A. Nait-Sidi-Moh et al. / Procedia Computer Science 141 (2018) 127–134
133
6
NaitSidiMoh et al./ Procedia Computer Science 00 (2018) 000–000
requests (𝑛𝑛
!"
), which are received at a given time. Furthermore, the found average charging rates (𝜆𝜆(𝑗𝑗)) and the
average charging times (𝑡𝑡
!
) based on accordingly the equations (2), and (3), (4), (5) will be main parameters to
determine the charging rates and times of EVs set individually.
The two considered scenarios: with and without considering required energy for EVs are used to show the
differences between the proposed charging policies. The Gantt chart of Fig. 5 (a)
shows the charging time and
waiting time for each EV without considering the needed energy for charging requests. Without considering this
energy, each EV is planned to be fully charged, consequently certain EVs have to wait for long time. Thus the
accumulation of waiting times becomes more and more raised with the arrival of other charging requests. However,
by considering the required energy of each demand and the inter-arrival of all charging requests, the predictive
charging time varies from minimum charging time 𝐶𝐶𝐶𝐶
!"#
, corresponding in our case to
the minimum accepted
charging rate 𝑃𝑃
!"#
, to the arrival time of the next accepted charging request, which is limited by 𝐶𝐶𝐶𝐶
!"#
. If the next
charging demand arrives after 𝐶𝐶𝐶𝐶
!"#
, the charging point is free and can be used at any time. In this case, the EVs
will be charged with at least an accepted amount of energy and the accumulated waiting of EVs can be considerably
reduced. The Gantt chart depicted in Fig. 5 (b) illustrates the results of this second case. Here, we consider that the
needed battery energy varies between 20 and 80 %. We keep our first charging condition that charges each battery at
least 50%, when the next EV charging demand arrives before the desired time and the battery
can be charged for
more than 50%.
(a) (b)
Fig. 5. Waiting time and charging time (a) without needed energy, (b) with considering needed energy
Furthermore, Fig. 6 (a) presents the difference between desired charging rates (according to the needed energy of
each EV) and proposed charging rates. The needed energy is considered in this case for showing the comparison
between the two approaches. We can see the difference of the desired charging rate and the proposed charging rate.
The proposed charging rate is less than the desired charging rate because it depends on the requested time interval.
In Figure 6 (b), distinguishable differences of the waiting times for the two studied cases are presented. The large
scale reducing of the waiting times is compared to the battery SoC and charging without this information. Battery
energy
information, such as the SoC and related charging technology (slow or fast charging) can be helpful for the
companies of energy distribution and smart grids in order to schedule daily loading costs. When the decision of
charging EVs with 100% of energy is not kept, the accumulation of EVs waiting times can be reduced till 49.8%.
NaitSidiMoh et al./ Procedia Computer Science 00 (2018) 000–000
7
(a) (b)
Fig. 6. (a) the difference between the desired and proposed charging rate, (b) the
waiting time evolution
Dostları ilə paylaş: