Research on key technologies of deduction of multinational power trading in the context of Global Energy Interconnection

1 Introduction

With the development of a domestic electric power trade simulation system [1],the development of a GEI [2,3] power trading simulation system considering the precision of the model,driver model of the data access,and external environmental factors (including politics and exchange rates) is considerably different in different countries.The existing studies [4] pertaining to the multinational longterm power trading quote considered the influence of the long-distance-transmission line loss,characteristics of both the parties involved,discounts,and political factors,among other factors.However,the matching based on the quoted price also involves a real financial settlement in the transnational power transaction.In such cases,the transnational settlement cycle is long,and the cost is high; consequently,the real-time intra-day power transaction among countries cannot be realized.To realize a rapid settlement by using digital currency,the electric power transaction deduction system must provide quotes using the digital currency.

The GEI multinational electricity traders are required to understand the associated energy information.Based on the changes in the real-time control of the global energy and information,the GEI framework provides a means to query the information; consequently,the cross-border power trading deduction system needs to obtain the change in the global energy information from the GEI to provide traders with a basis for the transaction.In addition,the multinational power trading deduction system also needs to optimize and match all the traders involved in the transnational power trading deduction and push the optimal quotation after the optimization and the optimal power generation of the various energy sources to the traders to assist in their decision-making.

In this study,based on the special drawing rights (SDRs) and the GEI data acquisition technology of the International Monetary Fund (IMF) and adopting the design framework of the microservices,a multinational power trading deduction system of the GEI is realized by considering the issues pertaining to the quotation settlement,acquisition of the transaction model data,and optimal transaction combination in the multinational power transaction deduction.

2 Establishment of a digital currency for cross-border electricity transactions

At present,compared with that of the other major energy cross-border transactions such as oil and natural gas,the overall scale of themultinational power trading is relatively small [5-9].Furthermore,the degree of marketization of power in various countries is considerably different [10-13] and the complexity and difficulty of the cross-border power transaction coordination are relatively high [14].In addition,the currency in which the cross-border electricity transactions are settled is influenced by the sensitive political issues that can generate a political influence similar to that of oil transactions.Chinese economist Xia Guo indicated that “in the future,the mainstream currency of human society will have the characteristics of digital currency,electronic currency,and global currency.The future world currency will surely be a globally consistent digital currency system” [15].With the increasing popularity of the digital currencies such as Bitcoin,which is based on the blockchain technology,the blockchain technology [16,17] has become a more widely used distributed accounting technology.

Energy coins (digital currencies) can be used in distributed books to record the behavior of multinational electricity transactions.At the time of the settlement,the energy coins in the multinational electricity transactions are equivalent to the SDRs.One dollar of an energy currency is equal to an SDR.The energy currency can be converted into a blue subcurrency (US dollar,Euro,RMB,Yen and Pound Sterling) and other global currencies.GEIDCO decides the number of energy coins to be issued according to the changes in the total energy generation and consumption worldwide in the next 10 years.Furthermore,this organization controls the number of energy coins purchased by the national power companies according to the demand for the energy generation in each country.

When conducting cross-border electricity transactions,each trading entity provides the quotes in the form of energy coins.Once the two trading entities form trading intentions,the consensus network completes the settlement of the transaction and the flow of energy.The framework of a transaction is shown in Fig.1.The trading bodies A and B are nodes in the consensus network.The consensus algorithm is the robust principal component analysis algorithm of the Ripple payment system,and the constructed alliance chain consensus mechanism helps realize an extremely fast transaction confirmation (completed within a few seconds ),while also ensuring that the entire consensus network has a strong extensibility and low consensus cost.

Fig.1 Energy currency transactions regulated by GEIDCO

3 Microservice-based transaction inference architecture

Under the global information technology development trend of shared services,many existing power systems are expected to use a micro application design or microservice design to provide the shared services [18,19].As a new subsystem of the power business,the GEI transnational transaction deduction system should also use the design route of microservices to provide services.

The traditional power market trading model is generally divided into two categories:centralized and decentralized trading.The GEI cross-border trading can thus adopt a decentralized trading model similar to that of the Nordic power market [20].Therefore,the multinational power transaction deduction system takes the calculations as the core,and the deduction process involves the input and output of the calculation data as well as the information interaction of the student simulation operation and the sharing function of the other power simulation services.When the multinational power transaction deduction system is microserviced,it is necessary to split the traditional power trading system and encapsulate it into multiple microservices such as a balance settlement and trend calculation.

As mentioned previously,the traders need to be aware of the dynamics of the world economy,as the GEI crossborder electricity trades are extrapolated.Most of these economic data are posted on the web pages of globally renowned websites,and thus the traders must regularly obtain the relevant data from these well-known websites.The multinational power transaction deduction architecture based on the microservices is shown in Fig.2.

4 Data acquisition of the multinational power trading calculation model

The calculation model of multinational power transactions requires the data regarding the change in the exchange rate of the countries,the wind and photovoltaic generation with meteorological changes,the changes in the electricity load with the meteorology,and other changes in the renewable energy and non-renewable energy aspects.The GEI thus involves a massive collection of data,which can be extracted from the GEI by using a vertical search engine [21,22].

Fig.2 Microservice-based transaction deduction architecture

4.1 Automatic data acquisition on a regular basis

The data published daily by many international organizations can be obtained on a regular basis because the data published daily are displayed on the web page in the same form; therefore,the JavaScript language within this web page can be analyzed.The data can then be extracted from such tables.For example,the IMF provides the daily corresponding values of the SDRs and the national currencies on its website in a tabular form.

Some official websites publish data in the form of multiple tables,which are related through the same columns.Therefore,this connection needs to be established when obtaining the data.For example,the meteorological data of the world’s cities,released by the World Meteorological Bureau on its official website,is related to the cities,and an ID number is assigned to each city.Therefore,one can first analyze the JavaScript language on the page to obtain the ID numbers for nearly 2,000 cities worldwide and then use the ID number of each city to transfer the weather page of the corresponding city to the website.

4.2 Automatically obtaining the data in the table

To enable users to customize the content on the web page to obtain the relevant data,the users need to interact with the browser.For this purpose,the Qt software is used to develop a browser to access the GEI web page data; this browser can remember the user interactions.

The page processing algorithm includes the preprocessing,user operation,automatic analysis,and other processes.This algorithm preprocesses the internal module structure of the database and displays all the user tables of the database and the database structure.

The user operation allows the user to establish a data mapping relationship between the data in the table on the web page and the data that is eventually stored in the database table,that is,a data block with several rows to several columns of the online table is stored in the corresponding column of the database table.According to the mapping relationship between the online table and the database table established by the user interaction,the online data is automatically extracted into the database.

5 Realization of the microservice of the transnational electric power transaction deduction

5.1 Realization of the microservices for the electricity transactions

The design architecture using microservices requires the construction of the power trading microservices.This service requires the load L of each node,number of units,quotation Pg,capacity Cg,reactance Xij of the branch (I,J),and maximum transmission active power PMij.Let the actual output of the decision variable set g be Ug.Its linear programming model can then be described as follows.

The optimal objective function is to purchase with the least cost:

The constraints include the energy balance,unit capacity,and branch capacity constraints,as described in equation (2)～(4),respectively:

The power transaction calculation program written in Python enters the input data required by the algorithm in the form of five two-dimensional vectors (L,Pg,Cg,X,Ug),providing the day,month,and year transaction decisionmaking functions.When the developed power transaction calculation program is upgraded to a microservice,the corresponding day,month,and year transaction decision microservice interface is provided.

When the user accesses the daily transaction policy processing interface for the power transactions to calculate the microservices,the user first writes the input data (L,Pg,Cg,X,Ug) of the daily transaction to the input data table.Because of the existing computational limitations,the daily trading algorithm performs simulation calculations for 24 hours in a day for one year or even several years.Several hours or more are required to complete these daily trading simulation calculations,which means that the caller must wait for a long time to obtain operational control.Therefore,this interface creates a new process to perform these daily transaction simulation calculations asynchronously.

Owing to the asynchronous operation of the MPTCMS in the new process,traders can only interact with the MPTC-MS asynchronously through the database.Therefore,the traders need to read the status record table in the database before they regularly read the calculation trend microservice.If the daily trading algorithm does not end,they call the calculation of the trend microservice and read the latest output value of the daily trading algorithm from the output table.This value is then used as the input of the trend calculation algorithm,and the calculated trend results are returned directly to the trader.

5.2 Power flow calculation and quotation conversion microservice

The branch capacity constraint needs to be determined by computing the microservice of the network DC current.In the case in which the energy (power) generated by the power generation module,energy (power) consumed by the load module,and reactance of each line are known,the grid DC power flow microservice can be used to calculate the flow of the energy between the node models.

Here,θi and θj denote the DC potentials at the two ends of the line,and Xij (or Bij for the thyristor) is the DC resistance of the line.Furthermore,

This aspect can be represented in the matrix form as

Here,B0 is the network node susceptance matrix during a normal operation; θ is the vector of the node voltage phase angle; and P is the active power vector injected by the node.

The quotation conversion microservice receives the quotation and currency of the transaction trader and obtains the conversion value of the currency used by the transaction trader to the energy coin by querying the database.Consequently,the quotation of the transaction trader can be converted into a new quotation with the energy currency as a unified unit.

6 Microservice synchronization and faulttolerant technology for transaction deduction

6.1 Synchronous control technology

In an object-oriented simulation development environment,the distributed simulation members synchronize by interacting with each other through timestamps or by sharing objects.In the distributed interactive simulation standard IEEE1516,a standardized solution algorithm based on the interactive timestamp has been proposed [23].The synchronization through the shared objects is typically applied to the computer supported cooperative work (CSCW).The cooperation mode of the CSCW is divided into time and space,and it can be divided into synchronization,distributed synchronization,asynchronous,and distributed asynchronous modes [24].

Because the ability of the microservices to communicate among each other is considerably lower than the communication ability between the processes,the exchange timestamps are used between the microservices [25].A large amount of code thus needs to be added to the microservices without being adopted in this article.The information exchange and synchronization control are realized by using the shared database table in the microservices.

In the process of the multinational electricity transaction deduction,the student needs to requote.In this case,the student sends the teacher a request to declare the price.The instructor publishes the requote request information by writing a simulation control form in the database.After reading the information from the simulation control table,the multinational power transaction computing microservice writes the current state to the simulation control table,pauses the transaction calculation process,and turns to the running state of the timed polling simulation control table.After reading the current state of the computing microservices for the cross-border electricity transactions from the simulation control table,the instructor distributes these states to the participants.The participants display this information on their own declaration interface,and after revising their requotations,the trainees send the instructor a new quotation.After the instructor receives the requotes of all the trainees,the simulation control form is established.After the multinational power transaction computing microservice reads this information from the simulation control table,it continues to run its calculation task.

6.2 Fault-tolerant technology

The microservice-running-fault-tolerance is a current research hotspot [26]; however,the MPTC-MS development also encounters the problem of the operational fault tolerance.Because the MPTC-MS creates a new process to asynchronously complete the daily transaction simulation calculations,many types of failures can occur during the implementation of the new process,such as a failure of the transaction trader program,failure of the instructor program,and failure of the MPTC-MS.

To tolerate these errors in a reasonable manner and avoid the occurrence of a wide-ranging impact,the transaction trader,instructors and corresponding process of the MPTCMS are merged into one deductive group,and each deduction of the group is numbered as the deduction number.Consequently,each deduction is determined by the deduction group number and the deduction number,and multiple deduction groups can be used for simultaneous deduction.

In a deductive group,the instructor is responsible for the initiation,suspension,and termination of the deduction,as well as running the synchronization and information exchange.These synchronized and exchanged information points are identified by the deduction team and deduction number.If the derivation numbers obtained by the members of the deduction group are inconsistent owing to the failure of the trader program,instructor program,or microservices in the multinational power transaction calculation,the entire deduction cannot be performed.

This derivation number requires the MPTC-MS to add one to the simulation data table at the end of the calculation.When the deduction number in the pair of the (deduction group,deduction number),which is read from the simulation control table,does not equal the deduction number of the MPTC-MS,and the control commands received by the caller are restarted,paused,and terminated,the MPTC-MS can adjust its own operating state.If the MPTC-MS fails,the corresponding fault identification is performed after the deduction number is added to the simulation data table before exiting,to ensure that the instructor program knows that the MPTC-MS has failed.

7 Extraction of the cross-border electricity transactions

The multinational electricity transaction deduction is first created by the instructor,and the participants subsequently join the deduction group.Upon confirmation that all the participants have joined,the instructor sends the participants the operating scenario of the transaction deduction,as shown in Fig.3.

After the instructor writes the input data of the annual daily trading (annual unit power declaration,daily unit price declaration,daily load electricity declaration,daily load price declaration,and annual price) into the input data table,the daily transaction policy processing interface of the multinational power transaction calculation microservice is called to start the calculation of the multinational power transaction microservice.After each one-hour transaction of the transnational power transaction calculation microservice,the output data of the linear programming are written into the database table.The instructor regularly queries the database table and sends the linear programming data read to the participants.

Fig.3 User interface of the internationalized power transaction deduction user interface

The students receive the data programmed linearly for themselves and display it.If they need to change,they apply for a rebid.The process for the rebid is shown in Fig.4.

Fig.4 Requote by the participants (standard requote)

During the rebidding process,the multinational power transaction deduction calculation microservice calls the quotation forecasting microservice and sends the predicted new quotation to the participants to guide the trainees to modify their quotations.

8 Conclusion

If the GEI partnership controls the amount of energy money issued,the use of energy coins in the multinational electricity transactions on the GEI can reduce the time required for the cash settlement to a few seconds.Furthermore,it also avoids the uncertainty of the crossborder transaction settlement caused by the changes in the exchange rate.To support the multiple deduction groups to perform the multinational power transaction deduction simultaneously,a microservice architecture must be employed.As the multinational power transaction computing microservice uses a linear programming algorithm to optimize the quotation and output power,the time to complete this service is extremely large,and it is necessary to use asynchronous methods to complete the calculation.Therefore,the problem of simultaneous communication and error tolerance between the traders and back-office services arises.The cooperative interaction technology based on the database table objects can accomplish the communication coordination and realization of the operation fault tolerance among the various microservices.

Acknowledgements

This work was supported by the State Grid Science and Technology Project (Research on Transnational Energy Interaction Simulation and Deduction Technologies of the Global Energy Interconnection,JS71-17-004).