数学建模社区-数学中国

标题: Energy-Aware Opportunistic Charging and Energy Distribution for Sustainable ... [打印本页]
作者: 杨利霞    时间: 2020-11-9 15:10
标题: Energy-Aware Opportunistic Charging and Energy Distribution for Sustainable ...
Energy-Aware Opportunistic Charging and Energy

: j& l7 p( k4 N4 X* O3 p
Distribution for Sustainable Vehicular Edge and Fog
) B) _1 m2 N1 j- R
Networks
1 Q) C: }( D7 `. c5 r# }- o& K5 H

/ O; R) D* _0 {' u
* G6 r! `+ d7 iThe fast-growing popularity of electric vehicles
# e# n1 \: j+ F0 s! l9 q. T(EVs) poses complex challenges for the existing power grid
0 p/ A8 {  d  N8 Y; rinfrastructure to meet the high demands at peak charging hours. 0 L+ J  U9 d2 a$ G' E
Discovering and transferring energy amongst EVs in mobile
2 e3 f! g9 Y9 |4 W5 q. \vehicular edges and fogs is expected to be an effective solution for 6 D7 S% g9 ~* K
bringing energy closer to where the demand is and improving the
; z' P$ Z) E8 b: xscalability and flexibility compared to traditional charging
/ G3 r1 O% g- L7 L( n4 Q, jsolutions. In this paper, we propose a fully-distributed energy; |" V- s. R1 R! P0 |$ J6 c
aware opportunistic charging approach which enables distributed ' e8 _$ X* d4 z- k2 E% |3 i7 O# I- D; \
multi-layer adaptive edge cloud platform for sustainable mobile
( s* V$ [* a3 A5 D& Qautonomous vehicular edges which host dynamic on-demand # ?& b' l  Q! v3 Z1 S! ]
virtual edge containers of on-demand services. We introduce a
7 [. d, d. g& z. unovel Reinforcement Learning (Q-learning) based SmartCharge
3 d3 Y  e- E( v% y/ y: dalgorithm formulated as a finite Markov Decision Process. We
* Q. `7 _1 O2 s* o; F. jdefine multiple edge energy states, transitions and possible actions
$ S! D  j2 _- G# U2 s% i1 j, E5 j/ \of edge nodes in dynamic complex network environments which 5 D" `& s2 m6 ~. g' Q2 m1 b
are adaptively resolved by multilayer real-time multidimensional : {- L; n2 _$ O/ R& _& M
predictive analytics. This allows SmartCharge edge nodes to more
4 I/ R* V) A3 q- M, H' A; v( Faccurately capture, predict and adapt to dynamic spatial-temporal
8 g  h8 T+ Y1 F" d- t8 U4 O! menergy supply and demand as well as mobility patterns when ! k/ E& a" w$ B% l) B9 q0 ~9 N
energy peaks are expected. More specifically, SmartCharge edge
# x0 _6 x  Q2 C7 M/ b0 ~nodes are able to autonomously and collaboratively understand
. C/ c# X  }- m: d5 i. h# _8 A- ewhen (how soon) and where the geo-temporal peaks are expected
7 k3 g2 C5 [: c5 G" Nto happen, thus enable better local prediction and more accurate
. o* Q" m4 m3 D5 Lglobal distribution of energy resources. We provide multi-criteria
& x! B7 a! N3 _7 ^: `9 w" uevaluation of SmartCharge against competitive protocols over 8 G  u" r# G* _$ B* l8 C4 w6 K
real-world San Francisco Cab mobility traces and in the presence 8 `9 L/ D, i+ }5 p$ Z" Y
of real-world users’ energy interest traces driven by Foursquare
- \# K' {; g$ Y% Z+ _. uSan Francisco dataset. We show that SmartCharge successfully 6 T3 i# @8 u7 j, q3 D8 n7 f" p
predicts and mitigates congestion in peak charging hours, reduces : P6 p8 l/ ~3 f: Z6 A6 o& |( G7 x, f
the waiting time between vehicles sending energy demand requests 6 |& ~9 e  \; w( m! P
and being successfully charged as well as significantly reduces the
1 B0 {! a( G" }$ L' ]4 Ktotal number of vehicles in need of energy.
6 J" s( e" M, D1 m0 `, ~3 ]% @, N- K% w6 K
" ^3 s& K' y3 m! `
# W, Y! w: T: B* E  s0 }. C! Z+ ^, ^

1 W) k& ~2 _! B  D

Energy-Aware Opportunistic Charging and Energy.pdf

1.17 MB, 下载次数: 0, 下载积分: 体力 -2 点



欢迎光临 数学建模社区-数学中国 (http://www.madio.net/) Powered by Discuz! X2.5