Page 47 - DP-2-2

Design+ EV charging capacity through queuing model

be >1 h. The charging waiting time for EV users is longer.
Taking the above analysis into account, the maximum value
of the number of charging terminals in a public fast-CS is
13. By considering the above two constraints of facility
utilization rate and charging waiting time, the number of
charging terminals in the capacity optimization model of
public fast-CS is limited to between 6 and 13.

3. Results
In the previous section, the parameters of the public fast-CS
capacity optimization model are set and the constraints
are analyzed. After calculation, the optimal configuration
of the public fast-CS is obtained, as shown in Figure 8. By Figure 8. Electric vehicle users’ queuing time for charging
analyzing the charging service system by queuing theory,
we found that with six charging terminals, the charging
of EVs in the charging service system could be satisfied
by applying the highest power at the charging terminals,
but the queuing time of EV users was found as long as
73.6 min. When the number of charging terminals was
increased to 7, the queuing time decreased significantly to
11 min, exceeding the maximum queuing time that users
can tolerate (i.e., 10 min). When the number of charging
terminals was increased to 8 and 9, the queuing time was
3 min and 1 min, respectively. With a continued increase of
number of charging terminals, the queuing time dropped
to 0 and did not change significantly.
The charging facility of this CS is a split-type charging Figure 9. Charging service system costs
equipment, and the charger is placed in an integrated
charging box-type transformer. When charging, the the number of terminals because the charging power
transformer distributes power to a plurality of charging becomes smaller and the charging time is prolonged.
terminals, thereby charging a plurality of EVs, so the The investment cost increases linearly and slowly as the
power for charging EVs is closely related to the number of number of terminals increases because the investment cost
charging terminals. As the number of charging terminals of CS mainly comes from the fixed investments such as
increases, the charging power of a single charging terminal land rent and box-type transformer, (a compact, enclosed
decreases, resulting in a longer charging time for the EV. unit integrating transformer and charging components
When the number of charging terminals is 8, the total which is less affected by the number of charging terminals.
time for queuing and charging is 39 min. When there are
eight charging terminals, total time is 39 min. Any number The queuing time, facility service rate, and the cost of
fewer or >8 results in total time exceeding 40 min. each for public fast-CSs configured with different numbers
of charging terminals are shown in Table 3. The sojourn
Public fast-CSs configured with different charging time is the total time that the user stays in the CS, including
terminals entail different investment costs and user time queuing time and charging time.
costs. Figure 9 shows the investment costs, user time costs,
and social costs in each case for a fast-CS configured with 4. Discussion
6 – 13 charging terminals.
The findings of this study provide valuable insights into
It can be found that the comprehensive cost of the the optimization of public fast-CS capacity, particularly in
charging service system is lowest when eight charging urban settings where charging demand exhibits distinct
terminals are configured. When the charging terminals temporal patterns. The proposed model, which integrates
are <8, the user time cost is affected by the queuing time, queuing theory with real-world operational data, offers a
and the user time cost decreases with the increase of the practical approach to balancing investment costs and user
number of terminals. When there are more than eight time costs. The results indicate that configuring 8 charging
charging terminals, the user time cost increases with terminals for an 800 kVA charging box transformer is the

Volume 2 Issue 2 (2025) 10 doi: 10.36922/dp.4225