Electricity service provision system and data preparation method

Abstract

This reduces the cost of creating applications in a power service delivery system that can perform data preparation processing on data collected from multiple external equipment resources. [Solution] The power service provision system of the present invention comprises: an application execution unit that executes an application capable of providing service information related to transactions in the power market; a data preparation flow creation unit that obtains requirements for input data of the application from an external source, creates a data preparation processing flow if one has not already been created to create input data that satisfies the requirements, and does not create a data preparation processing flow if one has already been created; and an input data creation unit that executes the data preparation processing flow to create input data and inputs the created input data to the application execution unit.

Description

Technical Field

[0001] The present invention relates to a power service providing system and a data preparation method.

Background Art

[0002] Conventionally, there is a system (hereinafter referred to as "power service providing system") that provides power services in the cloud to tenants such as distributed power suppliers, power generation BG (Balancing Group) operators, aggregators, etc. who participate in multiple power markets. In the power service providing system, data collected from a wide variety and a huge number of external facility resources (for example, distributed power sources, consumers, etc.) connected via a network is utilized to provide services related to power market transactions in, for example, the supply-demand adjustment market, LFM (Local Flexibility Market), etc. for multi-tenants. Specifically, in the power service providing system, data collected from a wide variety and a huge number of facility resources is input into an application corresponding to the provided service, and by executing the application, service information related to power supply-demand and power market transactions is provided to the tenant. / / 你提供的原文中这段内容日语语法有些混乱,我按照大致意思进行了翻译,你可以检查下是否符合你的需求。

[0003] Also, conventionally, various methods for providing data collected from external facility resources to an application have been proposed (see, for example, Patent Document 1). In the method for providing heterogeneous system data in the distributed system disclosed in Patent Document 1, the following processing steps (1) to (3) are performed. (1) Divide into layers of "reception", "analysis", "conversion", and "transmission", and make the processing modules of each layer pluggable. (2) Determine the priority based on the type, importance, frequency, etc. of the data and the processing timing of the application using the corresponding data, and distribute the data. (3) Separate and accumulate / manage the data that cannot be converted due to a mismatch with the conversion definition, and also present a list of the corresponding data to the application. In Patent Document 1, by using this method, problems such as data stagnation and inability to meet the performance requirements from the application, data loss, etc. due to, for example, a large amount of data generation and high load of conversion processing are solved. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Patent No. 5809743 [Overview of the project] [Problems that the invention aims to solve]

[0005] Incidentally, in power service provision systems, power supply and demand data collected from multiple external equipment resources typically differs for each data set in terms of specifications (models), granularity, collection cycle, etc. Furthermore, data collection can result in issues such as data loss, delays, or partial arrival. Therefore, before application execution, it is necessary to process the collected data to meet the data requirements demanded by each market / tenant (hereinafter referred to as "data preparation processing").

[0006] Another possible data preparation method is to extract the necessary data from a common database containing data / protocol-converted collected data and then perform the data preparation process. However, this method requires creating a separate data preparation flow for each application, which increases the cost of developing the application.

[0007] This invention was made to solve the above problems. The object of this invention is to provide a technology that can reduce the cost of creating applications in a power service provision system that can perform data preparation processing on data collected from multiple external equipment resources. [Means for solving the problem]

[0008] To solve the above problems, the power service provision system of the present invention comprises an application execution unit, a data preparation processing flow creation unit, and an input data creation unit. The application execution unit executes an application capable of providing service information related to transactions in the power market. The data preparation processing flow creation unit obtains requirements for the application's input data from an external source. If a data preparation processing flow for creating input data that satisfies the requirements has not yet been created, the data preparation processing flow creation unit creates a data preparation processing flow; otherwise, it does not create a data preparation processing flow. The input data creation unit executes the data preparation processing flow to create input data and inputs the created input data to the application execution unit.

[0009] Furthermore, in order to solve the above problems, the data preparation method of the present invention is performed in the power service provision system of the present invention. The data preparation method of the present invention includes a data preparation processing flow creation unit obtaining requirements for application input data from an external source. The data preparation method of the present invention includes the data preparation processing flow creation unit creating a data preparation processing flow if a data preparation processing flow for creating input data that satisfies the requirements has not already been created. The data preparation method of the present invention also includes the data preparation processing flow creation unit not creating a data preparation processing flow if a data preparation processing flow for creating input data that satisfies the requirements has already been created. Furthermore, the data preparation method of the present invention includes an input data creation unit executing the data preparation processing flow to create input data. [Effects of the Invention]

[0010] According to the present invention with the above configuration, the cost of creating applications can be reduced in a power service provision system that can perform data preparation processing on data collected from multiple external equipment resources. [Brief explanation of the drawing]

[0011] [Figure 1]This is a schematic diagram of a power supply and demand-related system, including a power service provision system, according to one embodiment of the present invention. [Figure 2] This is a hardware configuration diagram of a computer device (server) applicable to each AP server, control plane, etc., included in a power service provision system according to one embodiment of the present invention. [Figure 3] This is a functional block diagram of a power service provision system according to one embodiment of the present invention. [Figure 4] This figure shows an example of the configuration of an application data requirements definition stored in the application data requirements definition storage unit of a power service provision system according to one embodiment of the present invention. [Figure 5] This figure shows an example configuration of an application dataset information table stored in the application dataset information storage unit of a power service provision system according to one embodiment of the present invention. [Figure 6] This figure shows an example configuration of a requirements-processing microservice correspondence table stored in the requirements-processing microservice correspondence table storage unit of a power service provision system according to one embodiment of the present invention. [Figure 7] This figure shows an example configuration of a common DB catalog / status information table stored in the common DB catalog / status information storage unit of a power service provision system according to one embodiment of the present invention. [Figure 8] This figure shows an example configuration of an application DB catalog / status information table stored in the application DB catalog / status information storage unit of a power service provision system according to one embodiment of the present invention. [Figure 9] This figure shows an example of the operation of a power service provision system according to one embodiment of the present invention. [Figure 10] This figure shows an example of the operation flow between components included in a power service provision system according to one embodiment of the present invention. [Figure 11] This flowchart shows the procedure for the pre-creation process of the data preparation flow performed in a power service provision system according to one embodiment of the present invention. [Figure 12]It is a flowchart showing the procedures from the runtime adjustment process of the data preparation flow to the application execution process in the power service providing system according to an embodiment of the present invention. [Embodiment for Carrying Out the Invention]

[0012] Hereinafter, a power service providing system and a data preparation method according to an embodiment of the present invention will be specifically described with reference to the drawings. Note that the power service providing system of this embodiment is a system capable of executing an application that provides various information related to transactions in the power market (for example, power supply and demand forecasting, power supply and demand planning, etc.) as service information.

[0013] [Configuration of Power Supply and Demand Related System] FIG. 1 is a configuration diagram of a power supply and demand related system 10 including a power service providing system 1 according to an embodiment of the present invention. Note that in FIG. 1, only the components related to the power service providing system 1 are shown for convenience of explanation.

[0014] As shown in FIG. 1, the power supply and demand related system 10 includes a power service providing system 1, a plurality of power market servers 2 that are communicatively connected to the power service providing system 1 via a network 9, and a plurality of tenant terminals 3. Further, the power supply and demand related system 10 includes a plurality of DERMS (Distributed Energy Resource Management System) servers 4, 5... that are communicatively connected to the power service providing system 1 via a network 9. Furthermore, the power supply and demand related system 10 includes distributed power sources (distributed power sources 6, 7, 8... in FIG. 1) that are communicatively connected to each DERMS server via a network.

[0015] In the power supply and demand related system 10 shown in FIG. 1, for example, the distributed power source 6 is connected to its controller 6a, and the controller 6a is connected to the DERMS server 4 via the network 4a. Further, for example, the distributed power sources 7 and 8 are respectively connected to the corresponding controllers 7a and 8a, and the controllers 7a and 8a are connected to the DERMS server 5 via the network 5a.

[0016] The power service providing system 1 acquires (collects) various measurement data (for example, active power, reactive power, solar power generation amount, battery discharge amount, etc.) related to the power supply of a plurality of distributed power sources 6, 7,... via a plurality of controllers 6a, 7a, 8a, a plurality of DERMS servers 4, 5,... and the network 9. Further, the power service providing system 1 receives information on data requirements (application data requirement definition described later) required for the input data of the application to be executed from each tenant terminal 3. Then, the power service providing system 1 creates a processing flow (hereinafter referred to as "data preparation flow") that enables execution of data preparation processing for creating input data that satisfies the data requirements. Further, when the application is executed, the power service providing system 1 executes the created data preparation flow to create various data (application data set described later) required for application execution, and inputs the various data into the application. Note that the internal configuration and operation of the power service providing system 1 will be described in detail later.

[0017] The plurality of power market servers 2 include, for example, operation servers for power markets such as a spot market, a demand response market, and LFM.

[0018] Each tenant terminal 3 is operated by a user, such as a tenant administrator. Each tenant terminal 3 transmits information regarding the application's data requirements (application data requirements definition described later) to the power service provision system 1 via the network 9, in response to the user's input operation. In addition, each tenant terminal 3 obtains various information such as power supply and demand forecast results and power supply and demand plans obtained from the power service provision system 1 through application execution via the network 9, and transmits this information to multiple power market servers 2 via the network 9.

[0019] Each of the multiple DERMS servers 4, 5… acquires various measurement data (e.g., active power, reactive power, solar power generation amount, battery discharge amount, etc.) related to the power supply of the corresponding distributed power source from a controller connected via the corresponding network. The multiple distributed power sources 6, 7, 8… include, for example, generators such as thermal and hydroelectric power plants, renewable energy (solar, wind, geothermal, etc.) power generation facilities, and batteries. In Figure 1, the distributed power sources are illustrated as external equipment resources connected to the power service provision system 1 via the network 9, but in this embodiment, for example, external equipment resources such as customer load equipment are also connected to the power service provision system 1 via the network 9 (see Figure 9 below).

[0020] [Hardware configuration of the power service provision system] As shown in Figure 1, the power service provision system 1 comprises multiple application servers (hereinafter referred to as "AP servers") 11-1, 11-2, 11-3, ..., 11-n. The power service provision system 1 also includes a firewall 12, a load balancer 13, a DB (Data Base) server 14, a control plane 15, and an internal network 16 connecting the various components within the power service provision system 1. The power service provision system 1 is configured as a cloud system that allows access from external information processing devices via the network 9, enabling virtual use of, for example, servers, networks, platforms, storage, software, etc.

[0021] Each of the multiple AP servers 11-1, 11-2, 11-3, ..., 11-n operates as a virtual machine (virtual server) and performs data preparation processing, application execution processing, etc., for creating application input data (see Figure 9 below). In other words, each AP server has the function of creating application input data and the function of executing the application. Each AP server is an example of the input data creation unit and application execution unit according to the present invention.

[0022] Firewall 12 is a security device that protects the power service provision system 1 from malicious attacks from external sources connected via network 9. Load balancer 13 is a device that distributes communication (traffic) from external sources to the power service provision system 1 to multiple AP servers 11-1, 11-2, 11-3, ..., 11-n, etc.

[0023] The DB server 14 centrally manages various data and programs usable by the power service provision system 1, and performs tasks such as searching, updating, saving, and backing up these. The internal configuration of the DB server 14 will be described in detail later.

[0024] The control plane 15 performs tasks such as controlling data transfer to multiple AP servers 11-1, 11-2, 11-3, ..., 11-n, and controlling the allocation of AP servers to execute applications. In this embodiment, the control plane 15 also receives application data requirement definitions (requirements for application input data) for the applications to be executed from each tenant terminal 3. The control plane 15 then performs pre-creation processing of data preparation flows (data preparation processing flows) that satisfy the received application data requirement definitions. Furthermore, the control plane 15 performs processing to adjust the execution timing of the data preparation flows. The functional configuration and operation of the control plane 15 will be described in detail later.

[0025] Here, we will describe an example of the hardware configuration of a computer device (server) that can be used as each AP server and control plane 15 included in the power service provision system 1. Figure 2 is a block diagram showing an example of the hardware configuration of a computer device 20 that can be used as each AP server and control plane 15.

[0026] As shown in Figure 2, the computer device 20 includes a processor 21, memory 22, local disk 23, I / O (Input / Output) devices 24, and bus lines 25 connecting them to each other.

[0027] The processor 21 consists of arithmetic processing units such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit). The processor 21 also reads program code for software that implements the various processing functions of the computer device 20 into the RAM (Random Access Memory) in the memory 22 and executes it. For example, the execution of applications in each AP server and the pre-creation of data preparation flows by the control plane 15 are executed by the processor 21 (CPU and MPU).

[0028] Memory 22 is called the main memory and consists of ROM (Read Only Memory) and RAM. Local disk 23 consists of storage devices such as HDD (Hard Disk Drive) and SSD (Solid State Drive). The ROM in memory 22 and local disk 23 store (remember) software program code and the like for realizing the various processing functions of the computer device 20.

[0029] The I / O device 24 consists of various interfaces used for inputting and outputting various types of data (various types of information) with external devices. The I / O device 24 also includes a communication function unit for data communication with the outside world via a communication network.

[0030] [Functional Configuration of the Electricity Service Provisioning System] Next, the functional configuration of the power service provision system 1 will be explained with reference to the diagram. Figure 3 is a functional block diagram of the power service provision system 1. Note that, in order to simplify the explanation, the firewall 12 is not shown in Figure 3.

[0031] (AP server) Each of the multiple AP servers 11-1, 11-2, 11-3, ..., 11-n is provided with multiple containers that functionally have the ability to execute a predetermined microservice, as shown in Figure 3. In the example shown in Figure 3, for example, AP server 11-1 is provided with container 31 capable of executing microservice 31a, container 32 capable of executing microservice 32a, etc. For example, AP server 11-2 is provided with container 33 capable of executing microservice 33a, container 34 capable of executing microservice 34a, etc. Also, for example, AP server 11-n is provided with container 35 capable of executing microservice 35a, container 36 capable of executing microservice 36a, etc.

[0032] When an application is executed, each AP server reads the necessary microservices from the microservice DB 43 (described later) in the DB server 14 or from an external source and executes them on the container. The distribution of the types of microservices executed by each AP server is appropriately controlled by the control plane 15 according to the processing load and other conditions of each AP server.

[0033] (DB server) As shown in Figure 3, the DB server 14 functionally comprises a common DB 41, an application DB 42, a microservice DB 43, and a preparation flow DB 44.

[0034] The common DB41 stores various measurement data related to power supply measured at each distributed power source connected to the power service provision system 1 (e.g., active power, reactive power, solar power generation amount, battery discharge amount, etc.). The common DB41 also stores various measurement data related to power demand measured at each customer's load equipment connected to the power service provision system 1 (e.g., power consumption amount, etc.).

[0035] The application DB42 stores the data sets input to the application (hereinafter referred to as "application data sets"), specifically the application data sets created by executing the data preparation process, for each application. The application data sets stored in the application DB42 consist of a set of data for each data item within the acquisition area and collection period defined in the application data requirements definition 201 (see Figure 4 below).

[0036] The microservices database 43 stores information such as identification details and programs for various microservices available during application execution. The microservices stored in the microservices database 43 may be existing microservices available in other systems, or they may be microservices dedicated to this system. Furthermore, the microservices stored in the microservices database 43 may include both existing and dedicated microservices.

[0037] Furthermore, microservices stored in the microservice DB43 include, for example, a microservice for extracting various data necessary for application execution from the common DB41 (hereinafter referred to as "data extraction MS"), a microservice for processing the extracted data into the required format (hereinafter referred to as "data processing MS"), and a microservice for formatting the data to be output to the application into a predetermined format (file format) (hereinafter referred to as "data structuring MS"). In this embodiment, as will be described later, a data preparation flow is created by appropriately combining the data extraction MS, data processing MS, and data structuring MS.

[0038] Furthermore, microservices stored in the microservice DB43 include, for example, a microservice for forecasting power supply and demand (hereinafter referred to as "forecast MS"), a microservice for formulating a power supply and demand plan based on the forecast results (hereinafter referred to as "planning MS"), and a microservice for placing bids in each power market (hereinafter referred to as "bid MS"). In this embodiment, the application execution flow is created by appropriately combining the forecast MS, planning MS, and bid MS.

[0039] The preparation flow DB44 stores, for example, identification information, programs, etc., of the data preparation flow that has been created in advance by the control plane 15.

[0040] (Control plane) As shown in Figure 3, the control plane 15 includes a data preparation function unit 100 that functionally pre-creates data preparation flows and adjusts the execution timing of data preparation flows.

[0041] The data preparation function unit 100 functionally comprises a data preparation processing control unit 101, an MS (microservices) service execution flow management unit 102, a resource management unit 103, a dataset management unit 104, and a DB / data collection status monitoring unit 105. These functional components are included, for example, in the processor 21 in Figure 2.

[0042] Furthermore, the data preparation function unit 100 functionally includes an application data requirements definition storage unit 111, an application dataset information storage unit 112, a requirements-processing microservice correspondence table storage unit 113, a common DB catalog / status information storage unit 114, an application DB catalog / status information storage unit 115, and a microservice information storage unit 116. These information storage units are included, for example, in memory 22 or local disk 23 in Figure 2.

[0043] The following describes the functions and operations of each of the above-mentioned functional components of the data preparation function unit 100, as well as the configuration of the data stored in each information storage unit.

[0044] (1)Each functional component As shown in Figure 3, the data preparation processing control unit 101 includes a pre-flow creation unit 101a and an execution-time adjustment unit 101b. The pre-flow creation unit 101a performs the data preparation flow creation process. The execution-time adjustment unit 101b performs adjustment control of the execution timing of the data preparation flow (including setting whether to execute or not execute the data preparation flow). The specific processing operations performed by the pre-flow creation unit 101a and the execution-time adjustment unit 101b will be described in detail later. Note that the pre-flow creation unit 101a is an example of a data preparation processing flow creation unit according to the present invention, and the execution-time adjustment unit 101b is an example of a data preparation processing flow execution setting unit according to the present invention.

[0045] The MS / Service Execution Flow Management Unit 102 manages, for example, the execution order of microservices (MS) when an application is executed on an AP server and when a data preparation flow is executed. The Resource Management Unit 103 manages the usage status (usage amount) of each equipment resource (for example, AP servers, control plane 15, etc.; hereinafter referred to as "cloud resources") that constitutes the power service provision system 1 (cloud system). The information on the specification status of cloud resources managed by the Resource Management Unit 103 is referenced, for example, in the allocation control of AP servers that execute applications and the determination control of AP servers to be used preferentially in data preparation processing (resource priority control in Figure 12 described later).

[0046] The dataset management unit 104 manages the data storage status in the application dataset information storage unit 112 and performs processing such as data updates to the application dataset information storage unit 112. The DB / data collection status monitoring unit 105 manages the data storage status in the common DB 41 and performs processing such as data updates to the common DB 41.

[0047] (2) Application data requirements definition storage unit The application data requirements definition storage unit 111 stores information on various requirements (hereinafter referred to as "application data requirements definitions") that are requested from each tenant terminal 3 for the input data of the application to be executed.

[0048] Figure 4 shows an example configuration of an application data requirements definition 201 stored in the application data requirements definition storage unit 111. In the application data requirements definition 201 (requirements), the following information is defined as a set of data requirements: "data item," "acquisition range (area)," "acquisition range (time)," "measurement granularity," "type," "completeness," "processing degree," "freshness," "collection degree," and "execution deadline." This application data requirements definition 201 is stored for each application.

[0049] The "Data Items" section specifies the types of measurement data (input data) used in the application. In the example shown in Figure 4, "Active Power," "Reactive Power," "PV (photovoltaics) Power Generation," and "Battery Discharge" are specified. The "Acquisition Range (Area)" section specifies the area where measurement data is collected. In the example shown in Figure 4, "City A" is specified. The "Acquisition Range (Time)" section specifies the period (acquisition time) for collecting measurement data. In the example shown in Figure 4, "Previous Week" is specified.

[0050] The "Measurement Granularity" field specifies the coarseness (fineness) of the measurement data, and in the example shown in Figure 4, "30-minute value" is specified. The "Type" field specifies the format (file type) of the data to be input into the application. Specifically, "Type" can specify information such as table (csv: Comma Separated Values), json (JavaScript® Object Notation), or xml (Extensible Markup Language), and in the example shown in Figure 4, "csv" is specified.

[0051] "Completeness" specifies whether or not the collected measurement data must be free from omissions or inconsistencies; in the example shown in Figure 4, "Required" is specified. "Degree of Processing" specifies whether or not processing is necessary for the collected measurement data, such as normalization, unit unification, logical operations, or particle size conversion, and if necessary, the type of processing to be performed; in the example shown in Figure 4, "Normalization" and "Unit Unification" are specified. "Freshness" specifies whether or not the latest measurement data is required; in the example shown in Figure 4, "Required" is specified.

[0052] The "collection rate" specifies the percentage of measurement data collected within the collection area, the percentage of measurement data collected at the collection time, the percentage of data items that have been collected, and information indicating the AND / OR relationships between these percentages, as requested by the tenant. In the example shown in Figure 4, the "collection rate" is specified as "OR," "Area: 80% or more," "Time: 70% or more," and "Data items: 80% or more." This means that the requirements for the "collection rate" are that the percentage of measurement data collected within the collection area is 80% or more, or the percentage of measurement data collected at the collection time is 70% or more, or the percentage of data items that have been collected is 80% or more. If all measurement data is available, the percentage of measurement data collection (collection rate) will be 100%.

[0053] The percentage of measurement data collected within the collection area and the percentage of measurement data collected at the collection time can be obtained by referring to the Common DB Catalog Status Information Table 204 (see Figure 7 below). However, the percentage of data items that have been collected can be determined based on the collection degree information for each data item (data identification information) specified in the Common DB Catalog Status Information Table 204 below. For example, the percentage of data items requested by the tenant that have been collected at a value equal to or greater than a predetermined collection degree may be used as the percentage of data items that have been collected.

[0054] The "execution deadline" (input deadline) specifies the deadline for providing input data (app dataset) to the application. In the example shown in Figure 4, this deadline is set to "30 minutes past every hour" (0:30, 1:30, ..., 23:30).

[0055] (3) Application Dataset Information Storage Unit The application dataset information storage unit 112 stores, in a table format, information (hereinafter referred to as "application dataset information") that associates the following for each application targeted by the tenant: application dataset identification information (hereinafter referred to as "dataset identification information"), application data requirements definition 201, application identification information (hereinafter referred to as "application identification information"), data preparation flow identification information (hereinafter referred to as "preparation flow identification information"), reference destination information in the application DB 42 (application dataset storage destination information: hereinafter referred to as "application DB reference destination information"), operating status information, and the update date and time of this information. Furthermore, within the application dataset information table 202, the application data requirements definition 201 is set with data items, acquisition range (area), acquisition range (time), measurement granularity, type, completeness, processing degree, freshness, collection degree, and execution deadline.

[0056] Figure 5 shows an example configuration of the application dataset information table 202 stored in the application dataset information storage unit 112. In the application dataset information table 202 shown in Figure 5, each row contains a set of various specified pieces of information, which constitutes the application dataset information. Specifically, the application dataset information is composed of the following associated pieces of information: dataset identification information 301, application identification information 302, application data requirements definition 201, preparation flow identification information 303, application DB reference information 304, operating status 305, and the update date and time 306 for this information.

[0057] For example, in the first row of the application dataset information table 202 shown in Figure 5, the following are associated and set: dataset identification information "Dataset_1", application identification information "App_A", each data requirement of the application data requirements definition 201 explained in Figure 4, preparation flow identification information "Flow_1", application DB reference information "AppDB_addrA", operating status "Normal", and update date and time "24 / 9 / 27 20:00:00".

[0058] In this embodiment, even if the target application is different, if the data requirements of the application data requirements definition 201 are the same except for freshness, the same data preparation flow will be used. For example, in the application dataset information (dataset identification information "Dataset_2") of the application identification information "App_B" specified in the second row of the application dataset information table 202 shown in Figure 5, the data requirements other than freshness 317 are the same as those of the application dataset information of the application identification information "App_A" specified in the first row. Therefore, in the application dataset information of the application identification information "App_B" specified in the second row of the application dataset information table 202 shown in Figure 5, "Flow_1" is specified in the preparation flow identification information 303, just as in the application dataset information specified in the first row.

[0059] (4) Requirements - Processing Microservice Correspondence Table Storage Unit The requirements-processing microservice correspondence table storage unit 113 stores a requirements-processing microservice correspondence table 203 which defines the correspondence between the values ​​(information) that can be specified for each data requirement of "completeness," "degree of processing," and "type" as defined in the application data requirements definition 201 (see Figure 4), and the target microservices used in accordance with those values ​​(information).

[0060] Figure 6 shows an example configuration of the requirements-processing microservice correspondence table 203 stored in the requirements-processing microservice correspondence table storage unit 113. As shown in Figure 6, the requirements-processing microservice correspondence table 203 defines the correspondence between each data requirement 321 of "completeness," "degree of processing," and "type," their possible values ​​(information) 322, and the target microservice (MS) 323 used in accordance with those values ​​(information).

[0061] In the example shown in Figure 6, the value "Required" for the data item "Integrity" is associated with a microservice called "Cleansing MS," which performs processing such as data completion and outlier removal. Note that if the value "Not Required" for the data item "Integrity" is not specified, no target MS is defined, and in this case, no processing based on the value of the data item "Integrity" is performed.

[0062] In the example shown in Figure 6, the value "Normalization" for the data item "Processability" is associated with a microservice called "Value Normalization Processing MS" that performs normalization processing on the measurement data. The value "Unit Unification" for the data item "Processability" is associated with a microservice called "Unit Alignment Processing MS" that performs processing to unify the unit of the measurement data. The value "Logical Operation" for the data item "Processability" is associated with a microservice called "Logical Operation Processing MS" that performs logical processing on the measurement data. Note that if the value of the data item "Processability" is "None", no target MS is defined, and in this case, no processing based on the value of the data item "Processability" is performed.

[0063] Furthermore, in the example shown in Figure 6, the value "csv" for the data item "Type" is associated with a microservice called "CSV Structured Processing MS" that converts the format (file type) of the measurement data to be input into the application into CSV type. The value "json" for the data item "Type" is associated with a microservice called "JSON Structured Processing MS" that converts the format of the measurement data to be input into the application into JSON type. The value "xml" for the data item "Type" is associated with a microservice called "XML Structured Processing MS" that converts the format of the measurement data to be input into the application into XML type.

[0064] (5) Common DB Catalog / Status Information Storage Unit The common DB catalog / status information storage unit 114 stores, in a table format, information defining the correspondence between the identification information of each data stored in the common DB 41 (hereinafter referred to as "data identification information"), the reference destination (storage location) of each data within the common DB 41 (hereinafter referred to as "reference destination pointer information"), the degree of collection of each data (hereinafter referred to as "collection degree information"), the latest collection time of each data, and the update date and time of this information.

[0065] Figure 7 shows an example configuration of the common DB catalog / status information table 204 stored in the common DB catalog / status information storage unit 114. As shown in Figure 7, the common DB catalog / status information table 204 defines the association between data identification information 331, reference pointer information 332, the latest data collection time 333, collection degree information (area) 334, collection degree information (time) 335, and the update date and time 336 of this information. In the example shown in Figure 7, the data identification information "Data X", reference pointer information "ComDB_addrX", latest collection time "24 / 9 / 27 18:30:00", collection degree information (area) "90%", collection degree information (time) "90%", and update date and time "24 / 9 / 27 19:00:00" are defined in association.

[0066] (6) Application DB Catalog / Status Information Storage Unit The application DB catalog / status information storage unit 115 stores, in table format, information that defines the correspondence between the dataset identification information, application identification information, preparation flow identification information, and application DB reference destination information specified for each application in the application dataset information table 202 in Figure 5, the latest creation time of the data preparation flow, and the update date and time of this information.

[0067] Figure 8 shows an example configuration of the application DB catalog / status information table 205 stored in the application DB catalog / status information storage unit 115. Note that Figure 8 is an example configuration of the application DB catalog / status information table 205 created in accordance with the application dataset information table 202 shown in Figure 5. As shown in Figure 8, the application DB catalog / status information table 205 defines the association between dataset identification information 341, application DB reference information 342, preparation flow identification information 343, application identification information 344, the latest generation time of the data preparation flow 345, and the update date and time 346 of this information.

[0068] In the example shown in Figure 8, the dataset identifier "Dataset_1", the application DB reference information "AppDB_addrA", the preparation flow identifier "Flow_1", the application identifier "App_A", the latest generation time "24 / 9 / 27 19:30:00", and the update date and time "24 / 9 / 27 19:50:00" are defined in association with each other. Also in the example shown in Figure 8, the dataset identifier "Dataset_2", the application DB reference information "AppDB_addrB", the preparation flow identifier "Flow_1", the application identifier "App_B", the latest generation time "24 / 9 / 27 19:40:00", and the update date and time "24 / 9 / 27 20:00:00" are defined in association with each other.

[0069] As explained in Figure 5, the data requirements other than freshness in the application data requirements definition 201 for application identifier "App_B" (dataset identifier "Dataset_2") are the same as those in the application data requirements definition 201 for application identifier "App_A" (dataset identifier "Dataset_1"). Therefore, in the example shown in Figure 8, the preparation flow identifier ("Flow_1") corresponding to application identifier "App_B" will be the same as that corresponding to application identifier "App_A".

[0070] (7) Microservices information storage unit The microservice information storage unit 116 (not shown in the figure) stores, for each microservice (MS) available in the power service provision system 1, for example, the identification information of the microservice and the address of the microservice's storage location (reference location) within the microservice DB 43, in association with each other. The data stored in the microservice information storage unit 116 is referenced when the control plane 15 (data preparation processing control unit 101) extracts the microservices necessary when creating a data preparation flow.

[0071] [Operation of the power service provision system] Next, referring to the diagrams, we will explain the series of operations performed in the power service provision system 1, from the pre-creation of the data preparation flow to application execution.

[0072] (Operation overview) Figure 9 is a diagram illustrating the series of operations from the pre-creation of the data preparation flow performed by the power service provision system 1 to application execution when it receives application data requirements definitions for a specified application execution from a tenant. The example shown in Figure 9 illustrates the operation of the power service provision system 1 when application data requirements definitions RA to RN for the target application are sent to the power service provision system 1 from each of tenants A to N.

[0073] In the example shown in Figure 9, the application executed by tenant A is designated as Application A, the application executed by tenant B is designated as Application B, the application executed by tenant C is designated as Application C, and the application executed by tenant N is designated as Application N. Furthermore, applications A, B, C, and N are to be executed simultaneously (simultaneous execution). In the example shown in Figure 9, tenant A could be, for example, a resource aggregator, tenants B and C could be, for example, solar power generation backgrowth operators, and tenant N could be, for example, a wind power generation backgrowth operator.

[0074] Furthermore, in the example shown in Figure 9, it is assumed that the data requirements other than freshness (data items, acquisition range (area), measurement granularity, etc.) are the same between the application data requirement definition RA requested by tenant A and the application data requirement definitions RB and RC requested by tenants B and C. In the application data requirement definition RA requested by tenant A, the value of freshness is set to "required," while in the application data requirement definitions RB and RC, the value of freshness is set to "not required." In addition, in the application data requirement definition RN requested by tenant N, the data requirements other than freshness (data items, measurement granularity) are different from those of the application data requirement definition RA. Furthermore, in the example shown in Figure 9, for the sake of simplicity, it is assumed that the application data requirement definitions are transmitted to the power service provision system 1 (data preparation function unit 100) in the order of tenants A, B, C, ..., N.

[0075] In the example shown in Figure 9, first, when the power service provision system 1 receives the application data requirements definition RA from tenant A, the pre-flow creation unit 101a (see Figure 3) within the data preparation processing control unit 101 creates a data preparation flow that satisfies the application data requirements definition RA (pre-creation process W1 of the data preparation flow in Figure 9).

[0076] In the pre-creation process W1 of the data preparation flow, the pre-flow creation unit 101a first selects a predetermined data extraction MS from the microservice DB 43. At this time, the pre-flow creation unit 101a sets the data search conditions for when the data extraction MS extracts the necessary measurement data from the common DB 41, specifying the data items, acquisition range (area), acquisition range (time), and measurement granularity values ​​defined in the application data requirements definition RA.

[0077] Next, the pre-flow creation unit 101a selects data processing MS from the microservice DB 43 that satisfy the data requirements for completeness and degree of processing specified in the application data requirements definition RA, as well as the requirements-processing microservice correspondence table 203 (see Figure 6). At this time, one or more data processing MS may be selected. If the completeness value is "not required" and the degree of processing value is "none", no data processing MS will be selected.

[0078] Next, the pre-flow creation unit 101a selects data structure MSs from the microservice DB 43 that satisfy the data requirements, based on the data requirements of the specified type in the application data requirements definition RA and the requirements-processing microservice correspondence table 203 (see Figure 6).

[0079] The pre-flow creation unit 101a then creates a data preparation flow by combining the selected data extraction MS, data processing MS, and data structuring MS so that they are executed in that order. At this time, the pre-flow creation unit 101a inputs existing data (e.g., historical data, test data, etc.) into the created data preparation flow and performs a test execution to calculate the predicted processing time of the data preparation flow. The identification information, program, and predicted processing time of the data preparation flow created by the pre-flow creation unit 101a based on the application data requirements definition RA are input from the pre-flow creation unit 101a to the runtime adjustment unit 101b (see Figure 3) and are also stored in the preparation flow DB 44 (see Figure 3) in the DB server 14.

[0080] Next, when the data preparation flow is input from the pre-flow creation unit 101a to the runtime adjustment unit 101b, the runtime adjustment unit 101b refers to the data requirements specified in the application data requirements definition RA, such as freshness, completeness, collection degree, and execution deadline, the predicted processing time in the data preparation flow, the management information in the resource management unit 103, and the application dataset information stored in the application dataset information storage unit 112 (see Figure 5), and adjusts the execution timing of the data preparation flow (including setting whether to execute or not execute the data preparation flow) as needed (runtime adjustment process W2 of the data preparation flow in Figure 9). In this example, since the value of freshness specified in the application data requirements definition RA that the runtime adjustment unit 101b refers to is "Required", the data preparation flow created in W1 in Figure 9 is set (input) to the predetermined AP server (AP server 11-1 in this example) in runtime adjustment process W2.

[0081] Next, the data preparation flow created in W1 in Figure 9 is set on the AP server 11-1, and then the data preparation flow is executed on the AP server 11-1 (data preparation process W3 in Figure 9). At this time, the AP server 11-1 reads the latest measurement data from the common DB 41, executes the data preparation flow, and creates the application dataset Ds1, which will be the input data for application A. The application dataset Ds1 is a dataset of each data item (e.g., active power, reactive power, PV generation amount, battery discharge amount, etc.) for the acquisition range (area) and acquisition range (time) defined in the application data requirements definition RA, and is stored in the application DB 42 (see Figure 3) at the address specified in the application DB reference information.

[0082] Then, after the creation of the application dataset Ds1, the application A, which is the target of execution for tenant A, is executed on the AP server 11-2 at a predetermined time using the application dataset Ds1 (application execution process W4 in Figure 9).

[0083] Furthermore, in the example shown in Figure 9, the power service provision system 1 receives an application data requirements definition RA from tenant A, and then receives an application data requirements definition RB from tenant B. Note that the data requirements of tenant B's application data requirements definition RB, other than freshness, are the same as those of tenant A's application data requirements definition RA. In this case, the data preparation flow created based on tenant A's application data requirements definition RA can be used. Therefore, when the power service provision system 1 receives the application data requirements definition RB from tenant B, the pre-creation process W1 of the data preparation flow by the pre-flow creation unit 101a and the runtime adjustment process W5 of the data preparation flow by the runtime adjustment unit 101b are not performed.

[0084] Furthermore, since the specified freshness value is "unnecessary" in the application data requirements definition RB for tenant B, the application dataset Ds1 created based on the application data requirements definition RA for tenant A can also be used as input data for application B, which is the target of execution for tenant B. Therefore, in the example shown in Figure 9, the data preparation process W6 based on the application data requirements definition RB for tenant B is not performed. Then, after the creation of the application dataset Ds1, application B is executed on the AP server 11-3 using the application dataset Ds1 at the same time as the execution timing of application A (application execution process W7 in Figure 9).

[0085] Furthermore, in the example shown in Figure 9, the power service provision system 1 receives the application data requirements definition RB from tenant B, and then receives the application data requirements definition RC from tenant C. Note that the data requirements of tenant C's application data requirements definition RC, other than freshness, are the same as those of tenant A's application data requirements definition RA. In this case, the data preparation flow created based on tenant A's application data requirements definition RA can be used. Therefore, when the power service provision system 1 receives the application data requirements definition RC from tenant C, the pre-creation process W1 of the data preparation flow by the pre-flow creation unit 101a and the runtime adjustment process W8 of the data preparation flow by the runtime adjustment unit 101b are not performed.

[0086] Furthermore, since the specified freshness value is "unnecessary" in the application data requirements definition RC for tenant C, the application dataset Ds1 created based on the application data requirements definition RA for tenant A can also be used as input data for application C, which is the target of execution for tenant C. Therefore, in the example shown in Figure 9, the data preparation process W9 based on the application data requirements definition RC for tenant C is not performed. Then, after the creation of the application dataset Ds1, application C is executed on the AP server 11-4 using the application dataset Ds1 at the same time as the execution timing of application A (application execution process W10 in Figure 9).

[0087] Furthermore, in the example shown in Figure 9, the power service provision system 1 receives an application data requirements definition RC from tenant C, and then receives an application data requirements definition RN from tenant N. Note that the data requirements of tenant N's application data requirements definition RN are different from those of tenant A's application data requirements definition RA. Therefore, upon receiving the application data requirements definition RN from tenant N, the pre-flow creation unit 101a creates a data preparation flow that satisfies the application data requirements definition RN (pre-creation process W11 of the data preparation flow in Figure 9).

[0088] In the data preparation flow pre-creation process W11, the pre-flow creation unit 101a first selects a predetermined data extraction MS from the microservice DB 43. At this time, the pre-flow creation unit 101a sets the data search conditions for when the data extraction MS extracts the necessary measurement data from the common DB 41, specifying the data items, acquisition range (area), acquisition range (time), and measurement granularity values ​​defined in the application data requirements definition RN.

[0089] Next, the pre-flow creation unit 101a selects data processing MS from the microservice DB 43 that satisfy the data requirements for completeness and degree of processing specified in the application data requirements definition RN, as well as the requirements-processing microservice correspondence table 203 (see Figure 6). At this time, one or more data processing MS may be selected. If the completeness value is "not required" and the degree of processing value is "none", no data processing MS will be selected.

[0090] Next, the pre-flow creation unit 101a selects data structure MSs from the microservice DB 43 that satisfy the data requirements, based on the data requirements of the specified type in the application data requirements definition RN and the requirements-processing microservice correspondence table 203 (see Figure 6).

[0091] The pre-flow creation unit 101a then creates a data preparation flow by combining the selected data extraction MS, data processing MS, and data structuring MS so that they are executed in that order. At this time, the pre-flow creation unit 101a inputs existing data (e.g., historical data, test data, etc.) into the created data preparation flow and performs a test execution to calculate the predicted processing time of the data preparation flow. The identification information, program, and predicted processing time of the data preparation flow created by the pre-flow creation unit 101a based on the application data requirements definition RN are input from the pre-flow creation unit 101a to the runtime adjustment unit 101b (see Figure 3) and are also stored in the preparation flow DB 44 (see Figure 3) in the DB server 14.

[0092] Next, when the data preparation flow is input from the pre-flow creation unit 101a to the runtime adjustment unit 101b, the runtime adjustment unit 101b refers to the data requirements specified in the application data requirements definition RN, such as freshness, completeness, collection degree, and execution deadline, the predicted processing time in the data preparation flow, the management information in the resource management unit 103, and the application dataset information stored in the application dataset information storage unit 112 (see Figure 5), and adjusts the execution timing of the data preparation flow (including setting whether to execute or not execute the data preparation flow) as needed (runtime adjustment process W12 of the data preparation flow in Figure 9). In this example, since the specified freshness value in the application data requirements definition RN referenced by the runtime adjustment unit 101b is "Required", the data preparation flow created in W11 in Figure 9 is set (input) to the predetermined AP server (AP server 11-m in this example) in the runtime adjustment process W12.

[0093] Next, the data preparation flow created in W11 in Figure 9 is set on the AP server 11-m, and then the data preparation flow is executed on the AP server 11-m (data preparation process W13 in Figure 9). At this time, the AP server 11-m reads the latest measurement data from the common DB 41, executes the data preparation flow, and creates the application dataset Ds2, which will be the input data for application N. The application dataset Ds2 is a dataset of each data item (e.g., active power, reactive power, PV generation amount, battery discharge amount, etc.) for the acquisition range (area) and acquisition range (time) defined in the application data requirements definition RN, and is stored in the application DB 42 (see Figure 3) at the address specified in the application DB reference information.

[0094] Then, after the creation of the application dataset Ds2, the application N targeted for execution by tenant N is executed on the AP server 11-n at the same time as the execution timing of application A, using the application dataset Ds2 (application execution process W14 in Figure 9).

[0095] The above-described allocation of AP servers for data preparation and application execution is merely an example, and is appropriately controlled by the control plane 15 according to the usage status (processing load) of each AP server managed by the resource management unit 103. For example, the data preparation process W3 and application execution process W4 of application A in Figure 9 may be executed on the same AP server (VM), or the data preparation process W13 and application execution process W14 of application N in Figure 9 may be executed on the same AP server (VM).

[0096] (Operation Flow) Figure 10 is a diagram showing the operation flow between each component corresponding to the operation example of the power service provision system 1 described in Figure 9. However, for the sake of explanation, Figure 10 only shows the operation flow when the power service provision system 1 receives the application data request definition RA from tenant A and the application data request definition RB from tenant B, and omits the illustration of the operation flow when the application data request definition RC from tenant C and the application data request definition RN from tenant N are received. Therefore, in the following explanation of the operation flow, the explanation of the operation flow when the power service provision system 1 receives the application data request definition RC from tenant C and the application data request definition RN from tenant N will also be omitted.

[0097] First, when the power service provision system 1 receives the application data requirements definition RA from tenant A (P1 in Figure 10), the pre-flow creation unit 101a performs the pre-creation process for the data preparation flow (P2 in Figure 10). In this process, the pre-flow creation unit 101a selects the data extraction MS, data processing MS, and data structuring MS that constitute the data preparation flow based on the various data requirements specified in the application data requirements definition RA (see Figure 4) and the requirements-processing microservice correspondence table (see Figure 6). Then, the pre-flow creation unit 101a combines these selected MS to create the data preparation flow. In addition, in this process, the pre-flow creation unit 101a tests the created data preparation flow using existing data (e.g., historical data, test data, etc.) and calculates the predicted processing time when the data preparation flow is executed.

[0098] Next, the pre-flow creation unit 101a inputs (sets) the program of the created data preparation flow, the predicted processing time, etc., into the runtime adjustment unit 101b (P3 in Figure 10).

[0099] Next, when it is time to execute the data preparation process, the runtime adjustment unit 101b adjusts the execution timing of the data preparation flow (including setting whether to execute or not execute the data preparation flow) (runtime adjustment process: P4 in Figure 10). This process is performed by referring to the data requirements specified in the application data requirements definition RA, such as freshness, completeness, collection degree, and execution deadline, the predicted processing time of the data preparation flow, the management information in the resource management unit 103, and the application dataset information stored in the application dataset information storage unit 112. In this example, since the freshness value specified in the application data requirements definition RA is "Required", the runtime adjustment unit 101b inputs the data preparation flow to the AP server 11-1 and performs the data preparation process startup (P5 in Figure 10).

[0100] Next, the AP server 11-1 reads the latest measurement data from the common DB 41, executes the data preparation flow (data preparation process), and creates the application dataset Ds1 (P6 in Figure 10). Then, the AP server 11-1 stores the application dataset Ds1 in the application DB 42 (P7 in Figure 10).

[0101] In the example shown in Figure 10, the AP server 11-1 performs the following processes at predetermined intervals: receiving various measurement data from multiple distributed power sources connected to the power service provision system 1, converting the received measurement data, and storing the converted measurement data in the common DB 41 (PL in Figure 10). However, the present invention is not limited thereto, and the collection of various measurement data from multiple distributed power sources connected to the power service provision system 1 may be performed by one or more AP servers other than the AP server 11-1.

[0102] Then, after the application dataset Ds1 has finished being stored in the application DB42, when it is time to execute application A, the AP server 11-2 reads the application dataset Ds1 from the application DB42 and executes application A (P8 in Figure 10).

[0103] In this example, the pre-flow creation unit 101a receives the application data requirements definition RB from tenant B at a specific timing after the power service provision system 1 receives the application data requirements definition RA from tenant A (P9 in Figure 10). However, in the example shown in Figure 10, the data requirements other than freshness specified in tenant B's application data requirements definition RB (data items, acquisition range (area), measurement granularity) are the same as those in tenant A's application data requirements definition RA. Therefore, for tenant B's request, the data preparation flow created based on tenant A's application data requirements definition RA can be used, so the pre-flow creation unit 101a does not perform pre-creation processing of the data preparation flow based on the application data requirements definition RB, and the runtime adjustment unit 101b also does not perform runtime adjustment processing of the data preparation flow.

[0104] Furthermore, in this example, since the specified freshness value in tenant B's application data requirements definition RB is "unnecessary," the application dataset Ds1 created based on tenant A's application data requirements definition RA can be used, and therefore no data preparation processing is performed. Consequently, after receiving the application data requirements definition RB from tenant B, when it is time to execute application B, i.e., application A, the AP server 11-3 reads the application dataset Ds1 stored in the application DB 42 and executes application B (P10 in Figure 10).

[0105] [Processing flow for pre-creation of data preparation flow] Next, with reference to Figure 11, we will explain the specific details of the data preparation flow pre-creation process (data preparation method) performed by the pre-flow creation unit 101a (see Figure 3) within the data preparation processing control unit 101. Figure 11 is a flowchart showing the procedure for the data preparation flow pre-creation process performed by the pre-flow creation unit 101a.

[0106] First, the pre-flow creation unit 101a reads (acquires) the application data requirements definition received by the power service provision system 1 (S1).

[0107] Next, the pre-flow creation unit 101a compares the application data requirements definition read in the S1 process with the application dataset information stored in the application dataset information storage unit 112 (see Figure 5) (S2). In this process, data requirements other than those specified in the application data requirements definition read in the S1 process are compared with data requirements other than those specified in the application data requirements definition 201 within each application dataset information.

[0108] Next, the pre-flow creation unit 101a determines, based on the comparison results of S2, whether or not there is application data set information corresponding to the application data requirements definition read in the processing of S1 (S3).

[0109] In the S3 judgment process, if there is app dataset information (see Figure 5) that includes app data requirement definition 201, which specifies a data requirement with the same value as the data requirement value other than the specified freshness value in the app data requirement definition read in the S1 process, it is determined that the corresponding app dataset information exists, and the S3 judgment result is YES. In this case, it corresponds to the case where a data preparation flow for creating an app dataset (input data) that satisfies the app data requirement definition read in the S1 process has already been created. On the other hand, if there is no app dataset information that includes app data requirement definition 201, which specifies a data requirement with the same value as the data requirement value other than the specified freshness value in the app data requirement definition read in the S1 process, it is determined that the corresponding app dataset information does not exist, and the S3 judgment result is NO. In this case, it corresponds to the case where a data preparation flow for creating an app dataset (input data) that satisfies the app data requirement definition read in the S1 process has not yet been created.

[0110] In S3, if the pre-flow creation unit 101a determines that there is application dataset information corresponding to the application data requirements definition read in the processing of S1 (i.e., S3 determines YES), that is, if the data preparation flow has already been created, the pre-flow creation unit 101a performs the processing of S12 described below. In this case, the pre-creation processing of the data preparation flow by the pre-flow creation unit 101a (processing of S10 described below) is not performed. Note that in the operation example explained in Figure 9, the processing performed by the pre-flow creation unit 101a when application data requirements definitions are received from tenant B and tenant C corresponds to the processing when the determination result of S3 is YES.

[0111] On the other hand, in S3, if the pre-flow creation unit 101a determines that there is no application dataset information corresponding to the application data requirements definition read in the processing of S1 (i.e., S3 is determined to be NO), that is, if the data preparation flow has not been created, the pre-flow creation unit 101a selects a data extraction MS (S4). In this process, the pre-flow creation unit 101a selects and obtains a predetermined data extraction MS from the microservice DB 43 that can extract various measurement data specified in the data items specified in the application data requirements definition from the common DB 41.

[0112] Next, the pre-flow creation unit 101a sets the data items, acquisition range (area), acquisition range (time), and measurement granularity values ​​specified in the application data requirements definition read in the S1 process as data search conditions in the data extraction MS selected in the S4 process (S5).

[0113] Next, the pre-flow creation unit 101a refers to the requirements-processing microservice correspondence table (see Figure 6) stored in the requirements-processing microservice correspondence table storage unit 113 and selects a data processing MS from the microservice DB 43 that corresponds to the specified completeness value in the application data requirements definition (S6). If the specified completeness value in the application data requirements definition is "not required", no data processing MS is selected in this process.

[0114] Next, the pre-flow creation unit 101a refers to the requirements-processing microservice correspondence table (see Figure 6) and selects a data processing MS from the microservice DB 43 that corresponds to the processing degree value specified in the application data requirements definition (S7). If the processing degree value specified in the application data requirements definition is "none", no data processing MS is selected in this process.

[0115] Next, the pre-flow creation unit 101a determines whether the selection process in S7 has been completed for all items with a specified level of processing in the application data requirements definition (S8). In S8, if the pre-flow creation unit 101a determines that the selection process in S7 has not been completed for all items with a specified level of processing in the application data requirements definition (S8 is a NO determination), the pre-flow creation unit 101a returns to the process in S7 and repeats the processes from S7 onwards described above.

[0116] On the other hand, in S8, if the pre-flow creation unit 101a determines that the selection process in S7 has been completed for all items of processing degree specified in the application data requirements definition (if S8 is determined to be YES), the pre-flow creation unit 101a refers to the requirements-processing microservice correspondence table (see Figure 6) and selects a data structure MS from the microservice DB 43 that corresponds to the value of the type specified in the application data requirements definition (S9).

[0117] Next, the pre-flow creation unit 101a creates a data preparation flow by combining the various MS selected in the above process (S10). Then, the pre-flow creation unit 101a performs a test execution of the generated data preparation flow and calculates the predicted processing time when the data preparation flow is executed (S11).

[0118] After processing S11, or if S3 is determined to be YES, the pre-flow creation unit 101a updates the application dataset information table 202 (see Figure 5) (S12). If processing S12 is performed after processing S11, in processing S12, the pre-flow creation unit 101a adds new application dataset information to the application dataset information table 202, which associates the application data requirements definition read in processing S1 with the data preparation flow information (preparation flow identification information) newly created in processing S10. On the other hand, if processing S12 is performed when S3 is determined to be YES, in processing S12, the pre-flow creation unit 101a adds new application dataset information to the application dataset information table 202, which associates the application data requirements definition read in processing S1 with the data preparation flow information (preparation flow identification information) specified for the corresponding application dataset information.

[0119] Then, after processing in S12, the pre-flow creation unit 101a finishes the pre-creation process for the data preparation flow.

[0120] [Processing flow from runtime adjustment processing to application execution processing in the data preparation flow] Next, with reference to Figure 12, we will explain the specific details of the series of processes from the runtime adjustment processing of the data preparation flow by the runtime adjustment unit 101b (see Figure 3) within the data preparation processing control unit 101 to the application execution processing by the AP server. Figure 12 is a flowchart showing the procedure of the series of processes from the runtime adjustment processing of the data preparation flow executed by the runtime adjustment unit 101b to the application execution processing executed by the AP server.

[0121] First, the runtime adjustment unit 101b refers to the application data requirements definition that has been read by the pre-creation process of the data preparation flow by the pre-flow creation unit 101a (S1 in Figure 11) (S21).

[0122] Next, the runtime adjustment unit 101b determines whether the specified freshness value in the application data requirements definition is "required" (S22).

[0123] In S22, if the runtime adjustment unit 101b determines that the freshness value is "required" (i.e., S22 is a YES determination), the runtime adjustment unit 101b performs the process in S25 described later. In this case, the execution process of the data preparation flow in S31 described later is always performed, so the execution of the data preparation flow is effectively set based on this determination result.

[0124] On the other hand, in S22, if the runtime adjustment unit 101b determines that the freshness value is not "required" (S22 is a NO determination), the runtime adjustment unit 101b refers to the application dataset information storage unit 112 and determines whether or not there is application dataset information corresponding to the application data requirement definition (see Figure 5) (S23). In this determination process, if there is application dataset information that includes an application data requirement definition 201 that defines a data requirement with the same value as the specified freshness value in the application data requirement definition being referenced, it is determined that there is corresponding application dataset information, and the determination result in S23 is a YES determination. On the other hand, if there is no application dataset information that includes an application data requirement definition 201 that defines a data requirement with the same value as the specified freshness value in the application data requirement definition being referenced, it is determined that there is no corresponding application dataset information, and the determination result in S23 is a NO determination.

[0125] In S23, if the runtime adjustment unit 101b determines that there is application dataset information corresponding to the application data requirements definition (if S23 is a YES determination), the runtime adjustment unit 101b extracts the corresponding application dataset (application input data) from the application DB 42 based on the application DB reference information 304 (see Figure 5) specified for the application dataset information (S24). After processing in S24, the runtime adjustment unit 101b performs the processing in S33 described below. Note that if S23 is a YES determination, the execution process of the data preparation flow in S31 described below is not performed, so this determination result effectively sets the data preparation flow to not be executed.

[0126] On the other hand, in S23, if the runtime adjustment unit 101b determines that there is no application data set information corresponding to the application data requirements definition (if S23 is a NO determination), or if S22 is a YES determination, the runtime adjustment unit 101b determines whether the completeness value specified in the application data requirements definition is "required" (S25). In S25, if the runtime adjustment unit 101b determines that the completeness value is not "required" (if S25 is a NO determination), the runtime adjustment unit 101b performs the process in S28 described below.

[0127] On the other hand, in S25, if the runtime adjustment unit 101b determines that the completeness value is "required" (S25 is a YES determination), the runtime adjustment unit 101b determines whether the current measurement data collection status meets the collection degree requirements specified in the application data requirements definition (S26). This determination process is performed based on the latest collection degree status (collection degree information 334, 335) for each measurement data specified in the common DB catalog / status information table 204 (see Figure 7) stored in the common DB catalog / status information storage unit 114. Specifically, if the percentage of measurement data collected within the collection area, the percentage of measurement data collected at the collection time, and the percentage of data items that have been collected, obtained based on the latest collection degree information 334, 335 values ​​for each measurement data, do not meet the collection degree requirements specified in the application data requirements definition, that is, if there is insufficient measurement data, the determination result in S26 is a NO determination. On the other hand, if the various percentages obtained based on the latest collection degree information values ​​334,335 for each measurement data meet the collection degree requirements specified in the application data requirements definition, that is, if there is sufficient measurement data, then the judgment result in S26 will be YES.

[0128] In S26, if the runtime adjustment unit 101b determines that the current level of measurement data collection meets the requirements for the level of collection specified in the application data requirements definition (i.e., if S26 is a YES determination), the runtime adjustment unit 101b performs the process described in S28 below.

[0129] On the other hand, in S26, if the runtime adjustment unit 101b determines that the current level of measurement data collection does not meet the collection level requirements specified in the application data requirements definition (i.e., if S26 is a NO determination), the runtime adjustment unit 101b adds a data replication / supplementation process MS (cleansing MS) to the data preparation flow as one of the data processing MS (S27). In this process, the runtime adjustment unit 101b selects a predetermined data replication / supplementation process MS from the microservice DB 43. The runtime adjustment unit 101b then incorporates the selected predetermined data replication / supplementation process MS into the data preparation flow created by the pre-creation process of the data preparation flow by the pre-flow creation unit 101a. The data replication / supplementation process MS performs processing such as replicating representative values ​​of the measurement data and supplementing the missing parts of the measurement data with the replicated representative values ​​to create the missing data.

[0130] If, after processing in S27, S25 is determined to be NO, or if S26 is determined to be YES, the runtime adjustment unit 101b determines whether the time until the execution deadline specified in the application data requirements definition is less than a predetermined threshold (S28). Specifically, in processing S28, the runtime adjustment unit 101b determines whether the difference between the period from the current time until the execution deadline specified in the application data requirements definition (the deadline for providing input data to the application) and the predicted processing time of the data preparation flow is less than a predetermined threshold. The predetermined threshold used in this determination process is set in advance as appropriate.

[0131] In S28, if the runtime adjustment unit 101b determines that the time remaining until the execution deadline specified in the application data requirements definition is not less than a predetermined threshold (i.e., if S28 is determined to be NO), the runtime adjustment unit 101b performs the process described in S30 below.

[0132] On the other hand, in S28, if the runtime adjustment unit 101b determines that the time remaining until the execution deadline specified in the application data requirements definition is less than a predetermined threshold (if S28 is a YES determination), the runtime adjustment unit 101b adds resource priority control processing to the data preparation flow (S29). In resource priority control processing, the runtime adjustment unit 101b refers to the management information in the resource management unit 103 and performs configuration control (resource priority control described later) of the cloud resource (AP server) that will prioritize the execution of the data preparation flow in question, which has little time remaining until the execution deadline. In other words, in the processing of S29, the runtime adjustment unit 101b reflects resource priority control in the data preparation flow.

[0133] After processing in S29, or if S28 is determined to be NO, the runtime adjustment unit 101b inputs the data preparation flow to the AP server where the data preparation flow is executed and performs the data preparation process startup (S30). If processing in S27 is performed before processing in S30, the data preparation flow input to the AP server will be a data preparation flow in which the data replication / replenishment processing MS is incorporated as one of the data processing MSs. If processing in S29 is performed before processing in S30, the data preparation flow input to the AP server will be a data preparation flow that reflects the settings of the cloud resource (AP server) that prioritizes the execution of the data preparation flow. Furthermore, if both processing in S27 and S29 is performed before processing in S30, the data preparation flow input to the AP server will be a data preparation flow in which the data replication / replenishment processing MS is incorporated as one of the data processing MSs and in which resource priority control is reflected. On the other hand, if both S27 and S29 processes have not been performed before S30, the data preparation flow input to the AP server will be the data preparation flow created by the pre-creation process of the data preparation flow by the pre-flow creation unit 101a.

[0134] Next, the AP server, upon receiving the data preparation flow, executes the data preparation flow (S31). At this time, the AP server reads the latest measurement data from the common DB41, inputs the latest measurement data into the data preparation flow, and executes it to create an application dataset (application input data). The application dataset created here is a dataset of each data item (e.g., active power, reactive power, PV generation amount, battery discharge amount, etc.) for the acquisition range (area) and acquisition range (time) defined in the application data requirements definition. Next, the AP server stores the created application dataset in the application DB42 (S32).

[0135] After processing in S32 or S24, the AP server inputs the application dataset created in S31 or extracted in S24 into the application and executes the application (S33). The AP server that executes the application in processing S33 may be different from or the same as the AP server that executes the data preparation flow in processing S31, as shown in the operation examples described in Figures 9 and 10. After processing S33, the power service provision system 1 completes the series of processes from the runtime adjustment processing of the data preparation flow by the runtime adjustment unit 101b to the application execution processing by the AP server.

[0136] [Various effects] As described above, in the power service provision system 1 of this embodiment, if a data preparation flow that satisfies the data requirements requested by a tenant for the application to be executed has already been created, the system can utilize the already created data preparation flow without creating a new data preparation flow for the tenant's request. Therefore, in this embodiment, the man-hours required for creating the data preparation process before application execution can be reduced, and the cost of creating the application can be reduced.

[0137] In the power service provision system 1 of this embodiment, if a data preparation flow that satisfies the data requirements requested by the tenant for the application to be executed has already been created, and the tenant does not require data freshness (the freshness value is "not required"), the application dataset obtained by executing the already created data preparation flow can be used as input data for the application without executing the data preparation process. Therefore, in this embodiment, the number of data preparation processes can be kept to the minimum necessary, and the amount of cloud resources used by the power service provision system 1 can be reduced.

[0138] In the power service provision system 1 of this embodiment, when an application data requirements definition is received from a tenant, a data preparation flow that satisfies the application data requirements definition is automatically created in advance. Therefore, in this embodiment, in the development of the power service provision system 1, it is possible to focus solely on the application logic without having to be aware of changes on the data source side.

[0139] In the power service provision system 1 of this embodiment, the data preparation flow can be created by combining existing MSs during the pre-creation process of the data preparation flow. Therefore, in this embodiment, even if the number of tenant services increases, updates, and customizations occur frequently, it is possible to facilitate the development and construction of services and suppress increases in service operation costs.

[0140] In the power service provision system 1 of this embodiment, when there is insufficient application input data (application dataset), a data replication and supplementation process MS (cleansing MS) is added to the data preparation flow as one of the data processing MSs. Therefore, in this embodiment, even when there is insufficient application input data (application dataset), service information such as power supply and demand forecast results and power supply and demand plans can be provided with good accuracy.

[0141] Furthermore, in the power service provision system 1 of this embodiment, resource priority control processing is added to the data preparation flow when the period until the execution deadline (the deadline for providing input data to the application) specified in the application data requirements definition is short. Therefore, in this embodiment, service information such as power supply and demand forecast results and power supply and demand plans can be reliably provided by a predetermined deadline (for example, a bidding deadline or a deadline requested by a tenant), thereby increasing reliability.

[0142] [Various variations] The embodiments described above are detailed and specific explanations of the device's configuration for the purpose of clearly illustrating the present invention, and are not necessarily limited to those comprising all the described configurations. Furthermore, the positions, sizes, shapes, and ranges of each component shown in the drawings may not represent the actual positions, sizes, shapes, and ranges, in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the positions, sizes, shapes, and ranges disclosed in the drawings. The present invention can be modified in various ways as long as it does not depart from the gist of the invention as described in the claims. For example, the following various modifications can be adopted.

[0143] (Variation 1) In the power service provision system 1 of the above embodiment, an example was described in which each AP server (VM) is composed of one server machine (see Figure 1), but the present invention is not limited to this. For example, two or more AP servers (VMs) may be composed of one server machine, or the power service provision system 1 may be composed of one server machine.

[0144] (Modification 2) In the above embodiment of the power service provision system 1, an example configuration was described in which the microservice DB 43 is provided within the system (DB server 14), but the present invention is not limited thereto. For example, the microservice DB storing the microservices available to the power service provision system 1 may be provided outside the power service provision system 1.

[0145] In this case, the microservice information storage unit 116 may store, for example, identification information for each microservice, information specifying the storage location, etc., and the control plane 15 may be configured to access the storage location of the microservices based on this information and obtain information such as the program of the microservice. In this case, the external DB server that serves as the storage location may differ depending on the type of microservice, or all of these various microservices may be stored as a library in a single DB server regardless of the type of microservice. Furthermore, in this case, if a dedicated microservice is available for the power service provision system 1, only that dedicated microservice may be stored in the DB server 14.

[0146] (Variation 3) In the above embodiment, an example was described in which the data requirements specified in the application data requirements definition are data items, acquisition range (area), acquisition range (time), measurement granularity, type, completeness, degree of processing, freshness, degree of collection, and execution deadline (see Figure 4). However, the present invention is not limited thereto. For example, depending on the type of tenant or the type of application to be executed, the application data requirements definition may not include some of these data requirements, or data requirements other than those shown in Figure 4 may be specified in the application data requirements definition.

[0147] (Modification 4) In the above embodiment, an example of applying the technology of the present invention to a service specification in which multiple applications are executed simultaneously (sequential execution) was described, but the present invention is not limited thereto. For example, the technology of the present invention can also be applied to service specifications in which the execution timing of multiple applications is different, and similar effects can be obtained. [Explanation of Symbols]

[0148] 1…Power service provision system, 2…Power market server, 3…Tenant terminal, 4,5…DERMS server, 6~8…Distributed power source, 9…Network, 10…Power supply and demand related system, 11-1~11-n…AP server, 12…Firewall, 13…Load balancer, 14…DB server, 15…Control plane, 16…Internal network, 31~36…Containers, 31a~36a…Microservices (MS), 41…Common DB, 42…Application DB, 43…Microservice DB, 44…Preparation flow DB, 100… Data preparation function unit, 101... Data preparation processing control unit, 101a... Pre-flow creation unit, 101b... Runtime adjustment unit, 102... MS / service execution flow management unit, 103... Resource management unit, 104... Data set management unit, 105... DB / data collection status management unit, 201... Application data requirements definition, 202... Application data set information table, 203... Requirements-processing microservice correspondence table, 204... Common DB catalog / status information table, 205... Application DB catalog / status information table, Ds1, Ds2... Application data set

Claims

1. An application execution unit that runs an application capable of providing service information related to transactions in the electricity market, A data preparation flow creation unit that obtains requirements for the input data of the application from an external source and creates the data preparation flow if a data preparation flow for creating the input data that satisfies those requirements has not already been created, and does not create the data preparation flow if the data preparation flow has already been created. The system includes an input data creation unit that executes the data preparation processing flow described above to create the input data and inputs the created input data to the application execution unit. Electricity service provision system.

2. The above requirements include whether or not the input data needs to be fresh. The system further includes a data preparation processing flow execution setting unit that, when the freshness of the input data is important, sets the execution of the data preparation processing flow by the input data creation unit, and, when the freshness of the input data is not important and input data that satisfies the requirements has already been created, sets the execution of the data preparation processing flow by the input data creation unit not to be performed. The power service provision system according to claim 1.

3. The data preparation processing flow creation unit creates the data preparation processing flow by combining multiple existing microservices. The power service provision system according to claim 2.

4. The data preparation processing flow execution setting unit can add data duplication and / or supplementation processing to the data preparation processing flow based on the data collection status of the input data. The power service provision system according to claim 2.

5. The application execution unit comprises multiple such units, The data preparation processing flow execution setting unit can configure the application execution unit to prioritize the execution of the data preparation processing flow based on the deadline for inputting the input data into the application. The power service provision system according to claim 2.

6. The data preparation processing flow creation unit, when it has created the data preparation processing flow in advance, performs a test execution of the data preparation processing flow and calculates the predicted processing time for the data preparation processing flow. The data preparation processing flow execution setting unit can configure the application execution unit to prioritize the execution of the data preparation processing flow based on the input deadline and the predicted processing time. The power service provision system according to claim 5.

7. The power service provision system comprises an application execution unit that executes an application capable of providing service information related to transactions in the power market, a data preparation processing flow creation unit that can create a data preparation processing flow for creating input data for the application, and an input data creation unit that inputs the input data to the application execution unit, wherein the data preparation processing flow creation unit acquires requirements for the input data of the application from an external source, If the data preparation processing flow creation unit has not yet created the data preparation processing flow for creating the input data that satisfies the requirements, it will create the data preparation processing flow. If the data preparation processing flow creation unit has already created the data preparation processing flow for creating the input data that satisfies the requirements, it will not create the data preparation processing flow. The input data creation unit includes creating the input data by executing the data preparation processing flow. Data preparation method.

Images