High-reliability elastic data storage management method compatible with multiple electricity markets

Abstract

The invention discloses a high-reliability data elastic storage management method compatible with multiple power markets, which comprises the following steps: carrying out generality extraction and difference analysis on an interaction scene, an interaction mechanism, resource characteristics and a data source, and establishing a unified data interaction model; constructing a network-operation two-stage collaborative interaction architecture based on a power grid system and an operator system; in combination with the interaction requirements of a power grid system and an operator system, mass multi-source heterogeneous data are accessed and collected and then processed, a mass data storage management system is formed, the functions of operator management and control, overall effect monitoring and the like are achieved, and development of vehicle network interaction aggregation management and control is completed. The invention belongs to the technical field of data management, and particularly provides a high-reliability elastic data storage management method compatible with multiple power markets, which is used for solving the problems of real-time performance, reliability and stability of mass multi-source heterogeneous data transmission and how to ensure the safety, stability and high efficiency of mass data storage.

Description

Technical Field

[0001] The present invention belongs to the technical field of data management, and specifically refers to a highly reliable data elastic storage management method compatible with multiple power markets. Background Art

[0002] With the accelerated development of low-carbon energy transformation, the popularization and application of new energy vehicles will become an important way for my country to reduce carbon emissions and absorb new energy. In order to meet the increasing demand for new energy vehicle travel, the construction of new energy charging facilities will usher in sustained long-term development in the future. Considering the growth in the number of electric vehicle users, the large-scale disordered charging of electric vehicles will further aggravate the uncertainty of the power grid. It is necessary to strengthen the efficient vehicle-grid interaction between new energy vehicles and power grid energy, that is, a new energy transmission mode in which electric vehicles can transmit electricity to the power grid through energy conversion facilities with bidirectional charging and discharging such as charging. When the power demand is high, it is allowed to transmit the energy stored in the battery to the power grid. When the power demand is low or the power generation of renewable energy is high, the dependence on fossil fuel power generation can be reduced by using electric vehicle charging, so as to achieve energy conservation and carbon reduction and balance the power load. In recent years, V2G has made continuous breakthroughs in technology, but because its commercial application and market operation model are not yet clear, it is difficult to stimulate large-scale user participation.

[0003] In terms of the evaluation of electric vehicle charging and discharging flexibility, the main research at home and abroad is on approximate evaluation methods that reduce computational complexity. For example, Harvard University in the United States uses internal approximate methods such as built-in rectangles and ellipses, and Tsinghua University has established an aggregated power and energy boundary model to characterize the adjustable range of charging and discharging. However, they have simplified the coupling constraints between time periods, resulting in evaluation errors. In terms of vehicle-grid interaction regulation, existing research mainly adopts distributed or hierarchical methods from the perspective of charging and discharging facility operators. For example, California Institute of Technology in the United States uses distributed iterative algorithms to optimize vehicle charging plans for electric vehicle clusters, Denmark has carried out a demonstration project on the interaction of electric vehicles and power grids to absorb wind power on Bornholm Island, Northeast Electric Power University uses a two-tier architecture of operators and charging stations to guide electric vehicle charging, and the Beijing-Tianjin-Tangshan power grid has realized the participation of charging operators in the peak-shaving auxiliary service market, but the regulation accuracy and real-time performance are not high. Existing charging and discharging flexibility evaluation methods are difficult to balance accuracy and computational complexity, and the evaluation accuracy is low. Existing research on vehicle-grid interaction and control is mostly targeted at operators, with less research on the control strategies of power grid companies. The scale of demonstration applications of vehicle-grid interaction is small, and there is insufficient research on the coordinated control of massive resources, power grid security constraint processing and global optimization brought about by large-scale vehicle-grid interaction. Practical issues such as the accuracy and real-time nature of vehicle-grid interaction control have not yet been deeply considered.

[0004] In response to the above situation, in February 2019, the State Grid Electric Vehicle Service Company developed "E-Vebko", which guides users to charge and discharge at different electricity price periods. When the electricity price is low, electric vehicles can be charged; when the electricity price increases, idle electric vehicles can be used to sell electricity to the grid through batteries. This method can effectively reduce users' charging costs. Although the above technology has studied the operation mode, management mode, control strategy and other aspects of electric vehicle V2G. However, the following problems still exist:

[0005] 1. How to access and process massive multi-source heterogeneous data based on streaming computing and incremental transmission to ensure data real-time, reliability and stability.

[0006] 2. How to build a massive data storage management system to ensure safe, stable and efficient data storage. Summary of the invention

[0007] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a high-reliability data elastic storage management method compatible with multiple power markets, which is used to solve the problems of real-time, reliability and stability of massive multi-source heterogeneous data transmission, and how to ensure the safe, stable and efficient storage of massive data.

[0008] The technical solutions adopted in this application are as follows:

[0009] This solution provides a highly reliable data elastic storage management method compatible with multiple power markets, including the following steps:

[0010] Step S1: extract commonalities and analyze differences among interactive scenarios, interactive mechanisms, resource characteristics, and data sources, and establish a unified data interaction model;

[0011] Step S2: Based on demand-side response, distribution network safety verification flexibility assessment and prediction, charging and discharging dynamic warning and functional requirements, a network-operation two-level collaborative interaction architecture based on the power grid system and the operator system is constructed;

[0012] Step S3: In combination with the interaction requirements of the power grid system and the operator system, research the access of massive multi-source heterogeneous data based on streaming computing and incremental transmission, access and collect massive multi-source heterogeneous data and process them to form a massive data storage management system, realize the operator's management and control and overall effectiveness monitoring functions, complete the development of vehicle-grid interactive aggregation management and control, and build a vehicle-grid interactive management and control support system, thereby achieving highly reliable data elastic storage and management compatible with multiple power markets.

[0013] Preferably, the unified data interaction model in step S1 includes a device layer, an edge layer, a network layer and a platform layer which are arranged in sequence from bottom to top.

[0014] Preferably, the network-operation two-level collaborative interaction architecture can be divided into application layer, platform layer, network layer, edge layer and resource layer, and the specific functions are as follows:

[0015] Resource layer: used to obtain the area load and charging pile status data of charging pile resources (community charging resources and charging and swapping station resources);

[0016] Among them, community charging resources include AC orderly piles, DC orderly piles, V2G piles and other types of piles; charging and swapping station resources include distributed photovoltaic equipment, V2G piles, DC piles and energy storage equipment;

[0017] Edge layer: used to collect data from the resource layer and transmit the data to the distribution system of the platform layer through the distribution transformer and edge control terminal;

[0018] Network layer: used to upload data obtained by the resource layer to the platform layer;

[0019] Platform layer: used to receive and collect various types of data from the edge layer, resource layer, etc., provide corresponding system underlying services according to user instructions, support data interaction with the National Monitoring and Management Platform for Electric Vehicles and other operators, and monitor and deal with the power supply status; among them, the platform layer is used to interact with the National Monitoring and Management Platform for Electric Vehicles for vehicle information, and the platform layer is used to interact with other operators for aggregated regulation; the platform layer includes a two-level interactive control support system for network operation, which interacts with the distribution system and the virtual power plant platform, performs charging pile monitoring and safety boundary services with the distribution system, and performs spot and auxiliary services and demand response services with the virtual power plant platform;

[0020] Application layer: used to receive request instructions and generate request instructions sent to the platform layer according to user instructions, including at least functions such as new energy consumption, green electricity trading, auxiliary services, demand response, local overload management, and network access.

[0021] Preferably, the network-operation two-level interactive control support system includes a grid-level interactive control support system located on the grid side, and the grid-level interactive control support system includes a performance monitoring module, a baseline calculation module, a market clearing module, a control strategy module, a substation early warning module, a trusted interaction module, etc.;

[0022] Among them, the grid-level interactive management and control support system supports system development. Combined with the interaction needs of the grid business system and the operator system, it studies the massive multi-source heterogeneous data access and processing technology based on streaming computing and incremental transmission, builds a massive data storage management system, and completes the development of the grid-level interactive management and control support system to realize functions such as operator management and overall performance monitoring.

[0023] It also includes an operator-level interactive control support system located on the operator side, which includes flexibility assessment and prediction module, AI model risk prevention and control module, aggregation decision module, access control module, online control module, trusted interaction module, charging facility dynamic early warning module, orderly charging and discharging strategy module, etc. Among them, the power grid side conducts two-way aggregation control with other operators, the operator side interacts with the national monitoring and management platform for electric vehicles for vehicle information; the resource layer conducts aggregation control status monitoring with the operator side.

[0024] Among them, the development of an operator-level interactive management and control support system is used to study technologies such as automatic grid-connected operation and debugging of charging and swapping facilities, dynamic stratification and grouping modeling, and real-time processing of time-series data, and to develop application modules for contactless and orderly charging and discharging and electric vehicle load aggregation for communities and fast charging stations. The development of an operator-level interactive management and control support system is completed to achieve aggregated access to massive charging and discharging resources, flexibility assessment and prediction, and online regulation.

[0025] In a further solution, the network-operation two-level interactive management and control support system can communicate downlink with the charging pile resources to carry out orderly charging and discharging control instructions / user behavior guidance schemes, and can also communicate uplink execution results;

[0026] The grid-level interactive management and control support system and the operator-level interactive management and control support system perform aggregate response and instruction decomposition interaction;

[0027] The grid-level interactive management and control support system is used to obtain the declaration information of the demand response platform and synchronize the load information to the demand response platform; and obtain the grid operation information of the new load management system, and publish the historical load data of the distribution area and the heavy overload warning information of the distribution area;

[0028] The operator-level interactive management and control support system sends a data request to the national electric vehicle monitoring and management platform to obtain the vehicle's real-time SOC information, vehicle real-time location and vehicle historical operation information.

[0029] Preferably, the vehicle-grid interactive control and management support system also includes a vehicle-grid interactive aggregation control and management support system, which includes at least a basic function module, a load organization module, a scheduling and monitoring module, and a clearing and settlement module. The basic function module includes station management, equipment management, and direct connection and control functions. The load organization module includes market demand, market declaration, and plan issuance functions. The scheduling and monitoring module includes an execution monitoring function. The clearing and settlement module includes a subsidy statistics function. The vehicle-grid interactive aggregation control and management support system and the network operation two-level interactive control and management support system have the functions of demand response and auxiliary services.

[0030] Preferably, the resource layer also includes an operator with its own cloud management platform, the network layer includes an interconnection protocol connected to the operator with its own cloud management platform, and a 4G / 5G public network connected to the charging pile resources, and the network layer also includes an APN wireless private network and a secure access area / firewall.

[0031] Preferably, the specific process of accessing and collecting massive multi-source heterogeneous data and processing it in step S3 is as follows:

[0032] 1) Data collection: Access equipment facilities, power grid business systems and operator systems through API interfaces or protocol conversion (middleware, ETL tools, model mapping management, etc.) to obtain massive multi-source heterogeneous data, including structured data such as equipment monitoring data, operation and maintenance data, and operational data, as well as semi-structured data such as XML and JSON documents, and unstructured data such as text, pictures, and videos generated during the monitoring process;

[0033] 2) Data integration and processing: Store and manage the collected massive multi-source heterogeneous data, perform pre-processing operations, and finally integrate the processed data into a target database;

[0034] 3) Data analysis: Further analysis of the integrated and processed data, such as data mining and statistical analysis, can be performed to optimize database performance based on business needs, such as creating indexes, adjusting query statements, or performing database partitioning.

[0035] Furthermore, the structured data specifically includes: data generated during the operation and maintenance of charging facilities, monitoring facilities, fire-fighting facilities, station auxiliary equipment and facilities in the power grid business system and operator system, as well as operational data generated by charging users during the charging process.

[0036] Furthermore, the data integration step specifically includes:

[0037] First, storage management includes: storing the collected massive multi-source heterogeneous data into different databases, where the databases use SQL databases, NoSQL databases and / or NewSQL databases to store structured data, semi-structured data and unstructured data; using database connection tools (such as JDBC, ODBC, etc.) to connect and associate different databases; among them, SQL databases represented by MySQL, Oracle, DB2, SQL Server, etc., NoSQL databases represented by Redis, HBase, MongoDB, Neo4j, etc., and NewSQL databases; this solution embodiment takes the use of MySQL database as an example;

[0038] Secondly, preprocessing includes: extracting required data from various databases according to business needs, which may involve full or incremental extraction, using ETL tools to clean and transform the data, cleaning the extracted data to remove duplicate, erroneous or invalid data; and performing data conversion, which includes: data type conversion, format conversion or encoding conversion to ensure data consistency and availability; using statistical methods such as PCA, LDA, ICA, KPCA, KICA, etc. to extract data features;

[0039] Finally, the preprocessed data is integrated into a target database. During the integration process, the consistency and integrity of the data are ensured to avoid data loss and damage, and the consistency of the data is maintained when synchronizing data between multiple databases.

[0040] By adopting the above scheme, the beneficial effects achieved by the present invention are as follows:

[0041] 1. Relieve the pressure on power grid: In this solution, the vehicle-grid interactive aggregation control system uses a large amount of stored energy from electric vehicles as a buffer between the power grid and renewable energy, which can alleviate the problem of low power grid operation efficiency and create certain benefits for electric vehicle users, achieving a win-win situation, which is also conducive to the popularization of electric vehicles. From the perspective of the State Grid, the system can play a role in regulating the power grid, that is, discharging to the power grid during peak hours and charging during non-peak hours, thereby reducing the load pressure on the power grid.

[0042] 2. Save charging costs: Electric vehicles participating in V2G can also provide hot standby capacity for the power grid, reduce investment in grid standby equipment, improve response speed, and reduce grid operation costs. They play a certain supporting role at the end of the grid, ensure the normal operation of the grid, and make the grid develop in a more intelligent direction.

[0043] 3. Provide efficient services: The data required for the collaborative interaction between vehicles, charging piles (stations) and the network has strong heterogeneity in terms of interaction scenarios, interaction mechanisms, and resource characteristics, and the data sources are wide-ranging. This plan intends to achieve standardization, unification, and normalization of data access and interactive sharing by building a cross-platform unified data interaction model, and utilize highly available, scalable streaming computing, distributed storage, metadata management and other technologies to achieve multi-source heterogeneous data separation, cleaning, processing, and storage, providing a data foundation and efficient services for the collaborative interaction of vehicles, charging piles (stations) and the network in multiple scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The overall architecture diagram of the vehicle-network interactive control support system provided for this solution;

[0045] Figure 2 This is a schematic diagram of the composition of the network-operation two-level collaborative interaction architecture in this solution;

[0046] Figure 3 This is a schematic diagram of the composition of the two-level interactive control support system for network and operation in this solution;

[0047] Figure 4 This is a schematic diagram of the architecture for implementing the interactive control support system functions in this solution;

[0048] Figure 5 This is a schematic diagram of the massive data access storage and processing architecture in this solution;

[0049] Figure 6 A schematic diagram of the demand response process provided in this embodiment;

[0050] Figure 7 for Figure 6 An enlarged diagram of the business development section;

[0051] Figure 8 for Figure 6 An enlarged schematic diagram of the medium load access part;

[0052] Fig. 9 for Figure 6 A magnified schematic diagram of the market operations section. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0054] Example 1

[0055] Please refer to Figure 1 As shown, this embodiment provides a highly reliable data elastic storage management method compatible with multiple power markets, including:

[0056] Step S1: extract commonalities and analyze differences among interactive scenarios, interactive mechanisms, resource characteristics, and data sources, and establish a unified data interaction model;

[0057] Step S2: Based on demand-side response, distribution network safety verification flexibility assessment and prediction, charging and discharging dynamic warning and functional requirements, a network-operation two-level collaborative interaction architecture based on the power grid system and the operator system is constructed;

[0058] Step S3: In combination with the interaction requirements of the power grid system and the operator system, research the access of massive multi-source heterogeneous data based on streaming computing and incremental transmission, access and collect massive multi-source heterogeneous data and process them to form a massive data storage management system, realize the operator's management and control and overall effectiveness monitoring functions, complete the development of vehicle-grid interactive aggregation management and control, and build a vehicle-grid interactive management and control support system, thereby achieving highly reliable data elastic storage and management compatible with multiple power markets.

[0059] like Figure 1 As shown, the unified data interaction model in step S1 includes a device layer, an edge layer, a network layer, and a platform layer arranged in sequence from bottom to top. The construction of the unified data interaction model in step S1 includes the following steps:

[0060] 1) First, we need to conduct in-depth analysis of different data sources or data systems. Data sources include but are not limited to databases, API interfaces, file systems, etc., and extract the commonalities between them. These commonalities may include data formats, data structures, data elements, etc. By finding these commonalities, we can determine a common data model that can accommodate the common elements in different data sources.

[0061] 2) While extracting commonalities, it is also necessary to analyze the differences between different data sources. These differences may include the specific definition, value range, precision, etc. of data elements. The purpose of difference analysis is to understand and deal with these differences to ensure that data from different data sources can be accurately represented and processed in a unified data interaction model;

[0062] 3) Select appropriate data interaction protocols and formats, such as RESTful API and MessagePack, to ensure cross-system compatibility and define a unified data model, including data fields, data types, and data constraints;

[0063] 4) After building a unified data interaction model, verify and optimize it to meet the needs of actual applications.

[0064] refer to Figure 2 As shown, the network-operation two-level collaborative interaction architecture can be divided into application layer, platform layer, network layer, edge layer and resource layer, and the specific functions are as follows:

[0065] Resource layer: used to obtain the area load and charging pile status data of charging pile resources (community charging resources and charging and swapping station resources);

[0066] Among them, community charging resources include AC orderly piles, DC orderly piles, V2G piles and other types of piles; charging and swapping station resources include distributed photovoltaic equipment, V2G piles, DC piles and energy storage equipment;

[0067] Edge layer: used to collect data from the resource layer and transmit the data to the distribution system of the platform layer through the distribution transformer and edge control terminal;

[0068] Network layer: used to upload data obtained by the resource layer to the platform layer;

[0069] Platform layer: used to receive and collect various types of data from the edge layer, resource layer, etc., provide corresponding system underlying services according to user instructions, support data interaction with the National Monitoring and Management Platform for Electric Vehicles and other operators, and monitor and deal with the power supply status; among them, the platform layer is used to interact with the National Monitoring and Management Platform for Electric Vehicles for vehicle information, and the platform layer is used to interact with other operators for aggregated regulation; the platform layer includes a two-level interactive control support system for network operation, which interacts with the distribution system and the virtual power plant platform, performs charging pile monitoring and safety boundary services with the distribution system, and performs spot and auxiliary services and demand response services with the virtual power plant platform;

[0070] Application layer: used to receive request instructions and generate request instructions sent to the platform layer according to user instructions, including at least functions such as new energy consumption, green electricity trading, auxiliary services, demand response, local overload management, and network access.

[0071] refer to Figure 1 As shown, the vehicle-network interaction control support system includes a two-level network-operation interaction control support system and a vehicle-network interaction aggregation control support system based on a two-level network-operation collaborative interaction architecture.

[0072] On the one hand, the network and operation two-level interactive management and control support system includes:

[0073] The grid-level interactive management and control support system located on the grid side includes a performance monitoring module, a baseline calculation module, a market clearing module, a control strategy module, an area warning module, a trusted interaction module, etc. The grid-level interactive management and control support system supports system development. In combination with the interaction needs of the grid business system and the operator system, it studies the massive multi-source heterogeneous data access and processing technology based on streaming computing and incremental transmission, builds a massive data storage management system, completes the development of the grid-level interactive management and control support system, and realizes functions such as operator management and overall performance monitoring.

[0074] It also includes an operator-level interactive management and control support system located on the operator side, which includes a flexibility assessment and prediction module, an AI model risk prevention and control module, an aggregation decision module, an access control module, an online control module, a trusted interaction module, a charging facility dynamic early warning module, an orderly charging and discharging strategy module, etc. Among them, the power grid side conducts two-way aggregation and control with other operators, and the operator side interacts with the national monitoring and management platform for electric vehicles for vehicle information; the resource layer and the operator side conduct aggregation and control status monitoring. Among them, the development of the operator-level interactive management and control support system is used to study technologies such as automatic grid-connected operation and debugging of charging and swapping facilities, dynamic stratification and grouping modeling, and real-time processing of time series data, develop non-sensing orderly charging and discharging for communities and fast charging stations, and electric vehicle load aggregation application modules, complete the development of the operator-level interactive management and control support system, and realize aggregated access, flexibility assessment and prediction, and online control of massive charging and discharging resources.

[0075] refer to Figure 3 As shown, the network-operation two-level interactive management and control support system can communicate downstream with the charging pile resources, carry out orderly charging and discharging control instructions / user behavior guidance plans, and can upstream execute results; the grid-level interactive management and control support system and the operator-level interactive management and control support system can perform aggregated response and instruction decomposition interaction.

[0076] The grid-level interactive management and control support system is used to obtain the declaration information of the demand response platform and synchronize the load information to the demand response platform; and obtain the grid operation information of the new load management system, and publish the historical load data of the substation and the heavy overload warning information of the substation for the distribution system; the operator-level interactive management and control support system sends a data request to the national monitoring and management platform for electric vehicles, so as to obtain the vehicle's real-time SOC information, the vehicle's real-time location and the vehicle's historical operation information.

[0077] On the other hand, reference Figure 4As shown, the vehicle-grid interactive control and management support system also includes a vehicle-grid interactive aggregation control and management support system, which includes at least a basic function module, a load organization module, a scheduling monitoring module and a clearing and settlement module. The basic function module includes station management, equipment management, and direct connection and direct control functions. The load organization module includes market demand, market declaration, and plan issuance functions. The scheduling monitoring module includes an execution monitoring function. The clearing and settlement module includes a subsidy statistics function. The vehicle-grid interactive aggregation control and management support system and the network-operation two-level interactive control and management support system perform demand response and auxiliary service functions. Among them, the resource layer also includes operators with their own cloud management platforms, the network layer includes an interconnection protocol connected to operators with their own cloud management platforms, and a 4G / 5G public network connected to charging pile resources. The network layer also includes an APN wireless private network and a secure access area / firewall.

[0078] refer to Figure 5 As shown, in step S3, during the operation of the system equipment, from the extensive data collection in the early stage, to the real-time calculation and analysis in motion based on streaming computing, and finally to the value extraction of data, the general process of multi-source heterogeneous data processing includes data collection, data integration and data analysis. The specific process of accessing and collecting massive multi-source heterogeneous data and processing it is as follows:

[0079] 1) Data collection: Access equipment facilities, power grid business systems and operator systems through API interfaces or protocol conversion (middleware, ETL tools, model mapping management, etc.) to obtain massive multi-source heterogeneous data, including structured data such as equipment monitoring data, operation and maintenance data, and operational data, as well as semi-structured data such as XML and JSON documents, and unstructured data such as text, pictures, and videos generated during the monitoring process;

[0080] The structured data specifically includes: data generated during the operation and maintenance of charging facilities, monitoring facilities, fire-fighting facilities, station auxiliary equipment and facilities in the power grid business system and operator system, as well as operational data generated by charging users during the charging process.

[0081] 2) Data integration and processing: Store and manage the collected massive multi-source heterogeneous data, perform pre-processing operations, and finally integrate the processed data into a target database;

[0082] 3) Data analysis: Further analysis of the integrated and processed data, such as data mining and statistical analysis, can be performed to optimize database performance based on business needs, such as creating indexes, adjusting query statements, or performing database partitioning.

[0083] In the above, data collection mainly realizes the accurate and real-time collection of a large amount of raw data, providing the original data source for the data integration stage. Data integration mainly realizes the database storage of data, data cleaning, conversion, dimensionality reduction and other pre-processing, and the construction of massive related databases, providing the pre-processed data source for the data analysis stage. Data analysis mainly uses technologies such as association analysis, classification clustering and deep learning to realize the value mining of data.

[0084] The data integration step specifically includes:

[0085] First, storage management includes: storing the collected massive multi-source heterogeneous data into different databases, where the databases are SQL databases, NoSQL databases and / or NewSQL databases for storing structured data, semi-structured data and unstructured data; using database connection tools (such as JDBC, ODBC, etc.) to connect and associate different databases;

[0086] Among them, SQL databases represented by MySQL, Oracle, DB2, SQL Server, etc., NoSQL databases represented by Redis, HBase, MongoDB, Neo4j, etc., and NewSQL databases; this solution embodiment uses the MySQL database as an example;

[0087] Secondly, preprocessing includes: extracting required data from various databases according to business needs, which may involve full or incremental extraction, using ETL tools to clean and transform the data, cleaning the extracted data to remove duplicate, erroneous or invalid data; and performing data conversion, which includes: data type conversion, format conversion or encoding conversion to ensure data consistency and availability; using statistical methods such as PCA, LDA, ICA, KPCA, KICA, etc. to extract data features;

[0088] In data integration, multi-source heterogeneous data integration is to integrate data from multiple data sources, shield the differences in data types and structures, solve the problems of complex sources and heterogeneous structures of multi-source heterogeneous data, so as to achieve unified storage, management and analysis of data, realize non-discriminatory access by users, and give full play to the value of data. At present, common database technologies include: SQL databases represented by MySQL, Oracle, DB2, SQL Server, etc., NoSQL databases represented by Redis, HBase, MongoDB, Neo4j, etc., and NewSQL databases. For cache data storage, the distributed write cache subsystem based on Redis is used to cache the collected metering data, and then write it in batches to the relational database, so that the system has the ability to cache data efficiently. HBase, which is implemented by Google Bigtable open source, has the characteristics of good scalability and perfect backup mechanism. When performing statistical analysis of multi-source heterogeneous data, HBase can be used to synchronize and integrate data from various subsystems for storage.

[0089] Finally, the preprocessed data is integrated into a target database. During the integration process, the consistency and integrity of the data are ensured to avoid data loss and damage, and the consistency of the data is maintained when synchronizing data between multiple databases.

[0090] Furthermore, the data analysis adopts association analysis, classification analysis, clustering analysis and / or streaming calculation method, wherein the association analysis method includes Apriori algorithm, FP-Growth algorithm, etc., the classification analysis method includes decision tree, artificial neural network, deep learning, support vector machine, etc., the clustering analysis method includes K-means algorithm, DBSCAN algorithm, BIRQH algorithm, etc., and the streaming calculation method includes real-time calculation and analysis based on flow data.

[0091] In the above system, combined with the interaction needs of power grid business systems, operators and other systems, research on massive multi-source heterogeneous data access and processing technology based on streaming computing and incremental transmission, build a massive data storage management system, complete the development of power grid-level interactive management and control support system, realize data management center, operator asset management, load organization, load response, clearing and settlement, effectiveness monitoring, and meet the needs of power grid, operator hierarchical regulation and full-link safety prevention and control.

[0092] The practical basis of the interactive control support system in this solution is:

[0093] (1) Bus charging scheduling based on peak-valley price difference

[0094] Time-of-use electricity prices reflect the most basic electricity load regulation needs of the power grid. For a large number of electric vehicle aggregates, the economy of charging can be achieved through the peak-valley price difference, thus producing research results and applications of electric vehicle charging strategies based on time-of-use electricity prices.

[0095] (2) Charging dispatch in substations

[0096] The primary goal of charging scheduling in the substation is to protect the transformer. By optimizing the scheduling, the transformer overload can be avoided after adding charging facilities, which effectively improves the transformer load rate and utilization rate.

[0097] (3) Aggregate electric vehicles to participate in peak load-shaving auxiliary services

[0098] The dispatch strategy based on peak-valley electricity price response is not sophisticated enough for power grid operation and is difficult to alleviate the increasing peak-shaving pressure of the power grid. The peak-shaving capacity of traditional peak-shaving units has reached its upper limit. As a large-scale load-side resource, electric vehicle aggregates have certain peak-shaving capabilities. At present, the power grid is also actively exploring load-side resources including electric vehicles to participate in power peak-shaving auxiliary services.

[0099] Aggregate electric vehicles to participate in peak-shaving auxiliary services. The aggregator first establishes an information exchange channel with the power grid dispatching center to exchange real-time data. The power grid exercises overall control over the aggregated charging resources by issuing dispatching instructions. The dispatching instructions are broken down and executed by the aggregator to meet the peak-shaving needs of the power grid.

[0100] (4) Partition aggregation and interactive control and demand response

[0101] In a province or city, when power supply pressure is high in some local areas, it is difficult to meet the peak load regulation and demand response of each area during the overall aggregation dispatch. In order to eliminate or alleviate the overload and heavy load problems of local electricity consumption, a solution has evolved to aggregate electric vehicles in different areas to accept grid dispatch. Aggregators regard each area as an aggregation unit, and the grid can control one or several aggregation units according to actual needs.

[0102] The above scheme has been implemented in Qingdao. As load-side resources, electric vehicles are aggregated by load aggregators according to power supply areas to participate in ubiquitous IoT interactive control and receive grid dispatch instructions. The control of the zoning is more refined and flexible. The division of specific areas can be determined according to the actual dispatch needs of the grid to meet the refined dispatch needs of the grid.

[0103] Example 2

[0104] refer to Figure 6-Figure 9As shown, this embodiment provides a demand response process diagram, which is applied to Tianjin City. In the horizontal direction, the participating users constitute the high-level operation platform, Tianjin Electric Power and access operators constitute the grid-level platform, and the Municipal Bureau of Industry and Information Technology and Tianjin Electric Power constitute the demand response platform. The vertical direction is composed of the business development part, the load access part and the market operation part.

[0105] in, Figure 7 As part of the business development component, Figure 8 is the composition of the load access part, Fig. 9 It is the component of the market operation part.

[0106] First, the grid-level platform collects market information and operates the high-level platform access application. The two parties conduct business negotiations together. If an intention is formed, a cooperation agreement is signed, and then it is checked whether the access conditions are met. If the access conditions are not met, the system needs to be improved; if the access conditions are met, the load access and access acceptance steps are carried out.

[0107] Secondly, the demand response platform publishes the implementation details of demand response. After obtaining the grid demand, the grid-level platform allocates capacity and negotiates the details with the high-level operation platform. After successful negotiation, the relevant agreement is signed, the entrusted registration is filled in, and the filled-in information is sent to the demand response platform for review.

[0108] Next, after the grid-level platform's control instructions are decomposed and issued, the advanced operation platform receives the instruction response and provides feedback on the grid-level platform's execution monitoring. After the grid-level platform executes the monitoring, it conducts a deviation assessment, and the advanced operation platform corrects the deviation. At the same time, the grid-level platform verifies the execution results to determine whether both parties have reached an agreement on the execution results. If an agreement is reached, the results will be reported to the demand response platform, which will then review the results. Then the three parties conduct clearing and settlement, and after the settlement, the grid-level platform issues a market evaluation report.

[0109] It should be noted that although the embodiments of the present invention have been shown and described, for ordinary technicians in this field, without departing from the purpose of the invention, structural methods and embodiments similar to the technical solution are designed without creativity and should all fall within the scope of protection of the present invention.

Claims

1. A highly reliable data elastic storage management method compatible with multiple power markets, characterized by: Step S1: extract commonalities and analyze differences among interactive scenarios, interactive mechanisms, resource characteristics, and data sources, and establish a unified data interaction model; Step S2: Based on demand-side response, distribution network safety verification flexibility assessment and prediction, charging and discharging dynamic warning and functional requirements, a network-operation two-level collaborative interaction architecture based on the power grid system and the operator system is constructed; Step S3: In combination with the interaction requirements of the power grid system and the operator system, a vehicle-grid interactive management and control support system is constructed, and the access of massive multi-source heterogeneous data based on streaming computing and incremental transmission is studied. After accessing and collecting massive multi-source heterogeneous data, it is processed to form a massive data storage management system, realize the operator's management and control and overall effectiveness monitoring functions, and complete the development of vehicle-grid interactive aggregation management and control, to achieve highly reliable data elastic storage and management compatible with multiple power markets.

2. A highly reliable data elastic storage management method compatible with multiple power markets according to claim 1, characterized in that: In step S2, the network-operation two-level collaborative interaction architecture can be divided into application layer, platform layer, network layer, edge layer and resource layer. Resource layer: used to obtain the area load and charging pile status data of charging pile resources; Edge layer: used to collect data from the resource layer and transmit the data to the distribution system of the platform layer through the distribution transformer and edge control terminal; Network layer: used to upload data obtained by the resource layer to the platform layer; Platform layer: used to receive and collect various types of data from the edge layer and resource layer, provide corresponding system underlying services according to user instructions, support data interaction with the National Monitoring and Management Platform for Electric Vehicles and other operators, and monitor and deal with the power supply status; use the platform layer to interact with the National Monitoring and Management Platform for Electric Vehicles for vehicle information, and use the platform layer to interact with other operators for aggregated regulation; the platform layer includes a two-level interactive management and control support system for network operation, which interacts with the distribution system and the virtual power plant platform, performs charging pile monitoring and safety boundary services with the distribution system, and performs spot and auxiliary services and demand response services with the virtual power plant platform; Application layer: used to receive request instructions and generate request instructions sent to the platform layer according to user instructions, including at least new energy consumption, green electricity trading, auxiliary services, demand response, local overload management, and network access functions.

3. A highly reliable data elastic storage management method compatible with multiple power markets according to claim 2, characterized in that: The interactive management and control support system includes: A grid-level interactive control support system located on the grid side, the grid-level interactive control support system includes a performance monitoring module, a baseline calculation module, a market clearing module, a control strategy module, an area early warning module and a trusted interaction module; the grid side conducts two-way aggregation control with other operators; the grid-level interactive control support system supports system development, combines the interaction needs of the grid business system and the operator system, studies the massive multi-source heterogeneous data access and processing technology based on streaming computing and incremental transmission, builds a massive data storage management system, completes the development of the grid-level interactive control support system, and realizes the control of operators and overall performance monitoring functions; The operator-level interactive management and control support system located on the operator side includes a flexibility assessment and prediction module, an AI model risk prevention and control module, an aggregation decision module, an access control module, an online control module, a trusted interaction module, a charging facility dynamic early warning module and an orderly charging and discharging strategy module; the operator side interacts with the national electric vehicle monitoring and management platform for vehicle information; the resource layer and the operator side perform aggregation and control status monitoring; the operator-level interactive management and control support system is developed to study the technology of automatic grid-connected operation and debugging of charging and swapping facilities, dynamic stratification and grouping modeling, and real-time processing of time series data, develop non-contact orderly charging and discharging for communities and fast charging stations, and electric vehicle load aggregation application modules, complete the development of the operator-level interactive management and control support system, and realize aggregated access, flexibility assessment and prediction, and online control of massive charging and discharging resources.

4. A highly reliable data elastic storage management method compatible with multiple power markets according to claim 3, characterized in that: The network-operation two-level interactive control support system can communicate with the charging pile resources downstream, carry out orderly charging and discharging control instructions / user behavior guidance schemes, and can upload execution results upstream; the grid-level interactive control support system and the operator-level interactive control support system can interact with each other in aggregated response and instruction decomposition; The grid-level interactive management and control support system is used to obtain the declaration information of the demand response platform and synchronize the load information to the demand response platform; and obtain the grid operation information of the new load management system, and publish the historical load data of the distribution area and the heavy overload warning information of the distribution area; The operator-level interactive management and control support system sends a data request to the national electric vehicle monitoring and management platform to obtain the vehicle's real-time SOC information, vehicle real-time location and vehicle historical operation information.

5. A highly reliable data elastic storage management method compatible with multiple power markets according to claim 4, characterized in that: The vehicle-grid interactive management and control support system also includes a vehicle-grid interactive aggregation management and control support system, which includes at least a basic function module, a load organization module, a scheduling monitoring module and a clearing and settlement module. The basic function module includes station management, equipment management, and direct connection and control functions. The load organization module includes market demand, market declaration, and plan issuance functions. The scheduling monitoring module includes an execution monitoring function. The clearing and settlement module includes a subsidy statistics function. The vehicle-grid interactive aggregation management and control support system and the network operation two-level interactive management and control support system have the functions of demand response and auxiliary services.

6. A highly reliable data elastic storage management method compatible with multiple power markets according to claim 5, characterized in that: The specific process of accessing and collecting massive multi-source heterogeneous data and processing them in step S3 is as follows: 1) Data collection: Access equipment facilities, power grid business systems and operator systems through API interfaces or protocol conversion to obtain massive multi-source heterogeneous data, including structured data such as equipment monitoring data, operation and maintenance data, semi-structured data such as XML and JSON documents, and unstructured data such as text, pictures and videos generated during the monitoring process; 2) Data integration and processing: Store and manage the collected massive multi-source heterogeneous data, perform pre-processing operations, and finally integrate the processed data into a target database; 3) Data analysis: Further analysis of the integrated and processed data, including data mining and statistical analysis, and optimization of database performance according to business needs, such as creating indexes, adjusting query statements, or partitioning the database.

7. A highly reliable data elastic storage management method compatible with multiple power markets according to claim 6, characterized in that: The data integration step specifically includes: First, storage management includes: storing the collected massive multi-source heterogeneous data into different databases, where the databases are SQL databases, NoSQL databases and / or NewSQL databases, for storing structured data, semi-structured data and unstructured data; using database connection tools to connect and associate different databases; the SQL databases include MySQL, Oracle, DB2 and / or SQL Server, and the NoSQL databases include Redis, HBase, MongoDB and / or Neo4j; Secondly, preprocessing includes: extracting required data from various databases according to business needs, involving full or incremental extraction, using ETL tools to clean and transform the data, cleaning the extracted data to remove duplicate, erroneous or invalid data; and performing data conversion, which includes: data type conversion, format conversion or encoding conversion to ensure data consistency and availability; using PCA, LDA, ICA, KPCA and / or KICA statistical methods to extract data features; Finally, the preprocessed data is integrated into a target database. During the integration process, the consistency and integrity of the data are ensured to avoid data loss and damage, and the consistency of the data is maintained when synchronizing data between multiple databases.

8. A highly reliable data elastic storage management method compatible with multiple power markets according to claim 7, characterized in that: The data analysis adopts association analysis, classification analysis, clustering analysis and / or streaming calculation method, wherein the association analysis method includes Apriori algorithm and FP-Growth algorithm, the classification analysis method includes decision tree, artificial neural network, deep learning and / or support vector machine, the clustering analysis method includes K-means algorithm, DBSCAN algorithm, BIRQH algorithm, and the streaming calculation method includes real-time calculation and analysis based on flow data.

9. The highly reliable data elastic storage management method compatible with multiple power markets according to claim 2 is characterized by: The charging pile resources include community charging resources and charging and swapping station resources; The community charging resources include AC orderly charging piles, DC orderly charging piles, V2G charging piles and other types of charging piles; the charging and swapping station resources include distributed photovoltaic equipment, V2G charging piles, DC charging piles and energy storage equipment; the resource layer also includes operators with their own cloud management platforms, and the network layer includes an interconnection protocol connected to operators with their own cloud management platforms, and a 4G / 5G public network connected to the charging pile resources. The network layer also includes an APN wireless private network and a secure access area / firewall.

10. The highly reliable data elastic storage management method compatible with multiple power markets according to claim 8, characterized in that: The structured data specifically includes: data generated during the operation and maintenance of charging facilities, monitoring facilities, fire-fighting facilities, and station auxiliary equipment and facilities in the power grid business system and operator system, as well as operational data generated by charging users during the charging process; the protocol conversion includes middleware, ETL tools and / or pattern mapping management.

Images