exceeds the allowed concurrent access quantity QPS according to the configuration parameters, and returns an error code and the corresponding prompt information to the user if it exceeds. If it is within the system allowed range, the data gateway parses the request message and performs access authentication according to the Token information in the message to determine whether the user in the request is authorized to access. If not authorized to access, an error code and the corresponding prompt information are returned to the user. If authorized to access, the data request is forwarded to the data proxy server on the center side (cloud). The data proxy server program determines whether the request IP is a center side IP address. If it is not a center side IP address, it is determined that the request is specifically from which power plant, and if it is a center side IP address, it is transferred to x. If the request IP is from a power plant that applies to a time series data processing system, the request is forwarded to the data proxy server of the power plant and the local (edge side) time series data interface is accessed. If the request IP does not belong to any power plant that applies to a time series data processing system, it is transferred to x. The power plant local time series data interface service determines whether the accessed measurement point name is correct and the data range is reasonable. If the measurement point name is incorrect or the data range is unreasonable, an error code and the corresponding prompt information are returned to the user. If the above information is not a problem, the data is queried from the power plant side time series database system and returned to the user. The time series data interface on the center side (cloud) is accessed. The center side time series data interface service determines whether the accessed measurement point name is correct and the data range is reasonable. If the measurement point name is incorrect or the data range is unreasonable, an error code and the corresponding prompt information are returned to the user. If the above information is not a problem, the data is queried from the center side time series database system and returned to the user.

[0059] The operation and maintenance work steps in the cloud-edge integrated time series data processing system are as follows: server monitoring: the time series data processing system server is monitored, and the monitoring content includes CPU usage, memory usage, memory overflow OOM, disk IO per second, disk data read / write situation, system time synchronization status, etc. Time series database instance monitoring: all database instances are monitored, and the monitoring content includes the number of groups, the amount of data written per second, the amount of data read per second, memory usage, interface concurrent access quantity QPS per second, query time consumption each time, the current idle connection number of the database connection pool, high time-consuming query and writing, the current access system user, etc. Log auditing: all operation information in the system is recorded, and the operation information is supported to be searched according to the access IP, operation type, operator, indicated unit, operation time.

[0060] The data analysis work steps in the cloud-edge integrated time series data processing system are as follows: basic data entry: the basic data related to nuclear power are entered, edited and displayed. For example, nuclear power unit data, safety index data, etc. Mapping grouping management: the nuclear power time series data index is classified and managed. First, the index classification is created, edited and deleted, and then the corresponding index is added in each index classification. Mapping configuration management: the corresponding measuring point is added in each index above. First, the measuring point meeting the requirements is found in the system, then the corresponding index is selected for the measuring point, the mapping configuration is completed, and finally the user can view the information corresponding to each index. Monitoring data management: through the mapping configuration management function, the nuclear power monitoring data is configured. For example, the power system monitoring group is added, and the power, thermal power, generator voltage and other indexes of the power system of each power plant and each unit are extracted and monitored. The cooling system monitoring group is added, and the coolant average temperature, coolant flow, coolant pressure and other indexes of each power plant and each unit are extracted and monitored. Historical trend analysis: first, the power plant and unit are selected, the required measuring point is queried and selected, then the time range is selected, and the system can draw the historical trend curve graph of the selected measuring point. The selected measuring points can be batch-labeled, and the historical trend curve can be compared according to the label information in the future. The historical trend analysis can support the comprehensive trend comparison of all measuring point data of different power plants and different units.

[0061] The data statistics work steps in the cloud-edge integrated time series data processing system are as follows: system data statistics: the measuring point data under the nuclear power process system is statistically analyzed. Including the total system quantity, the total measuring point quantity, and the measuring point quantity under each system. Equipment data statistics: the nuclear power equipment measuring point data is statistically analyzed. Including the total equipment quantity, the total measuring point quantity, and the measuring point quantity under each equipment. System equipment statistics: the association information of system and equipment is statistically analyzed. Including the total system quantity, the total equipment quantity, the total measuring point quantity, the selected current system name, the equipment quantity under the current system, and the information list of all equipment under the current system. Statistical report export: after the system data statistics, equipment data statistics or power production statistics information of the power plant unit are configured, these information are exported as a report.

Claims

1. A cloud-edge integrated time-series data processing system, characterized in that: Deploy power plant-side time-series databases in each power plant to collect and store time-series data locally; deploy central-side time-series databases in the data center to upload and connect power plant-side time-series databases as needed; deploy a time-series data processing system to manage power plant-side and central-side time-series data in a unified manner.

2. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The time-series data processing system is deployed on a server at the central side or any power plant side to manage and access time-series database instances at the central side and each power plant side.

3. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The time-series data processing system is compatible with standalone, dual-active, and distributed versions of time-series databases. The stored data is mounted on multiple SSD solid-state drives, and the underlying hardware storage supports RAID planning and design.

4. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The power plant-side time-series database supports the storage of power plant DCS system, PI system, sensor data, inspection data, chemical data, and calculation point data.

5. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The power plant-side time-series database parses the received data and places different types of data in different storage areas; for structured data, it converts it into a time-series data storage format before storing it.

6. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The power plant-side time-series database registers time-series data interfaces on demand in the industrial service open platform; the time-series data service is opened through the industrial service open platform, supporting access authentication, concurrent access control, and data access scope control for time-series data.

7. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The time-series database on the power plant side uses one or more channels for data upload and access, depending on the data volume of each nuclear power plant.

8. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The power plant-side time-series database accesses data according to the power plant, generating unit, and measuring point.

9. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The power plant-side time-series database implements granular permission control for authorized data based on the needs of different users. Administrators are granted global data permissions, power plant users are granted database permissions, applications are granted entity device permissions, and testers are granted test point permissions.

10. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The central time-series database collects and accesses time-series data from various power plants on demand, supports the management of full measurement point metadata for all power plants, and supports data permission granularity control at four levels: global, database, entity, and measurement point.

11. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: The time-series data in the power plant-side time-series database and the central-side time-series database serve as copies of each other. If data is lost or other data is inconsistent in one of them, the data in the other is used to complete and repair it. Multiple power plant-side time-series databases and the central-side time-series databases synchronize real-time and historical data at the measurement point level and storage file level through synchronization settings.

12. The cloud-edge integrated time-series data processing system according to claim 6, characterized in that: Time-series data interfaces are deployed in a manner where one interface corresponds to one microservice, or multiple interfaces are deployed in a microservice.

13. The cloud-edge integrated time-series data processing system according to claim 6, characterized in that: The time-series data interface includes: retrieving attribute information of one or more specified measurement points; retrieving index information of one or more high-frequency measurement points within a time period; retrieving the latest actual value of one or more specified measurement points with the most recent timestamp; retrieving interpolation values ​​of one or more specified measurement points at a specified time point; retrieving all actual values ​​of one or more specified measurement points within a specified time period; retrieving interpolation values ​​of one or more specified measurement points within a specified time period or time interval; retrieving aggregated query results of one or more specified measurement points within a specified time period; calculating point values ​​and writing them to the time-series database; calculating point values ​​and writing them to the time-series database in batches; retrieving the original values ​​of high-frequency data at a given time point; and registering measurement points.

14. The cloud-edge integrated time-series data processing system according to claim 1, characterized in that: Determine whether the power plant's time-series data needs to be connected to the central time-series database. If so, the data is uploaded via the MQTT protocol.

15. The cloud-edge integrated time-series data processing system according to claim 14, characterized in that: When using the MQTT protocol for data upload, determine the scale of the data to be uploaded. If it is necessary to upload all the data from all the measuring points of a power plant to the central side, multiple MQTT connections need to be configured, and the data of each unit is uploaded through one MQTT connection.

16. The cloud-edge integrated time-series data processing system according to claim 6, characterized in that: Registering a time-series data interface: Fill in the gateway address, service address, request method, and interface name information of the data API to complete the measurement point registration. After registration, the data interface will be exposed to the outside world through the time-series data processing system. Access authentication will be configured, and online verification will be performed using JWT+Token. If the verification passes, the data in the system can be accessed. Alternatively, the interface can be exposed to the outside world through the industrial service open platform. Add an interface to the platform, configuring the service name, interface address, interface description, request type, maximum number of supported connections, and maximum number of supported requests within a specified time period. Reviewers will conduct the review. After approval, the platform will generate a token bound to the user and the interface, specifying the token's effective start and end times. When using the interface, the user adds the corresponding token to the request header to access system data.

17. The cloud-edge integrated time-series data processing system according to claim 6, characterized in that: Data Access: Power plant applications access data in the time-series data processing system through a unified time-series data service. Data requests are forwarded to the data gateway. The data gateway checks whether the number of data requests in the current system exceeds the allowed concurrent access per second according to configuration parameters. If it does, it returns an error code and corresponding prompts to the user. If it is within the allowed range, the data gateway parses the request message and performs access authentication based on the token information in the message to determine whether the user requesting the request is authorized. If the access is not authorized, it returns an error code and corresponding prompts to the user. If the access is authorized, the data request is forwarded to the central data proxy server. The data proxy server determines whether the requesting IP is a central IP address. If it is not a central IP address, it determines which power plant the request originates from. If it is a central IP address, it forwards the request to [the central data proxy server]. If the requesting IP originates from [the central data proxy server], it forwards the request to [the central data proxy server]. The power plant using the time-series data processing system forwards the request to its data proxy server and accesses its local time-series data interface. If the requesting IP does not belong to any power plant using the time-series data processing system, it is redirected to X. The power plant's local time-series data interface service determines whether the accessed measurement point name is correct and whether the data range is reasonable. If the measurement point name is incorrect or the data range is unreasonable, an error code and corresponding prompt message are returned to the user. If the above information is correct, data is queried from the power plant's time-series database system and returned to the user. The system then accesses the central-side time-series data interface. The central-side time-series data interface service determines whether the accessed measurement point name is correct and whether the data range is reasonable. If the measurement point name is incorrect or the data range is unreasonable, an error code and corresponding prompt message are returned to the user. If the above information is correct, data is queried from the central-side time-series database system and returned to the user.

18. A cloud-edge integrated time-series data processing method, characterized in that: Includes the following steps: Step 1: Deploy a power plant-side time series database at each power plant, collect time series data from each power plant and store it in the local time series database at the power plant. Step 2: Deploy the central-side time-series database in the data center, and upload and connect the power plant-side time-series database to the central-side time-series database as needed; Step 3: Deploy a time-series data processing system to manage the time-series data from both the power plant and the central data center in a unified manner.

19. The cloud-edge integrated time-series data processing method according to claim 18, characterized in that: The time-series data processing system provides connection management, time-series data storage, configuration management, data management, operation and maintenance management, data analysis, and data statistics functions. Users can manage and analyze the time-series data of one or more power plants according to their actual needs.

20. The cloud-edge integrated time-series data processing method according to claim 19, characterized in that: The connection management steps in the time series data processing system are as follows: Add a new connection: The user sets the database instance connection name, host or IP address, port, username, and password information, and then tests the connection. If the connection is successful, confirm that the connection is correct and save the configuration information; otherwise, check whether the port between the time series data processing system and the database server to be connected is open to ensure normal communication of the link. Continue adding new connections or deleting expired connections. Configure the display information for all connections, supporting one or more of the following: display name, number of groups, host, port, number of test points, username, database, creator, and creation time. Search for a specific connection by connection name, view connection information, and switch connections according to the data needed, switching the current connection to the specified database instance.

21. The cloud-edge integrated time-series data processing method according to claim 19, characterized in that: The configuration management steps in the time series data processing system are as follows: System Management: Supports displaying nuclear power time series data models in a tree diagram; User Management: Allows the creation and deletion of users for time series database instances, setting user permissions, supporting the selection and setting of data ranges, and supporting the setting of five types of permissions: creating storage groups, creating time series, inserting data, querying data, and deleting time series or data. Supports the configuration of resource permissions related to users, roles, UDFs, and triggers; Role Management: Supports the addition and deletion of roles and their binding with users; Group Management: Create time series data groups, each group managing data from a single unit's measurement points or data from a specific business area; Data Query: Users can create new queries and enter SQL-like statements in the console to query time series data. This supports the selection of measurement points and commonly used functions, including aggregate functions, mathematical functions, string functions, selection functions, trend calculation functions, and time functions.

22. The cloud-edge integrated time-series data processing method according to claim 19, characterized in that: The data management steps in the time-series data processing system are as follows: General Services: Select a specific unit under a power plant, and select the time range, storage path, equipment, measurement point, sorting method, and query method to query the measurement point data. The retrieved data supports sequential and reverse sorting by time. Query methods support historical query, data query, and aggregate query. Equipment Services: Query measurement point attribute information by equipment. The retrieved measurement point attribute information includes power plant, unit, equipment name, equipment code, measurement point ID, measurement point description, engineering unit, measurement point type grouping, measurement point value type, measurement point level, sensor category, measurement point source type, maximum range value, minimum range value, measurement type, whether to calculate the point, and import time. Query measurement point attributes by system. The retrieved measurement point attribute information includes power plant, unit, system name, system code, measurement point ID, measurement point description, engineering unit, measurement point type grouping, measurement point value type, measurement point level, sensor category, measurement point source type, maximum range value, minimum range value, measurement type, whether to calculate the point, and import time.

23. The cloud-edge integrated time-series data processing method according to claim 19, characterized in that: The operation and maintenance management steps in the time series data processing system are as follows: Server monitoring: Monitoring content includes CPU utilization, memory utilization, memory overflow (OOM), disk I / O per second, disk data read / write status, and system time synchronization status; Time series database instance monitoring: Monitoring content includes the number of groups, data written per second, data read per second, memory usage, concurrent access to interfaces per second, time consumed per query, the number of currently idle connections in the database connection pool, high-time-consuming queries and writes, and the users currently accessing the system; Log auditing: Recording all operation information and supporting retrieval of this operation information by access IP, operation type, operator, indicated unit, and operation time.

24. The cloud-edge integrated time-series data processing method according to claim 19, characterized in that: The data analysis steps in the time-series data processing system are as follows: Basic data entry: Input, edit, and display basic nuclear power-related data, including nuclear power unit data and safety indicator data; Mapping group management: Classify and manage nuclear power time-series data indicators, creating, editing, and deleting indicator categories, and adding corresponding indicators to each category; Mapping configuration management: Add corresponding measurement points to each indicator, find suitable measurement points in the system, select the appropriate indicators for each measurement point, complete the mapping configuration, and allow users to view the information corresponding to each indicator; Monitoring data management: Configure nuclear power monitoring data through the mapping configuration management function, including... This includes adding a power system monitoring group to extract and monitor the electrical power, thermal power, and generator voltage indicators of the power system for each power plant and each generating unit; adding a cooling system monitoring group to extract and monitor the average coolant temperature, coolant flow rate, and coolant pressure indicators for each power plant and each generating unit; historical trend analysis: select the power plant and generating unit, query and select the required measurement points, select the time range, and draw the historical trend curve of the selected measurement points. The selected measurement points are batch-tagged, and the historical trend curves are compared based on the tag information. The historical trend analysis supports comprehensive trend comparison of all measurement point data under different power plants and different generating units.

25. The cloud-edge integrated time-series data processing method according to claim 19, characterized in that: The data statistics steps in the time-series data processing system are as follows: System data statistics: Statistical analysis of measurement point data under the nuclear power process system, including the total number of systems, the total number of measurement points, and the number of measurement points under each system; Equipment data statistics: Statistical analysis of measurement point data of nuclear power equipment, including the total number of equipment, the total number of measurement points, and the number of measurement points under each equipment; System equipment statistics: Statistical analysis of the correlation information between systems and equipment, including the total number of systems, the total number of equipment, the total number of measurement points, the selected current system name, the number of equipment under the current system, and an information list of all equipment under the current system; Statistical report export: After configuring the system data statistics, equipment data statistics, or power production statistics information of the power plant units, export this information as a report.

Images