Model of decentralized cross-chain energy trading for power systems

0 Introduction

With the development of the energy Internet,conventional power grids allow wind power,photovoltaic and other distributed power,and flexible loads such as plugin hybrid electric vehicle (PHEV)to be connected to the grid on a large scale.In power networks [1],with the further integration of the energy Internet,some conventional power users have been transformed into power consumers with generation capacities.Consequently,heterogeneous energy trading led by these prosumers will become an important part of electricity trading.

Distributed heterogeneous energy trading [2] is characterized by high frequency and small quantity.Based on the prediction by the State Grid Energy Research Institute(2018),the disorderly charging of electric vehicles will increase the peak load of China by 13.61 and 153 million kW in 2020 and 2030,equivalent to 1.6% and 13.1% of the regional peak load (without the electric vehicle charging scenario),respectively.Wind power,a photovoltaic system,and PHEV [3] are distributed in different geographical locations,and the power generation situation is greatly affected by the natural environment and the charging and discharging habits of consumers.Centralized energy trading scheme cannot enable all kinds of distributed generation to perform efficient communication,and the flexibility is poor.If considering only the conventional method,the complementary ability of heterogeneous energy may be ignored,resulting in the formation of energy island,which will cause wind and solar energy to be disregarded.The heterogeneous energy blockchain technology uses the decentralized trading system,which can be flexibly adjusted and is characterized by peer-to-peer transaction,which can effectively solve the aforementioned problems.

Currently,domestic and foreign researchers have performed active research on the application of blockchain in energy and power trading.In [4],a distributed power trading system architecture based on blockchain technology was proposed to realize the optimal allocation of resources.In [5],a smart contract was proposed to store the electricity transaction information and automatically execute capital settlement.The central organization only conducts security check and congestion management for the completed transactions.Based on the Ethereum (ETH)network of proof of stake (PoS)and the Hyperledger Fabric network based on proof of Byzantine fault tolerance (PBFT),energy trading experiments were performed to simulate the regional energy trading lead by prosumers [6].By using smart contract as a virtual aggregator,an energy management platform based on blockchain was built to realize the optimization and coordination of information flow and energy flow in a microgrid [7].A safety charging system for electric vehicles based on blockchain was proposed,which can prevent replay and man in the middle attack and improve system security [8].By using the transactive grid instrument with the blockchain protocol [9],the power and transaction data are transferred from the intelligent instrument of the consumer to the blockchain account,and the execution of the transaction is completed by the smart contract,which improves the transaction efficiency.However,the energy Internet comprises several energy types,such as electric heating and water vapor.The increase in the number of nodes inevitably leads to slow transaction speed and long verification time,and the single chain architecture is not conducive for the expansion and maintenance of the system in the future.Therefore,in [10-13],a multichain architecture was introduced to solve these problems.Author[10] showed that different areas in a smart grid must adopt an independent blockchain system structure and transaction technology to realize the effective use of energy.In terms of the security of multichain architecture,the security problems of blockchain in a multifunctional system were analyzed [11],and the risks that are mainly derived from a single chain and cross chain interconnection were considered.By using the side-chain security mechanism based on simplified payment verification (SPV),the use of third-party supervision organization is proposed; however,the introduction of such a third-party organization could result in reduced decentralization.In addition,the credibility of the third party would need to be ensured.Aiming at the problems of mutual trust between heterogeneous energy blockchains and the lack of grid connection scalability,an energy supply consensus algorithm based on PBFT was proposed in [12],with the introduction of digital signature and the Merkle tree.However,in terms of efficiency,the efficiency of PBFT consensus decreases sharply with the increase in nodes.In cross-chain transactions,the efficiency of PBFT consensus decreases sharply.Moreover,the reliability of data on the Merkle tree must be guaranteed.To improve the transaction performance of the model,multichain fusion was used to separate the process of power matching and capital settlement [13].

Because energy transaction involves capital settlement,the security of the transaction should be guaranteed first.Simultaneously,because there are many participating nodes,a high transaction rate must be ensured.Therefore,to solve the aforementioned problems discussed in literature,this study was aimed at designing a heterogeneous energy trading model based on the cross-chain technology,which improves the efficiency and security of a transaction.In intrachain transaction,the intrachain aggregation signature authentication is realized based on the aggregation signature technology,which reduces the communication complexity of the nodes and improves the consensus efficiency.In addition,the pledge mechanism increases the cheating cost of the nodes and improves the security of the intrachain transactions.In the cross-chain transactions,the relaychain technology is used to improve the communication efficiency and transaction throughput,and the fisherman mechanism allows honest acceptors to provide invalid block information,and thus the security of cross chain transaction is improved.

1 Related technologies

1.1 Technologies of cross-chain blockchain

Owing to the introduction of Bitcoin (BTC)in 2008,there have been thousands of blockchain projects.However,because of the constraints of throughput and network isolation,these projects have not been applied in largescale businesses.The cross-chain technology is proposed to realize the interconnection of heterogeneous blockchain,break the information barrier and improve the transaction throughput.

Cross-chain technology [14],by the notary mechanism,side/relay-chain [15],and Hash locking technology.

1)The notary mechanism can be divided into three types: single signature,multisignature,and distributed signature.In the single and multi-signature notary mechanisms [16] with a strong degree of centralization,notaries are composed of specific institutions or consortia,which are responsible for cross-chain data collection,transaction confirmation,and verification; the distributed signature notary mechanism adopts the idea of multiparty computation (MPC).In each transaction,the notary will randomly extract and will receive only the processed secret key fragments; only when the number of signatures reaches the preset requirements,the complete secret key can be obtained,and this improves the security and decentralization of the whole system.

2)Side/Relay-chain: The relay-chain only participates in data collection,when the data is sent to the relay chain,it completes the transaction confirmation and verification by employing SPV,the block header data of the data transmission chain,and the number of signature verification nodes.Polkadot [17] assumes the collector,fisherman,nominator,and verifier roles to take on different functions,with the help of the reward and punishment mechanism and an incentive model to complete cross-chain communication.

3)Hash locking: By running smart contracts or scripts,interoperability triggers,such as a random number hash value,are set between different chains.Combined with a time lock,if the original random number is solved based on the hash value within a specified time,the asset transfer is performed.The lighting network uses BTC transmission to construct a two-way payment channel to conduct cryptocurrency transactions.

In the heterogeneous energy blockchains,based on the cross-chain technology,the solar power generation,wind power generation,and PHEV energy networks are integrated [18] to achieve the goal of multifunctional complementary trading in the energy Internet,where users connect,exchange multi-energy information,accelerate energy circulation,and improve transaction throughput.

1.2 Technologies of heterogeneous energy trading

The conventional electricity trading market is a centralized structure,which needs to rely on audit companies,banks,and other institutions,with a long transaction cycle and high cost.The information flow between different energy sources is slow,and the cost of establishing trust between users is high.Therefore,blockchain technology is applied to the energy Internet [19-21].In power transactions,all consumers make decentralized decisions equally,and direct energy transactions are conducted in a Peer-to-Peer (P2P)form.Each node in the blockchain has equal status and is interconnected and interacts with each other in a flat topology structure.Energy information is fully shared,which can reduce transaction and trust costs.

Blockchain technology can realize the data transparency of the power trading market and improve mutual trust among users [22].Currently,there are two management modes of power transaction: centralized and distributed,both of which need to build trust for the following reasons:

Security issues of power transaction: power transaction involves the privacy of power users.However,in the process of data transmission and processing,privacy leakage and network attacks may occur,such as the authentication of the parties to the transaction,repeated violation of electronic invoices,forgery of bills,transaction denial,fraud,and other issues [23].

Cost issues of power transaction: in the centralized power transaction system,to build a reliable and secure network,it needs to spend a lot of costs,leading to the rise of transaction cost [24].Therefore,in the transaction,a high level of cooperation can be achieved through the exchange relationship based on trust,which reduces the transaction cost of both parties and improves the competitive advantage[25].The blockchain technology,through time stamp,Merkle tree,asymmetric encryption and other technologies,realizes the unforgeability,and traceability of power transaction data,which is shown in Fig.1; thus,it can ensure the security of transaction information,reduce the cost of mutual trust,and improve the participation of users.

Fig.1 Energy blockchain structure

In the case of a centralized power trading center,it is difficult to deal with the transaction requests of numerous distributed devices in the energy Internet in real time.The transaction mechanism,pricing strategy,payment settlement,and other functions can be programmed into intelligent contracts that can be executed automatically through blockchain technology,to ensure the execution efficiency of transactions.Simultaneously,the transaction records are placed on the chain to avoid default disputes.

2 Construction of heterogeneous energy blockchain transaction model

Aiming at the problems of low mutual trust,low efficiency of power transaction,and poor information communication among heterogeneous systems in energy Internet,based on the design principles of atomicity,efficiency,security,generality,and friendliness,a heterogeneous energy blockchain transaction model based on energy Internet (EI-HEB)is proposed,which is shown in Fig.2.This section details the model in detail in terms of the intrachain and cross-chain transaction consensus mechanisms.

Fig.2 Heterogeneous energy transaction model based on cross-chain technology

2.1 Consensus mechanism of energy intrachain transaction based on aggregate signature

The three main consensus mechanisms are Proof of Work (PoW),PoS,and PBFT.Although the three algorithms have their advantages,they also have their shortcomings.PoW consensus algorithm consumes a lot of power resources.Based on the Daily Telegraph,UK,the energy consumed by BTC mining has reached 77.78 TWh per year,equivalent to the total energy consumption of Chile.PoS consensus algorithm is beneficial to nodes with more currency holdings and will cause nothing-at-stake problems.The PBFT consensus algorithm is only applicable to alliance or private chain,and the communication complexity is extremely high; as the number of nodes increases,the performance decreases exponentially.A heterogeneous energy blockchain transaction model must simultaneously meet the requirements of energy saving,efficiency,security,and scalability.Therefore,combined with the mainstream consensus mechanisms,we propose a transaction consensus algorithm for a heterogeneous energy system (HEST-CA).The implementation process of HESTCA is described as follows (taking PHEV chain as an example):

(1)When PHEV users are connected to the blockchain,if they want to be the acceptors of the PHEV chain,they must mortgage certain energy tokens.The more the energy tokens mortgaged,the greater is the right of voting.Simultaneously,the pledged tokens will generate dividends to reward the holders.

(2)PHEV-chain nodes exercise voting rights based on the weight of pledged tokens; proposers are then elected,who become speakers responsible for submitting the scheme.The elected proposers rotate regularly and confiscate their pledged tokens once cheating is detected.

The right of pledged tokens is denoted as

where m is the total amount of tokens pledged and t is the currency holding time.

(3)During the transaction,PHEV-chain nodes continuously broadcast information,such as ID,total production capacity,and electrical energy quotation,to search for PHEV power purchasers to reach the transaction intention.

(4)Assuming 1000 power transactions in PHEVchain during a certain period,the speaker packages all the transactions and distributes them to the acceptors.Concurrently,the signature aggregation is performed,and a multisignature process is constructed to collect the acceptors’ votes based on the BLS [26-28] signature scheme.Each acceptor only needs to accept one multisignature,and the communication complexity is reduced from O(n2)to O(n):

where Si represents the digital signature of each transaction.

(5)When the acceptor with 2/3 voting share signs the message released by the proposer,the 1000 PHEV transactions at this period are packaged and uploaded,and the next round of consensus is started.

HEST-CA is proposed based on PoS and PBFT consensus algorithms.PoS and PBFT are voting consensus algorithms,which require mutual authentication between nodes.Therefore,the communication complexity between nodes is usually O(n2).In addition,with the increase in the number of nodes,the consensus efficiency will decrease.The improved HEST-CA consensus algorithm adopts the BLS signature scheme for signature and key aggregations;this reduces the communication complexity between nodes to O(n)(Fig.3),improves the power transaction speed,reduces the storage space required for signature data,and reduces network congestion.In addition,to prevent nothingat-stake attacks,the stacking economy is introduced to enhance the system security.

Fig.3 Consensus process of intrachain transaction based on BLS aggregate signature

2.2 Consensus mechanism of heterogeneous energy cross-chain transaction based on relay chain

The cross-chain mechanism is mainly divided into notary mechanisms,side/relay-chain,and Hash Lock.The notary mechanism is highly centralized,and its security depends on the notary’s credibility.The side/relay-chain mechanism mainly uses data of block headers for SPV,which consumes fewer resources.Consequently,it is impossible to verify global transactions,and the security of this mechanism is low.Hash locking is vulnerable to denialof-service attack (DoS)owing to its message dissemination mechanism.Considering that energy transactions should meet both efficiency and security,the three aforementioned mechanisms have shortcomings.To overcome these drawbacks,the cross-chain consensus algorithm for a heterogeneous energy system (HESC-CA)is proposed,which is implemented as follows:

(1)The main chain of the State Grid is responsible for the relay verification of the whole model; thus,the verifiers are very important.HESC-CA selects verifiers by default and through voting; enterprises can always reserve a certain number of verifier seats.The voting method for selecting verifiers is the same as that of HEST-CA,and hence has not been described again.

(2)The photovoltaic chain sends abundant electricity and energy-related information to the main chain of the State Grid through cross-links.The proof method shows that the photovoltaic chain submits the data of the block header to the main chain of the State Grid.In the Challenge period,if there is no acceptor submitting invalid block information in the photovoltaic chain,proceed to Step 3; otherwise,confiscate the tokens pledged by the cheating acceptor and reject the transaction application.

(3)The main chain of the State Grid verifies the acceptor signature data on the photovoltaic chain in the header of the block,and if the verification is correct,it gives the validity certificate,hashes the transaction data,and recursively operates until only one Merkle root is generated to ensure the security and traceability of the transaction data.

(4)The main chain of the State Grid sends the certification message and transaction information of the photovoltaic chain to the PHEV-chain.

(5)After the PHEV-chain receives the message sent by the main chain of the State Grid,the acceptor will vote.If the transaction is completed,the certification data and transaction information will be recorded in the chain and the message will be returned to the main chain of State Grid.

(6)The main chain of the State Grid uses the Merkle tree to record transaction data,writes it into the block header,and sends proof of data to PHEV-chain,such that the consensus of cross-chain power transaction is completed and the next consensus is started.

HESC-CA introduces the concept of the State Grid main chain,which is responsible for broadcasting news,verifying the effectiveness of cross-chain transactions of energy subchains,and cross-link with the latest blocks in energy subchains from time to time.The introduction of the main chain of the State Grid has the following advantages: 1)To improve the transaction throughput,the internal transactions of heterogeneous energy sub-chains do not affect each other and improve the scalability of the system,2)The communication cost and communication complexity of cross-chain transactions are reduced,and the main chain of the State Grid is used for relaying,replacing the energy sub-chain broadcast information in P2P networks,and the communication complexity is reduced from O(n2)to O(n),3)To improve the safety of energy sub-chain,the energy sub-chain always tends to the chain that has been crosslinked with the main chain of the State Grid,resulting in the attacker having to attack both the energy sub-chain and the main chain of the State Grid to succeed.

In this paper,cheating refers to the problem of data validity [29].To ensure the transaction performance,the relay chain cannot verify the block,it can only verify how many verifiers have signed a block in the submission,and then recognize the data validity of the block.For example,the PHEV chain is captured,and the malicious role generates invalid block B (assuming that block B added 100 tokens to Alice’s account).Next,the malicious nodes generate an effective block C (in other words,the transaction in Block C is valid)immediately after block B,to cover up the invalid block B; subsequently,it initiates a cross-chain transaction and sends the 100 tokens to Bob’s account on the PV chain.Henceforth,the 100 tokens created maliciously become completely legal and valid in the PV chain.To enhance the security of cross-link transactions and make up for the deficiency that the main chain of the State Grid cannot verify global transactions,this model also introduces the cross-chain router and Fisherman mechanism,which is shown in Fig.4.

Fig.4 Cross-chain transaction confirmation based on Fisherman

(1)Cross-chain routers responsible for bridging each energy sub-chain with the main chain of the State Grid,are used for initiating cross-chain transactions,transmitting messages of transactions,and their Merkle certificates.

(2)Fisherman mechanism: The core idea of Fisherman is: as long as the block header information (cross-chain transaction occurs)is transferred between cross blocks,a period of the challenge time interval is set for honest verifiers to provide invalid block proof.Several simple structures can prove that the block is invalid [30,31]; thus the communication cost of receiving this kind of proof is far less than that of receiving the whole block.The Fisherman mechanism makes the system safe as long as there is an honest acceptor in the energy sub-chain.This paper uses the following mechanism to prove the validity of block data.

Determine the fisherman nodes: the fisherman can be some nodes specified by the system or ordinary user nodes.When the fisherman’s report is correct,the fisherman will be motivated.

Fishermen detect invalid data state: the submission of the block requires the node to verify the whole block,which means that any node can access each transaction and keep a local state.Through the local copy of the state that has not been tampered with,the fishermen can independently calculate the Merkle root of the correct state and submit it to the relay chain.The relay chain only needs to compare whether the Merkle root Hash of the two is different.

Dealing with cheating behavior: using social consensus.Because the identity of the cheater of the energy blockchain is recognizable,the cheater can be punished by the law or other external legal system if necessary.

3 Simulation of the running process

In this model,the type of distributed energy determines the type of energy blockchain.Based on the type of distributed generator (DG),photovoltaic chain,PHEV chain,etc.,are defined.In the system,in addition to the conventional energy consumer and energy provider,there are prosumers (combined with professional producers and consumers).Energy users connected to these energy chains can be energy providers (companies responsible for providing electricity to communities),prosumers(households or individuals holding DG),and energy consumers.Taking the transaction process in the PHEV chain as an example,energy consumers still account for the majority in the PHEV chain.Thus,to meet the energy needs of PHEV users and profit pursuit,more energy providers will be connected to the PHEV chain.These energy providers can purchase energy from other heterogeneous energy chains (such as PV chains)through cross-chain technology and sell it to energy consumers in the PHEV chain.When the PHEV consumer does not need to use the PHEV for a period and the electricity on the PHEV is still sufficient,then he can sell the energy to other users.In the above process,energy purchase is divided into an intrachain transaction and a cross-chain transaction.

(1)Intrachain transaction (see Fig.5):

Fig.5 Flow chart of in-chain energy transaction based on BLS aggregated signature

1)Publish transaction information: Consumers pledge a certain number of tokens with smart contracts and connect them to the PHEV chain,which is to prevent nothing-atstake attack and reduce the number of the hard fork and false transaction requests in the PHEV chain.After pledging tokens,producers and consumers publish their information such as energy supply level and stability to the flexible electricity price intelligent contract.

2)System pricing: In this model,the pricing rules of electricity are written in the intelligent contract,and the electricity price is calculated according to the electricity supply and demand data submitted by all users to the intelligent contract of electricity price within a certain period,and the users subsequently decide whether to accept the pricing and trade.

3)Intention to reach a deal: PHEV chain broadcasts the electricity price information of users,matches the users intent on transacting according to certain rules,and achieves the preliminary transaction.

4)Verify the legality of the transaction: The speaker packages all transactions,and based on the BLS aggregate signature scheme in HEST-CA,the acceptor respectively verifies the legality of transactions and checks for problems such as double spending and false transaction data.Electricity trading is different from ordinary commodity trading and the transmission of electricity needs to be restricted by physical rules.When the transaction legitimacy is verified,it is necessary to check the security and judge the power flow of each branch of the PHEV.If the validity verification and the safety check are both okay,the electricity transaction is completed through the intelligent contract.If the check fails owing to the legitimacy of the transaction,the token pledged by the user will be confiscated.If the check fails owing to the transaction security problem,the transaction will be canceled and the users will wait to match the next transaction.

5)Upload data to the chain: Transaction settlement is performed,the hash value of the transaction data is calculated,Merkle root is updated,and then the transaction data are wound to complete one round of transactions.

(2)Cross-chain transactions (see Fig.6):

Fig.6 Flow chart of cross-chain energy transaction based on the relay-chain technology

1)Publish transaction information: The main chain of the State Grid selects the Acceptor in two ways: preselection and voting.When the producer and consumer in the photovoltaic chain need to trade with the producer and consumer in the PHEV chain,the latest block header data of the photovoltaic chain must be synchronized to the main chain of the State Grid.

2)System pricing: After receiving the block header of the photovoltaic chain,the main chain of the State Grid needs to wait for the challenge period to observe whether there is an invalid block in the photovoltaic chain.After the challenge period,the main chain of the State Grid verifies the signature data in the block header of the photovoltaic chain to judge the validity of the block header.Finally,the purchase and sale prices are calculated by the smart contract,and the user decides whether to trade or not.

3)Intention to reach a deal: The main chain of the State Grid broadcasts the electricity quantity and electricity price information,and the intelligent contract matches the users with transaction intention based on certain rules.

4)Verify the legality of transactions: According to the consensus process in HESC-CA,the main chain of the State Grid acts as a relay and provides validity proof,whereas the acceptor is responsible for verifying the legality of transactions,and the smart contract is responsible for security check.If they both pass these checks,the smart contract locks the assets through the hash lock and completes the transaction.

5)Upload data to the chain: Transaction settlement is performed,the hash value of the transaction data is calculated,the cross-chain Merkle root is updated,and then the transaction data are uploaded to complete one round of cross-chain transactions.

4 Case study

To verify the feasibility of the proposed mechanism,the Geth client was installed on a Ubuntu 19.04 system,and a private chain was built by using Truffle.The EI-HEB transaction intelligent contracts were written using Solidity and the intelligent contracts were interacted with by web3.js to simulate the cross-chain heterogeneous energy transaction scenario and conduct simulation tests.

The improved IEEE33 nodes power distribution system was adopted in the distribution network structure,and the system structure is shown in Fig.7.

Fig.7 IEEE 33 nodes power distribution system

The efficiency of the consensus algorithm is mainly measured by the block speed and the number of transactions per second (TPS).The block-out speed is affected by the communication efficiency.TPS is the transaction throughput.In the actual system,the faster the blockout speed and the larger the transaction throughput of the consensus algorithm,the higher the efficiency of the consensus algorithm.

Table 1 shows the participant information of prosumers.Other nodes are connected to ordinary power users,and each account has mortgaged one digital token in the system.

Table 1 Participant Information Form

Node Distributed power supply Electricity purchased/sold (kW·h)Type A Photovoltaic power generation 0/6.38 Prosumer B Photovoltaic power generation 3.82/0 Prosumer C Wind power generation 0/6.52 Prosumer D Wind power generation 3.56/0 Prosumer E PHEV 3.21/0 Prosumer F PHEV 0/2.63 Prosumer

4.1 Analysis of model communication efficiency

In the blockchain,the consensus mechanism determines how the nodes in the chain agree on specific data and influences the speed of transaction confirmation; this is the foundation and core of the whole system.In a consensus process,communication and data transmission are needed between participating nodes.Therefore,the communication efficiency between nodes will affect the consensus efficiency.The proposed algorithm is compared with the algorithm proposed in [12].In the PBFT-based and HESTCA consensus mechanisms,based on aggregate signature in this model,1-30 nodes are randomly selected from the IEEE33 nodes power distribution system,and the number of volumes required to reach a consensus,with the increase of nodes,is simulated.The simulation results are shown in Fig.8.

Fig.8 Diagram of communication efficiency changing with the number of nodes

The consensus efficiency of this model was analyzed.When the number of communication nodes is less than 10,the consensus efficiency of the algorithm of [12] was slightly higher than that of the proposed algorithm; however,with the increase of nodes in the system,its communication efficiency increases exponentially,and the consensus efficiency decreases,which results in numerous transactions not being confirmed in time.However,the communication volumes of HEST-CA show a continuous linear increase with the increase of nodes.Thus,when a large number of nodes need to communicate with each other,this model achieves higher consensus efficiency and faster transaction confirmation speed.

4.2 Analysis of model throughput

In the blockchain,transaction throughput refers to the number of transactions handled by the system per unit time,which is generally expressed by Transactions Per Second(TPS).The higher the throughput of a system,the stronger its ability to handle transactions.Thirty nodes are randomly selected from the IEEE33-node system,and 10 nodes are set as a group,in the PBFT-based single-chain architecture[6],the number of nodes is increased by 10 each time in groups,and three single chains are set up in the cross-chain transaction model based on relay-chain to enter the mainchain of the State Grid.New node groups are added with three single chains in turn.The nodes trade based on the energy demand and obtain the TPS data of a single chain.The simulation results are shown in Fig.9.

Fig.9 Diagram of throughput changing with number of nodes

The throughput of the model is analyzed.When the number of nodes is less than 10,there is a little difference between the single-chain architecture trading model based on PBFT and TPS of the multichain architecture model.However,with the increase in the number of nodes,they are all added to a chain in the single-chain architecture,which increases the communication load in the chain and decreases the throughput.In the HESC-CA algorithm,the newly added nodes are scattered in three chains,and the nodes in the same chain will trade first.If the energy supply and demand in the chain cannot be balanced,then call the relay-chain for a cross-chain transaction.It can be seen from Fig.9 that when the 26th node is added,the throughput of the model declines,the reasons are as follows: first,with the increase in the number of side chains,when the energy in the chain cannot meet the users’ supply and demand,the number of cross-chain transactions of the system will continue to grow,resulting in a significant increase in the communication pressure of the relay chain (verified side chain block information,responsible for cross-chain consensus communication),which cause a decrease in the transaction volume.Second,the bandwidth limitation of each energy sub-chain.With the increase in the number of transactions between nodes,the number of messages to be processed increases,and the message forwarding time lengthens.Simulation results show that the multichain architecture can disperse the communication pressure of nodes in the chain,increase the system throughput,and has good scalability; however,its maximum TPS is constrained by the performance of the relay chain.

4.3 Analysis of model security

The proposed model differs from the centralized transaction mechanism in terms of security [11],as shown in Table 2.The centralized power trading center needs to spend a lot of costs to build a database to ensure the security of data,and has a strong ability to resist external attacks;however,there is a risk that insiders may leak and tamper with data.The proposed model uses the blockchain to build the underlying system,which makes data tampering impossible.Using BLS aggregate signature technology,encryption and decryption operations can be completed without random numbers to avoid communication with random number prediction machine systems which enhances security.The pledge mechanism is introduced to bind the interests of users and the interests of the system,such that the proposed model can resist the nothing-as-stake results.Fisherman mechanism is introduced,which makes the relay chain to have a global verification ability,thus enhancing the security of the entire system.

Table 2 Participant Information Form Safety Analysis of each Model

Index Centralized trading mechanism Reference [15] Proposed Algorithm No interest attack Preventable Unpreventable Preventable Global block verification Available Unavailable Available Random number Not needed Needed Not needed Data tampering Probably Impossible Impossible Data traceability Unavailable Available Available

5 Conclusion

This model takes advantage of the blockchain in data storage and transaction,and a heterogeneous energy transaction model EI-HEB based on blockchain cross-chain technology is proposed,which improves the information transparency of heterogeneous energy transactions and realizes the cross-chain resource complementation of multienergy systems in the energy Internet.

(1)Through aggregate signature and relay chain technology,the consensus mechanism of power transaction is improved,the complexity of nodal communication is reduced,the consensus efficiency is improved,the number of transaction confirmation rounds is reduced,and the multichain architecture is introduced to enhance the scalability of the model.

(2)The introduction of the pledge mechanism can prevent unprofitable attacks,and the introduction of the fisherman mechanism makes the relay chain have global verification ability and improves the security of the entire system.

(3)Compared with the energy transaction system based on the single-chain architecture,the cross-chain system in this paper has a larger energy transaction throughput,which is very necessary for the energy transaction system with numerous nodes.

The smart contract deployed in the blockchain ensures the efficient execution of transactions in the multi-energy system.However,the research on the application of blockchain technology in the energy Internet is still in its infancy.For example,how to design a more efficient trading mechanism in the relay chain and enhance the performance of the relay chain needs to be further studied in combination with the actual situation.

Acknowledgements

This work is supported by the Fundamental Research Funds for the Central Universities of Ministry of Education(2018 ZD06).

Declaration of Competing Interest

We declare that we have no conflict of interest.