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
7 O7 s, }7 s: O3 L$ ]+ u
Distribution for Sustainable Vehicular Edge and Fog

0 P9 O. F' h1 a$ O! S1 g
Networks
/ @- N: E+ t6 L
1 b2 `$ _; F/ J' b& k8 p8 }

* g9 x# |% d7 tThe fast-growing popularity of electric vehicles
, r; h8 Y1 o4 s! e. t4 U2 g3 {(EVs) poses complex challenges for the existing power grid
8 N4 W% b/ l: c7 e2 A+ ~infrastructure to meet the high demands at peak charging hours.
0 \. y9 @, @- z3 d8 KDiscovering and transferring energy amongst EVs in mobile / ], D( C" w+ e% \6 a! G9 d  l3 @
vehicular edges and fogs is expected to be an effective solution for
; A$ z+ m* J, {- ubringing energy closer to where the demand is and improving the
+ Q7 [# Z' x/ i* P3 w$ O" A) m6 yscalability and flexibility compared to traditional charging
' H" w" r; _. Z8 osolutions. In this paper, we propose a fully-distributed energy
2 y, I  p- m, Aaware opportunistic charging approach which enables distributed   T" a% [, o$ G" u$ H3 ^7 @( Z- q
multi-layer adaptive edge cloud platform for sustainable mobile
% k+ s# \1 C$ c- Bautonomous vehicular edges which host dynamic on-demand + `2 |, t+ C( Q8 j1 B# h
virtual edge containers of on-demand services. We introduce a
$ c. a1 v) u/ Z- x1 V1 U# C/ Knovel Reinforcement Learning (Q-learning) based SmartCharge
3 @2 \' B. v/ N6 U+ S" M; jalgorithm formulated as a finite Markov Decision Process. We
3 i6 n% j4 s# k% Vdefine multiple edge energy states, transitions and possible actions
' f7 C6 j' B, s" H) j3 P7 h5 b6 R" bof edge nodes in dynamic complex network environments which
9 w" g1 F- T5 j% h$ r- `( l% x% uare adaptively resolved by multilayer real-time multidimensional ) }* D7 w1 B0 A- d
predictive analytics. This allows SmartCharge edge nodes to more 5 B* g( W/ r& P# m
accurately capture, predict and adapt to dynamic spatial-temporal
# w) S  K& {6 }% tenergy supply and demand as well as mobility patterns when + R+ a/ G' o( g9 [( u
energy peaks are expected. More specifically, SmartCharge edge
" u* w: A2 i+ n! ~8 f" X! znodes are able to autonomously and collaboratively understand ) S3 I& A8 N( `
when (how soon) and where the geo-temporal peaks are expected
: I% a+ Q# k; w2 I7 m4 S% R5 G- ato happen, thus enable better local prediction and more accurate $ y' d4 K* b" y
global distribution of energy resources. We provide multi-criteria
# M! Z) G0 e8 t+ D# k6 Z' W* b0 kevaluation of SmartCharge against competitive protocols over
9 w- g8 @+ q3 k: M3 Greal-world San Francisco Cab mobility traces and in the presence 8 P! m1 d( D4 l, y
of real-world users’ energy interest traces driven by Foursquare
6 ?8 I7 I8 W: k  c- \San Francisco dataset. We show that SmartCharge successfully & e: d: G# W1 e  q- w3 X- W
predicts and mitigates congestion in peak charging hours, reduces
4 P& Z2 S% I' ]the waiting time between vehicles sending energy demand requests 9 E1 y" ]6 s6 p7 B- M1 W$ p
and being successfully charged as well as significantly reduces the
) l3 O5 T8 z& L  c4 F; i% @' Stotal number of vehicles in need of energy. . ?, H8 u+ H. ?6 r' o  a6 B6 T; h

) `" q* A- t2 p/ W3 P1 F
/ J: e3 X3 Y- g2 k
9 j# M/ p( i8 l' B# U0 u# q) }4 F1 f9 `6 T, g' t

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