Keywords

1 Introduction

The 13th five year plan for power development (2016–2020) put forward in 2016 clearly stated that we should deepen the reform of power system, improve the power market system, establish relatively independent and standardized power trading institutions, establish fair and orderly power market rules, and initially form a fully functional power market. In the information age, power data can be regarded as an important enterprise asset of power enterprises, and the effective management of massive data is an essential task. In the process of data capitalization, the value of data assets can be developed and improved by creating data products, carrying out data services and data circulation, in which data traceability is an important link.

Since the 11th five year plan, China has gradually incorporated energy conservation and carbon reduction into the national economic and social development plan. In September 2020, China put forward the “dual carbon” goal, striving to achieve carbon peak by 2030 and carbon neutralization by 2060. The power sector is one of the key areas to achieve the goals of carbon peak and carbon neutralization. Green power trading can promote clean and low-carbon production of energy, ensure low-carbon operation of power systems, and help enterprises accelerate the completion of the overall goal of carbon peak and carbon neutralization, so as to achieve a high degree of unity of economic, social and environmental benefits, so as to form a new power system dominated by clean energy, It is an important means to realize the reform and efficient development of the power industry, an inevitable need to adapt to the development trend of technological progress in the energy field and promote the transformation and upgrading of the system, and an urgent need to promote the transformation of energy to clean and low-carbon. The traceability of green power product information can fully ensure that the user can purchase pure “green power”, and ensure the consistency of green power products and consumption from the perspective of business relationship and power generation and supply relationship, so as to avoid production waste and achieve real “zero carbon emission”.

In the current green power transaction process, there are many problems, such as difficult to identify the authenticity of power data, unreliable transaction records, imperfect supervision and so on. Therefore, it is necessary to put forward a reliable traceability solution for power data. The challenge of power data traceability makes the use of new technologies an inevitable trend. Blockchain integrates multiple core technologies, such as distribution, cryptography, consensus mechanism and smart contract, and has been widely used in all walks of life, including finance, food, logistics, etc. The characteristics of blockchain, such as transparency, multi-party consensus, tamper proof and traceability, have brought a new cracking idea to the reliable traceability of power data.

2 Related Work

The development of blockchain technology has brought a new development direction to the energy industry and can promote the intelligent development of power grid system. Reference [1] proposed the concept of double chain blockchain, which separates data blocks and transaction blocks, only adds power data asset summary to the consensus, and uses the interactive process of biological gene expression for reference to take the blockchain storing operation data as DNA blockchain, and the blockchain storing transaction information and data summary as RNA blockchain. Reference [2] designed a power data asset trading platform, in which a random authentication model for identity management was designed through Ethereum’s smart contract, POA consensus mechanism was adopted, periodic monitoring synchronization mechanism and data access tracking mechanism based on B+tree were proposed. Reference [3] studies the power data management architecture, analyzes the power data recording process, and ensures that the data is not tampered with by means of asymmetric encryption between block nodes and power data recording nodes; At the same time, a process of power data sharing is proposed by using asymmetric encryption technology; The security management guarantee system of power data is proposed from two aspects of storage security and transmission security.

At present, there is still a lack of recognized concept of data traceability. The concepts vary due to different application fields. Here, refer to the definitions of some scholars. In reference [4], it is defined as the origin of data and its movement between databases in the field of database system. In reference [5], data traceability is defined as a kind of metadata, which is used to record the evolution process of workflow, annotation information, experimental process changes and other information; Reference [6] puts forward that all kinds of data have a life cycle, and data traceability is to record the changes and processed information of data in the whole life cycle from generation to extinction or transformation. The concept of W3C is that “the traceability of each resource is a record that describes the entity and the contained production and delivery processes or otherwise affects the resource.” At the same time, there are many forms of data traceability models. Reference [7] studies the OPM model, which has three core concepts: Artifact, Process and Agent. Artifact can be a physical object or a digital representation in a computer system. Process represents a single or a series of actions generated by Artifact. Agent is the main influencing factor of process, which can promote, control and affect the execution of Process. Reference [8] studies the PROV-DM model. Its core structure also includes three common aspects of Traceability Technology: Entity, Activity and Agent. Entity can be concrete objects such as buildings, or abstract concepts such as ideas; Activity is the action on entity; Agency, as an object, plays a certain role for the existence of Entity and Activity, which is essentially a special Entity or Activity.

At present, there are two solutions to the attribute based data access control mechanism, namely CP-ABE (ciphertext policy attribute based encryption) and KP-ABE (key policy attribute based encryption). In Reference [9], the CP-ABE mechanism is used and the original Fabric CA module in hyperledger fabric is combined to realize user level fine-grained secure access control of blockchain data.

3 Power Data Traceability Based on Blockchain

In this paper, a three-tier architecture is established to realize the reliable traceability of power data based on blockchain. As shown in Fig. 1, the structure is composed of a data generation layer, a data conversion layer and a blockchain network layer. The functions of each layer are described as follows:

Data generation layer: it is composed of various sensors to obtain the power data generated in the whole life cycle of power data, including the whole process of power generation, transmission, transformation, distribution and power consumption.

Data conversion layer: it consists of local nodes. The power data of the data generation layer is converted into data meeting the storage requirements through the traceability data model, and the original power data is stored locally.

Blockchain network layer: it cosists of blockchain nodes. The identification and hash of the original power data in the data conversion layer are stored in the blockchain to realize the lightweight management of power data. The blockchain uses the raft consensus mechanism to ensure the credibility of the data on the chain. At the same time, it uses the smart contract to query the data records, and prevents the data on the chain from being tampered by comparing with the original power data stored locally.

Fig. 1.
figure 1

Power data traceability architecture based on blockchain

Full size image

The data storage process is shown in Fig. 2. In the power generation stage, the type, source and power generation of green electricity are collected through the power generation terminal; In the transmission stage, the information of substation, distribution, data collector, watt hour meter and other equipment passing through the transmission process is collected through edge equipment; In the transaction stage, the transaction data, consumer data and trader data are recorded through the power grid system; In the three stages, there are user access links, which collect access records while users access data. Finally, all data is stored through blockchain nodes.

Fig. 2.
figure 2

Data storage process

Full size image

3.1 Power Data Traceability Model

This paper constructs and describes the application scenarios based on the prov-dm model. The nodes and edges in the power data traceability model are defined according to three types and seven relationships in the PROV-DM model. Power generation companies trade electricity with power grid companies through market operation agencies, and power grid companies sell electricity to small and medium-sized end users through acquisition equipment and marketing systems. In this process, there are also operations such as data query, modification and statistics.

This paper designs the following three types of nodes:

  1. (1)

    Entity node: used to represent data. It is represented by En. E is the abbreviation of Entity;

  2. (2)

    Agent node: used to represent an individual or enterprise. It is represented by An. A is the brief form of Agent;

  3. (3)

    Activity node: used to represent operations on data. It is represented by ACn. AC is the abbreviation of Activity.

This paper designs the following five types of connecting edges:

  1. (1)

    Used: used to connect the Active node and the Entity node used in the activity process. The Active node points to the Entity node;

  2. (2)

    WasGeneratedBy: used to connect the Active node and the new Entity node generated by the activity. The Entity node points to the Active node;

  3. (3)

    WasAttributedTo: used to connect the Agent node and the Entity node. The Entity node points to the Agent node;

  4. (4)

    WasAssociatedWith: used to connect the Agent node and the Active node. The Active node points to the Agent node;

  5. (5)

    WasDerivedFrom: used to connect two Entity nodes. The new Entity node points to the old Entity node.

According to the above description, this paper constructs the power data PROV-DM model as shown in the Fig. 3. The measurement data, marketing data and other data belong to Entity nodes, which are shown as oval nodes in the figure; Power trading, data collection and other operations are Active nodes, which are shown as rectangular nodes in the figure; Power generation enterprises, power grid enterprises and other enterprises are Agent nodes, which are shown as pentagon nodes in the figure. Connect the nodes according to the five edges defined above, and finally get a directed acyclic graph.

Fig. 3.
figure 3

Power data PROV-DM model

Full size image

3.2 Power Data Model

Data Model

As shown in Fig. 4, the traceability data model includes master data, processing data and remark data.

Fig. 4.
figure 4

Data model

Full size image

Master data includes two dimensions: initiator and traceability data. The data structure of initiator and traceability data is shown in Table 1 and Table 2.

Table 1. Data structure of initiator
Full size table
Table 2. Data structure of traceability data
Full size table

Processing data includes two dimensions: data source and data operation. The data structure of data source and data operation is shown in Table 3 and Table 4.

Table 3. Data structure of data source
Full size table
Table 4. Data structure of data operation
Full size table

For the remark information in the remark data, it’s able to add some user-defined information.

Data Identification Scheme

This paper constructs a set of identification rules to make the traceability data of power supply chain unique in the world. This section mainly describes in detail the identification scheme of the Agent and Entity nodes mentioned above.

  1. (1)

    Agent: different Agents must be distinguished in the traceability system. Specific examples of Agent may include power generation enterprises, market operation institutions, power grid companies or individuals. When enterprises are involved, they can be identified by using Global Location Number (GLN); When an individual is involved, the Global Service Relationship Number (GSRN) can be used for identification. See Table 5 for details.

Table 5. GS1 standard
Full size table
  1. (2)

    Entity: identify power data according to physical dimension, application dimension and time dimension, as shown in Fig. 5.

Fig. 5.
figure 5

Three dimensions scheme

Full size image

Data Access Control

CP-ABE (Ciphertext Policy Attribute Based Encryption) means that the ciphertext corresponds to an access structure and the key corresponds to the attribute set. When decrypting, only the attribute values in the attribute set can meet the access structure in the ciphertext. All users obtain the key from the attribute organization through their own conditions and attributes, and the encryptor formulates the access control of the message.

According to the characteristics of Hyperledger Fabric with channel isolation, this paper designs three attributes: channel ID, organization ID and user ID. The specific access structure example is shown in Fig. 6. This access structure indicates that the specified users usei and userj in orgi of the specified organization in the specified channel channeli have access rights. In this paper, the Fabric CA module is used to realize the initialization, key generation and distribution of CP-ABE, and the ciphertext is stored on the chain to achieve more secure data access control.

Fig. 6.
figure 6

Access structure

Full size image

4 Experiments and Analysis

4.1 Storage Module

Storage modules are divided into local storage module and blockchain storage module, both of which are run in docker. The local storage module selects the Neo4j diagram database, and Fig. 7 shows the Neo4j container. Hyperledger Fabric is selected for the blockchain storage module. Figure 8 shows the Hyperledger Fabric container.

Fig. 7.
figure 7

Local storage module

Full size image
Fig. 8.
figure 8

Blockchain storage module

Full size image

4.2 Data Operation Module

The project includs entity layer, mapper layer, service layer and controller layer. Entity layer corresponds to entities in the database; The mapper layer mainly does the work of the data persistence layer. Some tasks responsible for contacting the database are encapsulated here, specifically encapsulating the basic operations of adding, deleting, modifying and querying; The service layer is mainly responsible for the application logic design of the business module. First, the interface is designed, then its implementation class is designed, and one or more mappers are encapsulated again to form a service; The controller layer is responsible for the request forwarding, accepts the parameters passed from the front end, calls the interface of the service layer to control the business process, and gets the return value before passing it to the page.

4.3 Data Display Module

The data display module selects the Vue framework and divides the front end into login module, registration module and display module. The display module is divided into three sub modules: top navigation bar, side navigation bar and main module.

In view of the lack of power data sets, the power data upload function is designed, which is mainly used to simulate the generation of power data. Power data query is divided into ordinary query and traceability query. Ordinary query displays the data in the form of list, including detailed data attributes. The data integrity ensures that the data is not tampered by comparing the local data with the hash data on the chain; The traceability query uses the open source visualization library Echarts to display the data in the form of a graph, including only the identification of the data, as shown in Fig. 9.

Fig. 9.
figure 9

System function

Full size image

According to the statistics, the time-consuming of the relevant functions of the system is shown in. Ordinary query needs to compare the data on the chain and the locally stored data to prevent data from being tampered with, so it takes a long time; Traceability query can quickly display local data in the form of graph and takes a short time; Data uploading requires data to be stored locally and on the blockchain, so it takes a long time (Table 6).

Table 6. System function time consuming
Full size table

5 Conclusion

This paper analyzes the power data traceability scenarios and entities involved in the power life cycle. With the help of PROV-DM model, the power data traceability model is discussed. A multi-dimensional data model is proposed and two data identification schemes are proposed for different types of data. The data access control method based on attribute encryption is discussed. The power data traceability system architecture is designed. The blockchain achieves high data consistency, tamper proof and traceability, and one of the costs is performance. Therefore, combined with the local data storage Neo4j, the lightweight management of data on the chain is realized. At last, the power data traceability system is realized.