Energy-Aware Opportunistic Charging and Energy Distribution for Sustainable ...

杨利霞

Energy-Aware Opportunistic Charging and Energy

Distribution for Sustainable Vehicular Edge and Fog

Networks

: ?8 f5 d! w6 P& H+ w
) F( [. T% T0 p4 I5 [" u0 @
8 {) r* q8 @7 O  i+ R+ o) Q* f1 O6 zThe fast-growing popularity of electric vehicles
5 h; ]% U5 |3 |# N" j4 B1 o7 O(EVs) poses complex challenges for the existing power grid
/ P6 L' h. h" b$ U/ zinfrastructure to meet the high demands at peak charging hours.
- s  q8 ?8 u4 [0 o4 G9 RDiscovering and transferring energy amongst EVs in mobile
vehicular edges and fogs is expected to be an effective solution for
- [/ H5 A, Q0 U% J) [3 Vbringing energy closer to where the demand is and improving the
; ]7 ~7 T+ L3 {scalability and flexibility compared to traditional charging
4 @/ L, R( U, {* d' u# Bsolutions. In this paper, we propose a fully-distributed energy
$ Y+ X. I1 j/ saware opportunistic charging approach which enables distributed
multi-layer adaptive edge cloud platform for sustainable mobile
( e0 |, e& ]/ {autonomous vehicular edges which host dynamic on-demand
9 e+ @' h, H+ bvirtual edge containers of on-demand services. We introduce a
( K5 {; {/ s; I8 H* j( inovel Reinforcement Learning (Q-learning) based SmartCharge
  Z4 i/ X+ P- malgorithm formulated as a finite Markov Decision Process. We
0 r! \0 ]( s* P' J% T4 a7 Hdefine multiple edge energy states, transitions and possible actions
; D# A  S8 k: w7 Xof edge nodes in dynamic complex network environments which
( [% O. |% n' J" @* V& Dare adaptively resolved by multilayer real-time multidimensional
predictive analytics. This allows SmartCharge edge nodes to more
' H/ |' x( g; ^! [2 maccurately capture, predict and adapt to dynamic spatial-temporal
% G9 {! c& _( U# k* i2 M" {7 Denergy supply and demand as well as mobility patterns when
energy peaks are expected. More specifically, SmartCharge edge
nodes are able to autonomously and collaboratively understand
! L0 V- J  P5 Z  b3 @when (how soon) and where the geo-temporal peaks are expected
to happen, thus enable better local prediction and more accurate
& x5 w) t) L% X8 i8 wglobal distribution of energy resources. We provide multi-criteria
3 Z6 Z" d+ G3 W8 E! }evaluation of SmartCharge against competitive protocols over
+ y8 b' T: P  u/ ereal-world San Francisco Cab mobility traces and in the presence
+ C4 Q7 [4 x# w! k8 j& O& c/ jof real-world users’ energy interest traces driven by Foursquare
San Francisco dataset. We show that SmartCharge successfully
1 j# x* r$ |) `" U" m, Apredicts and mitigates congestion in peak charging hours, reduces
4 j6 H5 n+ [! u2 j9 W, K/ sthe waiting time between vehicles sending energy demand requests
and being successfully charged as well as significantly reduces the
. n2 ?+ ]) w4 Y2 t" Y5 utotal number of vehicles in need of energy.
) L( P8 `1 z& m8 h
7 {* C) Z8 P" I) |
+ I1 s# S2 ]8 R4 X9 u: u- G9 l

: _2 C6 p' J3 s. j+ O8 {