Hindawi

Scientific Programming
Volume 2021, Article ID 1198794, 11 pages
https://doi.org/10.1155/2021/1198794

Research Article
An Intelligent Scheduling Access Privacy Protection Model of
Electric Vehicle Based on 5G-V2X

Cheng Xu ,1 Hongjun Wu ,1 Hongzhe Liu ,1 Xuewei Li,1 Li Liu,2 and Pengfei Wang3
1
Beijing Key Laboratory of Information Service Engineering, Beijing Union University, Beijing, China
2
Information Center, Yunnan Power Grid Co., Ltd., Kunming, China
3
Communication and Information Center of Ministry of Emergency Management of the People’s Republic of China,
Beijing, China

Correspondence should be addressed to Hongzhe Liu; [email protected]

Received 20 August 2021; Revised 11 October 2021; Accepted 18 October 2021; Published 9 November 2021

Academic Editor: Punit Gupta

Copyright © 2021 Cheng Xu et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

It is more and more important to optimize electric power system scheduling in the development of the Internet of Vehicles. How
to improve the applicability and scientific nature of electric vehicle charging is an urgent problem to be solved. This paper proposes
an intelligent scheduling access model for electric vehicles based on blockchain. Firstly, the layout simplification calculation is
carried out for the layout of the traditional distributed power grid. Then, a data storage and consensus system is built using
blockchain smart contracts to ensure that all historical data are not tampered with and are traceable. Finally, the model forms an
electricity price guidance model in the intelligent scheduling and access of electric vehicles, optimizes the multivehicle line
congestion in operation, and can dynamically schedule and correct the model. In terms of the experiment, 13 test electric vehicles
were dispatched based on 12 real power station nodes and 36 test nodes of Yunnan Power Grid Co. Information Center for
verification. The result analysis shows that the model can effectively and quickly solve the blocking and unblocking of the Internet
of Vehicles and can develop a scheduling scheme conforming to the safety constraints of electric vehicles in a relatively short time.

1. Introduction stations, and difficulty in charging transaction settlement

[2, 3]. For example, the introduction of a central organi-
Electric vehicles (EVs) replacing traditional fuel cars are of zation to manage charging services and charging transaction
great significance in alleviating urban air pollution. Com- information can improve the current situation, but it re-
pared with traditional fuel cars, electric vehicles are ex- quires high cost and risks of information leakage [4].
pensive, take longer to charge, and have a shorter range. In Therefore, we need to seek new charging service mode and
particular, the lack of charging piles greatly affected the transaction information management system. With the
promotion of electric vehicles. All countries are committed increasing level of digitalization and decentralization of
to the research and development of new energy vehicle key energy systems, the need for reliable defense against
technologies, but the penetration rate is not high [1]. In the cyberattacks in the IoV has increased dramatically. If
future, the proportion of electric vehicles on the Internet of properly applied, blockchain technology can fix vulnera-
Vehicles (IoV) or Internet of things (IoT) will be increas- bilities in networking and data communication. It can
ingly large, because of its great advantages; for example, improve data confidentiality and privacy, so as to effectively
logistics enterprises can provide a fixed charging station. defend against various threats from the inside and outside
Compared with fuel cars, electric trucks are environmentally [5]. In terms of data communication vulnerability repair,
friendly and cost-saving. At present, EV charging service has blockchain technology can integrate the pricing and set-
problems such as difficulty in finding a pile for charging tlement functions of the wholesale power market into a
users, difficulty in information connection between charging secure blockchain system. It can greatly reduce the risk of
2 Scientific Programming

fake data injection and price manipulation. It ensures that clearing and auditing efficiency. Alliance chain, power plant,
the dataset or communication series specified between the power grid, electricity selling company, and other subjects
trading parties is immutable. At the same time, blockchain are adopted among the trading subjects in the power market
can improve data confidentiality and privacy. As digitally to reach contract agreements, improve efficiency, and fa-
distributed systems become the norm in the energy sector cilitate inquiry and statistics. The public link is adopted
and billions of energy-using devices are integrated into between the electricity selling company and users, and
power systems, malicious attackers will see more opportu- automatic meter reading, metering, and billing are adopted
nities to break into these systems and attempt to disrupt to improve transparency and credibility [11].
them [6]. Blockchain has a natural network threat defense capa-
Due to the wide network coverage, high data informa- bility, which benefits from the following characteristics
tion transmission rate, low end-to-end latency, and support [12, 13]:
for massive connections, 5G networks are rapidly spreading
(1) Tamper-proof data: after proper application of
[7]. As the underlying network communication technology,
blockchain, all calculations will be performed in hash
5G can deeply empower upper-layer vertical applications.
form and will be tamper-proof when the data are
The integration of 5G and blockchain can cause them to
generated, thus avoiding the risk of being tampered
empower each other. With its high speed and low latency,
with in transmission.
5G helps the blockchain accelerate transactions (full network
broadcasting), avoiding stalls and long-term unrespon- (2) Complete data availability: blockchain can store data
siveness [8]. Moreover, because 5G drives V2X, it has in a decentralized form through multiple nodes.
spawned a lot of data and scenarios, which is conducive to Under such an architecture, even if some nodes or
the implementation of blockchain applications. servers are hacked, the user can still get the complete
In this context, how to improve the applicability and dataset.
scientific nature of electric vehicle charging is an urgent (3) Redundancy: the operation of blockchain has no
problem to be solved. In order to improve the scheduling central failure point, so this architecture is inherently
optimization of EV in the development of IoV in the real highly reliable through redundancy.
environment, this paper proposes an intelligent scheduling (4) Privacy and control: blockchain users can choose
access model for EV based on blockchain. which data are permanently transparent and which
The rest of this paper is organized as follows. Section 2 data are encrypted, so only the designated receiver
explores the latest methods about blockchain, IoV, and can view the data content.
intelligent contract. Section 3 presents the access model of
(5) Outsourcing calculation: the encrypted data can be
intelligent scheduling of EV based on blockchain. Com-
sent to a third party for processing, and the data
parisons of experimental results with other works and
content will not be displayed in the whole process.
discussion are provided in Section 4. Section 5 summarizes
the paper and offers directions for future research.
2.2. IoV in the Grid Intelligent Scheduling Algorithm.
2. Related Work Vehicle scheduling problem was first proposed by Dantzig
and Ramser [2]. The goal is under certain constraints, such
2.1. Blockchain Security in Power Grid Industry. The de- as the limitation of vehicle load and the volume of the goods
centralization, openness, intelligence, and sharing of to seek the best assembly, also called the vehicle scheduling
blockchain are in line with the ideas of the IoV and the problem (VSP), or seek vehicles’ best path, also called the
energy Internet. They can be done by means of data en- vehicle routing problem (VRP), which makes the total
cryption, timestamps, distributed consensus, and economic distribution cost minimum. Therefore, the vehicle sched-
incentives. Point-to-point transaction, coordination, and uling problem is a typical nondeterministic polynomial
collaboration can be realized based on decentralized credit in problem. When the electric vehicle is used for trans-
a distributed system where nodes do not need to trust each portation, the problem is more difficult to solve due to the
other [9]. The application of blockchain in IoV will effec- influence of charging and endurance. In recent years, the
tively support the open interconnection of multitype systems logistics scheduling problem based on new energy vehicles is
and the extensive and deep participation of multiple users. It collectively known as the green vehicle routing problem
can not only solve network access, long-distance trans- (GVRP). As GVRP is relatively new, there is not much
mission loss, demand-side response, network security, and classical research literature. Literature [6] summarizes the
other problems, but also form some new models for the research status of EVRP before 2014. It summarizes the
inherent market mechanism and business model of the technical background of electric vehicles, including vehicle
power industry [10]. type and battery, cost comparison with traditional fuel car,
At the same time, in terms of power market transactions, vehicle grouping, route selection, and optimal route, and
blockchain can exist in three decentralized forms with gives possible research directions.
different degrees of the public chain, private chain, and In the past two years, representative research results
alliance chain. The internal finance of power grid adopts a included the following: vehicle scheduling problems under
private chain, which is conducive to the transparency of the constraints of time and charging station are widely
internal management universities and the improvement of concerned [9–11]. Literature [9] adopted a new hybrid
Scientific Programming 3

heuristic algorithm to solve this problem. The new algorithm nodes are set up to manage the choice and transaction of
combined variable domain search and tabu search algorithm user charging mode, but the implementation structure of a
to test on the standard set and achieved good results. Lit- specific alliance chain is not mentioned. In [16], electric
erature [10, 11] studied the hybrid formation problem of vehicle users and distributed energy operators were set up
different types of electric vehicles. In the model, the im- for the application scenario of microgrid. Literature [17]
proved branch-and-price algorithm based on Adaptive used an alliance chain to design charging and discharging
Large Neighborhood Search was adopted to solve the transactions between electric vehicles and set up local
problem with the constraints of transportation time, aggregators to act as service nodes, without involving user
transportation cost, vehicle load, etc. The experiment transactions in different aggregators.
showed that the algorithm was effective. Moreover, literature The sorting and analysis of the above research status
[12] studies the different trolley charging conditions under show that blockchain can effectively improve the defects of
different scheduling paths under time constraints and uses grid security access and optimize the charging model of
branch-price-and-cut algorithms to determine whether electric vehicles.
there are multiple charging stations on each scheduling path
and whether each charge must be filled to get 4 combina-
tions. The study shows that in the case of 100 customers and 3. Intelligent Scheduling Access Model of EV
21 charging stations, all four combinations have solutions.
3.1. Electric Vehicle Charging Security Access Trading Design.
The comparison of the four kinds of path planning reveals
EV charging trading system architecture can be divided into
that multiple charging piles have more advantages than the
application layer, intelligent contract layer, consensus layer,
other three cases without full charging. Literature [13]
network layer, and data layer [18]. The data layer stores the
proposes a universal EVRP model, which takes minimum
transaction data in LevelDB or CouchDB in the data
transportation time, minimum energy consumption, and
structure of Merkle Bucket tree and blockchain list [19]. The
minimum transportation vehicle as constraints and passes
network layer selects HTTP/2-based P2P protocol as the
through each customer only once. Different from previous
network transmission protocol and makes the node listen to
models, the problem of power consumption under different
verify whether the new block or new transaction of the
loads is considered in this modeling. In this model, the
broadcast is valid. The consensus layer adopts a practical
feasible solution space is not large, so the computer is used to
Byzantine fault-tolerant algorithm. The intelligent contract
solve it accurately. However, the above work did not solve
layer encapsulates the electric vehicle charging trading in-
the problem of electricity price optimization.
telligent contract written in Go language. The application
layer supports EV charging applications with query charging
2.3. Intelligent Contract for Secure Access. As a new appli- stations and transfer transaction functions [20–24].
cation technology of the new form of IoV, blockchain has the Figure 1 shows the network structure of EV charging
characteristics of weak centralization, distrust, and not being transaction model. Participants include application, charg-
tampered with easily. It can realize the undifferentiated ing station nodes of charging operators and utility com-
records of multiple nodes and promote the interconnection panies, sorting service nodes, and regulatory authorities of
of information [5, 6]. Intelligent contracts can be widely used charging transaction market.
in various interconnection scenarios of energy transactions Each charging operator or utility serves as an organi-
such as power market transactions and energy demand zation in the transaction network structure, with multiple
response [9, 10]. Literature [11] proposes to use blockchain charging station nodes per organization. All nodes are ac-
technology to realize transparent and trust-based charging counting nodes, responsible for validating transactions and
of public charging piles. Literature [12] proposed Ethereum writing them into the books. On this basis, a node can also
EV charging transaction mode, in which EV users choose the act as master node, anchor node, and endorsement node.
best bidding charging station. Literature [13] proposed a The master node is the node responsible for communicating
charging pile sharing platform based on a lightning network, with the sorting service node, which receives the transaction
blockchain, and intelligent contract. Literature [14] pro- containing the endorsement signature. It sorts unpackaged
posed electric vehicle charging transaction based on a transactions, generates blocks, and broadcasts them to the
lightning network, which solved the possible security billing node. Anchor nodes are nodes that can communicate
problems in a lightning network. with the nodes of other organizations. By running its in-
The above studies generally use the traditional bitcoin or stalled smart contract, the endorsement node can sign and
Ethereum blockchain architecture, which requires tokens endorse the charging transaction proposal put forward by
and public chain, which is not conducive to improving the client and feed back the result. The transaction model
transaction efficiency and reducing transaction cost. calculates the endorsement experience value M according to
Therefore, it is necessary to select a blockchain architecture the endorsement times of the nodes and refers to this index
more suitable for the performance requirements of EV in service evaluation, so as to encourage each charging
charging transaction business. Many scholars use the weakly station node to act as the endorsement node and maintain
centralized alliance chain to improve the efficiency and the stable operation of the blockchain platform. The ex-
security of EV transaction. In literature [15], multiple agent pression of endorsement experience value M is
4 Scientific Programming

Charging carrier Charging carrier node

identity The The
authentication endorsement accounting
node node
Electric car trading The master The anchor
book information node node
Public power
Public power supply
supply enterprise
enterprise node Electric vehicle trading market
identity
certification The The supervision department
endorsement accounting
node node Aisle Charging Public power
Electric car carrier supply
trading book identity enterprise
The master The anchor identity
information node node information information

Sort service
Application node

Figure 1: Network structure of EV charging transaction model.

lc,n subject under the consensus mechanism. Firstly, EV users

M�β , (1) select charging stations according to service rating and use
Ln
the charging clients to send a request to the endorsement
where lc,n is the number of account updates of the charging node in the channel and invoke charging service operation.
station node as the endorsement node up to this transaction. After receiving the request message, the endorsement node
Ln is the total number of ledger updates. β is the weight of the verifies the client’s identity, performs intelligent contract
adjusted endorsement experience value. simulation transactions, and evaluates the charging service.
Data exchange is realized by establishing a trading The endorsement node outputs the result of the intelligent
channel, which is a virtual channel of atomic broadcasting. contract, that is, a set of key values read or written in the
The design of the trading channel makes it impossible for the intelligent contract. The transaction request response with
nodes outside the channel to access the data inside the the endorsement signature is sent back to the client. The
channel, which provides safe and efficient data exchange. client broadcasts it to the sorting service node. The sorting
The channel is managed by the regulatory department of the service node passes ordered transactions as blocks to all
EV charging trading market and connects charging station nodes on the channel. Each node verifies that the execution
nodes and sequencing service nodes of multiple companies. result can be written to the ledger status database and no-
Users who register with any charging operator or utility can tifies the client.
have the account authentication of the trading platform and
deposit their wallet address, account balance, current
credibility, certificate, and public/private key pair. Through 3.3. Smart Contract for EV Charging Trading Model.
the application, users can access all endorsement nodes in Smart contracts are programs that run on a ledger of
the channel and select charging services provided by mul- blockchain data and are automatically executed by com-
tiple charging operators and utility companies. puters [27]. The main functions of the intelligent contract for
EV charging transactions include EV charging charge set-
tlement, charging service, user credit evaluation, and
3.2. The Process of Trading Electric Cars in IoV. The opera- transaction order query from users, charging operators,
tional process of EV charging transaction model is based on utility companies, and other institutions.
practical Byzantine fault tolerance (PBFT) algorithm Every time the owner completes the transaction settle-
[25, 26]. PBFT is a copy replication algorithm for state ment of electric vehicle, the credibility of the owner is
machines that replicate on different nodes. The client sends a evaluated, Rv , according to it transaction behavior, and the
request to the master node to invoke the service operation, formula is
and upon receiving the request information, the master node
enters the preparatory stage to broadcast to other nodes. 1 J
Rv � 􏽘C , (2)
After receiving the broadcast, other nodes simulate the J j�0 vj
transaction, generate the transaction result, generate the
hash value of the new block according to the result infor- where Cvj is the integral reward that EV account v obtains or
mation, and enter the preparation stage to carry out the deducts after the j-th charging transaction is completed, and
broadcast within its scope. its value is set. Let its value meet the set Cvj ∈ {−1, +1}. When
As shown in Figure 2, the operation of EV charging the user completes the correct payment of the previous
trading mode is the process of transaction initiation, con- charging service, it can get a bonus point. Otherwise, it can
tract execution, and transaction verification by the trading deduct a credit point. If the credit is lower than the
Scientific Programming 5

Electric vehicle
Electric car user
trading network
1 Price increase
transaction application 5 Book write situation
+ Identity Information Charging
Charging 2 Authentication + station Smart
transaction Gateway Transaction contract
node
application endorsement
6 Transaction
result response 3 Request sort
4 Send an orderly
transaction

Sort service node

Figure 2: Operation process of EV charging transaction model.

minimum threshold set by the system, that is, Rv < Rv, min , reasonable. Moreover, they can stimulate charging station
the user will not be able to use charging service normally. In nodes to improve charging service level, contribute com-
this model, the initial value of credit and the threshold value puting power, and maintain stable and efficient operation of
are 0. the trading platform. A charging station is set up to provide
The transaction model recommends charging stations charging service for EV. The charges that EV users need to
with high evaluation of nearby charging services to users pay to charging operators and utility power companies
with normal credibility for charging transactions. The include electricity fee and service fee, and the formula is
charging service evaluation Rcs,n of the charging station node
Fcharging(v) � fcharging(v) Ee − Es 􏼁,
is constantly updated with the increase of n times of (4)
transactions. The following equation calculates the charging Fsercice(v) � fservice(v) Ee − Es 􏼁,
service evaluation of the charging station according to the
score given by users with different credibility and the en- where Es is the initial electric quantity of the electric vehicle.
dorsement experience value of the charging station node. Ee is the power after charging. fcharging(v) is the unit price of
electricity charge. fservice(v) is the service fee unit price. The
⎪
⎧ 1 total cost of charging an electric car v is
⎪
⎪ Rcs,n � Rcs,n−1 + (1 + M)ϕ􏼐Rcs,n−1 􏼑Rv Wn − En 􏼁,
⎪
⎪ α
⎪
⎪ Fv � Fcharging(v) + Fservice(v) � fcharging(v) Pv t + fservice(v) Pv t,
⎪
⎪
⎪
⎪
⎪
⎨ 1 (5)
⎪ ϕ􏼐Rcs,n−1 􏼑 � 1 − ,
⎪ −(Rcs,n−1 − D)/σ
⎪
⎪ 1+e where Pv is the charging power. t is the charging time.
⎪
⎪ The specific steps of the transaction fee settlement
⎪
⎪
⎪
⎪ function are as follows. The first step is to receive the pa-
⎪
⎪ R
⎩ En � cs,n−1 , rameters including charging amount Echarging(v) ; The second
D
(3) step is to apply the API of querying the ledger to get the
balance of the charging station and EV account, as well as the
where Rcs,n−1 represents the n − 1 transaction service eval- unit electricity price corresponding to the period number.
uation of charging station node, and the adjustment pa- The third step is to complete the calculation of the charging
rameter α > 1 determines the speed of change of service transaction amount payment function according to the
evaluation after each transaction score. The value of α can be formula and transfer the corresponding amount in the EV
adjusted so that after the service capacity of charging station account to the charging station account. If the balance of the
nodes with low evaluation is improved, the evaluation will electric vehicle account is lower than the threshold value, the
not always be affected by the previous bad evaluation. M is credit points will be deducted and the bonus points will be
the calculated endorsement experience value. Wn is the score increased. The fourth step is to write deals into the blocks.
made by users whose credit rating is higher than the Administrators of electric vehicle users and charging
minimum threshold, and in this paper Wn � {1, 2, 3, 4, 5}. operators can query various data of charging bills through
The En is the expected score of the charging station node. intelligent contracts, taking the function of bill inquiry by
The D is the highest level in service evaluation, and this users as an example. The first step is to receive the EV
model takes D � 6. The ϕ(Rcs,n−1 ) is the damping function number (EVID) from the client’s SDK. The second step
that makes the change of charging service evaluation value constructs the query string based on the EVID, sends the
tend to be gentle. The σ is the acceleration factor in the query string proposal to the ledger, and retrieves the query
damping function, and σ � 0.7. result. Query results include transaction time, charging cost,
In the calculation of charging service evaluation, the unit electricity price, charging station number, organization
credibility of EV users and endorsement of charging station number of the user, and organization number of the
nodes are considered to make the evaluation value more charging station. The third step is to aggregate all query
6 Scientific Programming

results and return them. Unlike other functions that need to two active distribution networks, ADN1 and ADN2.
be written into the ledger, such as invoking the transaction According to the line and geographical location, the ADN1
ledger function, the node runs the intelligent contract to includes 1 wind turbine (W2), 2 tram stations (EV2 and
access the status data and directly outputs the running re- EV3), 1 set (DG3), and 1 load node (L3). The ADN2 consists
sults to the client without consensus. of 1 wind turbine (W1), 1 tram station (EV1), 2 units (DG1
and DG2), and 1 load node (L2). The unit information is
4. Experiment and Analysis shown in Table 1, where a, b, and c are the constant, primary,
and secondary coefficients of the generation cost, respec-
4.1. The Experiment Platform. In terms of the experiment, tively. The Pmin and Pmax are the minima and maxima of
the hardware platform includes 13 unmanned logistics ve- output, respectively. The optimal regulation of generating set
hicles, as shown in Figure 3, 12 charging nodes, laptop is shown in Table 2.
computers, and servers. The software platform includes 36
test nodes and blockchain based on alliance chain.
4.3. The Experimental Analysis. In the experiment, a
common algorithm without model optimization is firstly
4.2. The Experiment Design. The experiment evaluates the
used to conduct vehicle scheduling and charging as a
computational efficiency of the model by using performance
comparison item. After that, the distributed optimiza-
indicators of data throughput, response time, and latency. As
tion algorithm and the dynamic adjustment model based
shown in Figure 4, under the same node, when using the
on blockchain were adopted to conduct V2G optimi-
model in this article, it has better throughput and response
zation of the power grid, distribution network, and ve-
time than the traditional model. Compared with the tra-
hicle charging schedule; meanwhile, the dynamic
ditional model, the proposed model has more data
constraint counting method was adopted for blocking
throughput, shorter response time, and lower delay time
management.
under the same node. As shown in Figure 5, with the same
In the scheduling process, as shown in Table 3, power
number of vehicles, the number of nodes is gradually in-
flow overload occurs in some nodes such as 15, 16, and 17.
creased, and the performance of the model in this paper is
By adding the 3 modulation constraints into the constraint
better than the traditional model.
set, the traditional algorithm presents large information
Under large-scale scheduling, the ratio of the ware-
blocking, which makes the vehicle scheduling appear
house’s storage capacity to the vehicle’s transportation ca-
sluggish, and the charging efficiency is greatly reduced. After
pacity is an important index affecting scheduling. In this
that, the proposed model was used for optimization. After
article, it is called the α coefficient, which is
the optimization, there was no longer the case of charging
􏽐ni�1 Wi overload in statistics, so the results were in line with the
α� , (6)
􏽐mj�1 Vj
optimal solution of security constraints. In the test, the
traditional algorithm presented information blocking. Un-
where Wi stands for storage capacity and Vi stands for der the optimization of the proposed model, the interference
vehicle transportation capacity. is quickly eliminated, and the charging of multiple vehicles is
In this experiment, 13 vehicles are dispatched using 36 not disturbed. Moreover, under the optimization model, the
nodes. The proposed algorithm and the traditional sched- electricity price of charging is lower than that without
uling algorithm are run 150 times each with different optimization.
number of vehicles, and the average of the running results As shown in Figure 8, when blockchain is used in
was applied for the comparison. As shown in Figure 6, under scheduling and charging transactions, the delay is greatly
the same number of vehicles, the model in this paper has reduced. The increase of information throughput
better data than traditional algorithms, which increases strengthens the optimization effect of the model and ach-
vehicle scheduling efficiency. ieves a good experimental result.
In order to test the safety of this model, the illegal vehicle The running time and cost after optimization are less
ID packet was synthesized with the correct vehicle ID packet than those without model optimization. The optimized
to try to link the charging device. When a blockchain-based optimal power flow distribution and the data are better
charging device identifies an illegal data packet, it will give than the data obtained without model optimization. It
alarm and detect the average alarm time. As shown in can be seen from the above results that when vehicle
Figure 7, the conventional model does not identify the illegal scheduling is carried out by using the proposed model,
vehicle ID data packet several times; thus, the charging the charging efficiency of the vehicle is higher, the cost is
device starts to work. When using this experimental model, lower, there is no information blocking, and it is safer and
all illegal vehicle ID data packets are detected and an alarm is more efficient. The optimal scheduling scheme is devel-
issued successfully. As shown in Figure 7, the detection time oped in a short time to make the vehicle running and
was significantly lower than the traditional model. charging more reasonable. The proposed model enables
The transmission network includes load L1 and two large logistics enterprises to make better profits and achieve
units G1 and G2. The distribution network is divided into higher logistics efficiency.
Scientific Programming 7

Figure 3: Unmanned logistics vehicles.

200 1.4
180
1.2
160
140 1

Response time (s)

Throughput (kg)

120 0.8
100
0.6
80
60 0.4
40
0.2
20
0 0
6 12 18 24 30 36 6 12 18 24 30 36
number of nodes number of nodes
AI algorithm AI algorithm
Manual algorithm Manual algorithm
(a) (b)

Figure 4: Throughput (a) and response time (b) for the same number of nodes.

70 1.2

60 1
50
Response time (s)
Throughput (kg)

0.8
40
0.6
30
0.4
20

10 0.2

0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13
number of vehicles number of vehicles
AI algorithm AI algorithm
Manual algorithm Manual algorithm
(a) (b)

Figure 5: Throughput (a) and response time (b) for the same number of vehicles.
8 Scientific Programming

18
16

scheduling evaluation index

14
12
10
8
6
4
2
0
1 2 3 4 5 6 7 8 9 10 11 12 13
number of vehicles
AI algorithm
Manual algorithm
Figure 6: Scheduling evaluation index varying with number of vehicles.

25 0.6

20 0.5
Average alarm time
Number of attacks

0.4
15
0.3
10
0.2
5 0.1

0 0
2 4 6 8 10 12 14 16 18 20 2 4 6 8 10 12 14 16 18 20
Number of alarms Number of alarms
AI algorithm AI algorithm
Manual algorithm Manual algorithm
(a) (b)

Figure 7: Number of alarms (a) and average alarm time (b) with the same vehicle number.

Table 1: Generator information.

Pmin/MW Pmax/MW a/yuan b/yuan ∗ MW^−1 c/yuan ∗ MW^−2
G1 20 160 450 28 15
G2 10 130 330 20 25
DG1 0 45 450 23 15
DG2 0 30 360 29 20
DG3 0 45 360 17 15

Table 2: Optimal generating.

Time DG1/MW DG2/MW DG3/MW G1/MW G2/MW
1 25.6 29.8 31.2 42.1 34.5
2 24.1 29.1 30.8 40.2 33.2
3 23.3 29.5 29.5 40.3 32.4
4 22.7 29.9 29.6 37.4 33.5
5 22.4 30.2 28.3 37.1 30.7
6 23.8 30.0 28.2 38.7 32.3
7 23.5 30.0 28.6 36.9 30.8
8 24.3 30.3 27.7 40.2 32.5
9 23.4 28.9 29.8 38.5 31.1
10 20.1 29.1 28.5 33.1 27.8
Scientific Programming 9

Table 2: Continued.
Time DG1/MW DG2/MW DG3/MW G1/MW G2/MW
11 25.8 29.0 24.1 41.4 34.3
12 29.1 28.7 30.1 41.8 38.9
13 30.1 30.1 35.5 47.5 44.8
14 20.8 30.2 36.4 51.5 37.9
15 31.1 30.2 35.3 45.2 39.8
16 20.9 30.1 26.5 44.3 38.1
17 23.8 29.7 29.3 50.9 45.8
18 30.1 29.8 28.5 57.4 39.2
19 30.4 30.2 36.3 50.3 40.2
20 29.6 30.2 40.5 43.6 45.6
21 30.1 30.0 37.8 45.3 40.1
22 30.8 29.8 37.5 49.6 40.3
23 30.1 30.0 35.2 39.3 39.8
24 23.5 30.1 29.0 40.1 35.6

Table 3: Branch information.

Branch number Outflow node Injection node Current limit/MW
1 1 3 60
2 1 10 60
3 2 21 80
4 3 26 70
5 4 25 65
6 5 26 65
7 3 34 65
8 11 30 65
9 12 31 65
10 16 35 65
11 2 22 70
12 6 21 70
13 6 20 70
14 7 9 70
15 9 8 70
16 13 6 70
17 15 2 70
18 19 1 60
19 23 23 60
20 22 28 60
21 28 27 60
22 26 31 80
23 26 31 80
24 29 34 75
25 34 35 75
26 33 25 65
27 36 22 65
28 30 10 65
29 16 9 65
30 18 9 65
31 11 7 60
32 34 5 60
33 30 2 60
34 31 17 60
35 29 15 60
36 28 11 60
10 Scientific Programming

Main distribution network total output (MW)

25

20

15

10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
time (h)
Scene 1 output
Scene 2 output
Figure 8: Sum generating of distribution and transmission grid.

5. Conclusion KM201911417001), the National Natural Science Founda-

tion of China (Grant nos. 62102033, 61871039, 62171042,
In view of the fact that the current electric vehicle dis- 61906017, and 61802019), the Collaborative Innovation
patching is mostly manual dispatching and its efficiency is Center for Visual Intelligence (Grant no. CYXC2011), and
low, we investigate how to improve the applicability and the Academic Research Projects of Beijing Union University
scientific problem of electric vehicle charging in the existing (nos. ZB10202003, ZK40202101, and ZK120202104).
distributed grid layout. Based on blockchain and 5G tech-
nologies, the proposed intelligent vehicle dispatching model
realizes the simplified calculation of the layout of the tra- References
ditional distributed power grid. It uses smart contracts to [1] K. Kaur, G. Kaddoum, and S. Zeadally, “Blockchain-based
build data storage and consensus systems to ensure that all cyber-physical security for electrical vehicle aided smart grid
historical data are not tampered with and is traceable. It ecosystem,” IEEE Transactions on Intelligent Transportation
forms the electricity price guidance model in the electric Systems, vol. 22, no. 8, pp. 5178–5189, 2021.
vehicle intelligent dispatching and access, optimizes the [2] R. Elshaer and H. Awad, “A taxonomic review of meta-
multivehicle line blocking in the operation, and dynamically heuristic algorithms for solving the vehicle routing problem
adjusts the model scheduling. It has achieved good results in and its variants,” Computers & Industrial Engineering,
application, such as optimizing charging running time, vol. 140, Article ID 106242, 2020.
saving power, and controlling cost. The simulation and [3] S. S. Fazeli, S. Venkatachalam, R. B. Chinnam, and A. Murat,
actual test show that the model is feasible and correct. “Two-stage stochastic choice modeling approach for electric
Compared with the traditional scheduling algorithm, the vehicle charging station network design in urban commu-
nities,” IEEE Transactions on Intelligent Transportation Sys-
proposed intelligent vehicle dispatching model realizes the
tems, vol. 22, no. 5, pp. 3038–3053, 2020.
dynamic adjustment and correction and better optimizes the [4] B. Wang, D. Zhao, P. Dehghanian, Y. Tian, and T. Hong,
distributed power system to make up for the current electric “Aggregated electric vehicle load modeling in large-scale
vehicle. The future research will discuss the blockchain in electric power systems,” IEEE Transactions on Industry Ap-
more power nodes and multivehicle collaborative optimi- plications, vol. 56, no. 5, pp. 5796–5810, 2020.
zation problems. [5] I. Makhdoom, I. Zhou, M. Abolhasan, J. Lipman, and W. Ni,
“PrivySharing: a blockchain-based framework for privacy-
Data Availability preserving and secure data sharing in smart cities,” Computers
& Security, vol. 88, Article ID 101653, 2020.
The datasets used and/or analyzed during the current study [6] G. Kumar, R. Saha, M. K. Rai, R. Thomas, and T.-H. Kim,
are available from the corresponding author on reasonable “Proof-of-Work consensus approach in blockchain technol-
request. ogy for cloud and fog computing using maximization-fac-
torization statistics,” IEEE Internet of Things Journal, vol. 6,
Conflicts of Interest no. 4, pp. 6835–6842, 2019.
[7] D. C. Nguyen, P. N. Pathirana, M. Ding, and A. Seneviratne,
The authors declare that they have no conflicts of interest. “Blockchain for 5G and beyond networks: a state of the art
survey,” Journal of Network and Computer Applications,
Acknowledgments vol. 166, Article ID 102693, 2020.
[8] G. Praveen, V. Chamola, V. Hassija, and N. Kumar,
This work was supported by the Beijing Municipal Com- “Blockchain for 5G: a prelude to future telecommunication,”
mission of Education Project (nos. KM202111417001 and IEEE Network, vol. 34, no. 6, pp. 106–113, 2020.
Scientific Programming 11

[9] C. Rus-Casas, G. Jiménez-Castillo, J. D. Aguilar-Peña, [24] P. Patil, M. Sangeetha, and V. Bhaskar, “Blockchain for IoT
J. I. Fernández-Carrasco, and F. J. Muñoz-Rodrı́guez, “De- access control, security and privacy: a review,” Wireless
velopment of a prototype for monitoring photovoltaic self- Personal Communications, vol. 117, no. 3, pp. 1815–1834,
consumption systems,” Electronics, vol. 9, no. 1, p. 67, 2020. 2021.
[10] V. Sharma, “An energy-efficient transaction model for the [25] J. H. Chen, M. R. Chen, G. Q. Zeng, and J. S. Weng, “BDFL: a
blockchain-enabled Internet of vehicles (IoV),” IEEE Com- byzantine-fault-tolerance decentralized federated learning
munications Letters, vol. 23, no. 2, pp. 246–249, 2019. method for autonomous vehicle,” IEEE Transactions on Ve-
[11] H. Wang, H. Qin, M. Zhao, X. Wei, H. Shen, and W. Susilo, hicular Technology, vol. 70, no. 9, pp. 8639–8652, 2021.
“Blockchain-based fair payment smart contract for public [26] A. Yazdinejad, R. M. Parizi, A. Dehghantanha, Q. Zhang, and
cloud storage auditing,” Information Sciences, vol. 519, K.-K. R. Choo, “An energy-efficient SDN controller archi-
pp. 348–362, 2020. tecture for IoT networks with blockchain-based security,”
[12] A. Miglani, N. Kumar, V. Chamola, and S. Zeadally, IEEE Transactions on Services Computing, vol. 13, no. 4,
“Blockchain for Internet of energy management: review, pp. 625–638, 2020.
solutions, and challenges,” Computer Communications, [27] K. Yu, L. Tan, M. Aloqaily, H. Yang, and Y. Jararweh,
vol. 151, no. 1, pp. 395–418, 2020. “Blockchain-Enhanced data sharing with traceable and direct
[13] J. Kang, Z. Xiong, D. Niyato, D. Ye, D. I. Kim, and J. Zhao, revocation in IIoT,” IEEE transactions on industrial infor-
“Toward secure blockchain-enabled Internet of vehicles: matics, vol. 17, no. 11, pp. 7669–7678, 2021.
optimizing consensus management using reputation and
contract theory,” IEEE Transactions on Vehicular Technology,
vol. 68, no. 3, pp. 2906–2920, 2019.
[14] A. S. Almasoud, F. K. Hussain, and O. K. Hussain, “Smart
contracts for blockchain-based reputation systems: a sys-
tematic literature review,” Journal of Network and Computer
Applications, vol. 170, Article ID 102814, 2020.
[15] G. Liang, S. R. Weller, F. Luo, J. Zhao, and Z. Y. Dong,
“Distributed blockchain-based data protection framework for
modern power systems against cyber attacks,” IEEE Trans-
actions on Smart Grid, vol. 10, no. 3, pp. 3162–3173, 2019.
[16] N. B. Truong, K. Sun, G. M. Lee, and Y. Guo, “GDPR-
compliant personal data management: a blockchain-based
solution,” IEEE Transactions on Information Forensics and
Security, vol. 15, pp. 1746–1761, 2020.
[17] T. Ding, Z. Zeng, J. Bai, B. Qin, Y. Yang, and
M. Shahidehpour, “Optimal electric vehicle charging strategy
with markov decision process and reinforcement learning
technique,” IEEE Transactions on Industry Applications,
vol. 56, no. 5, pp. 5811–5823, 2020.
[18] Z. Zhou, B. Wang, M. Dong, and K. Ota, “Secure and efficient
vehicle-to-grid energy trading in cyber physical systems:
integration of blockchain and edge computing,” IEEE
Transactions on Systems, Man, and Cybernetics: Systems,
vol. 50, no. 1, pp. 43–57, 2020.
[19] F. Ahmad, M. Khalid, and B. K. Panigrahi, “An enhanced
approach to optimally place the solar powered electric vehicle
charging station in distribution network,” Journal of Energy
Storage, vol. 42, Article ID 103090, 2021.
[20] L. Yin and S. Li, “Hybrid metaheuristic multi-layer rein-
forcement learning approach for two-level energy manage-
ment strategy framework of multi-microgrid systems,”
Engineering Applications of Artificial Intelligence, vol. 104,
Article ID 104326, 2021.
[21] U. Bac and M. Erdem, “Optimization of electric vehicle re-
charge schedule and routing problem with time windows and
partial recharge: a comparative study for an urban logistics
fleet,” Sustainable Cities and Society, vol. 70, Article ID
102883, 2021.
[22] B. Zhou, J. Zou, C. Yung Chung et al., “Multi-microgrid
energy management systems: architecture, communication,
and scheduling strategies,” Journal of Modern Power Systems
and Clean Energy, vol. 9, no. 3, pp. 463–476, 2021.
[23] H. F. Atlam, M. A. Azad, A. G. Alzahrani, and G. Wills, “A
review of blockchain in Internet of things and AI,” Big Data
and Cognitive Computing, vol. 4, no. 4, p. 28, 2020.