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
& o5 [" b6 b7 K) y6 d( D
Distribution for Sustainable Vehicular Edge and Fog
# {. q# A; i! C/ m# @Networks
9 ~) P$ d' M9 K
3 ~) @; U. ?6 B# x% O8 u1 @) W
# K$ A( Y! L- `6 Q9 f' t+ t E
The fast-growing popularity of electric vehicles 2 O; D$ p2 [( |: a# i( O- J* f1 O8 f
(EVs) poses complex challenges for the existing power grid
& ?1 W+ @, A% k8 g1 S; Ainfrastructure to meet the high demands at peak charging hours.
/ f4 u) l! c" t# j1 X$ q' bDiscovering and transferring energy amongst EVs in mobile , ?* ]' j0 c9 q4 q4 j! T
vehicular edges and fogs is expected to be an effective solution for & Q/ `* P* B$ E% _; F9 P
bringing energy closer to where the demand is and improving the
7 _) k% d8 L" L' Y+ Z; w2 S8 @! Ascalability and flexibility compared to traditional charging , Y7 q' n2 ?* n
solutions. In this paper, we propose a fully-distributed energy; M J; ?2 |% J" F* A( F2 U+ e8 [
aware opportunistic charging approach which enables distributed
8 w8 Z7 `; Y, x! ~5 w7 Dmulti-layer adaptive edge cloud platform for sustainable mobile
) D$ H g4 J3 C: R4 S$ l' @autonomous vehicular edges which host dynamic on-demand . Y+ f( G9 v- W+ d
virtual edge containers of on-demand services. We introduce a ' r' {# D; ]! I
novel Reinforcement Learning (Q-learning) based SmartCharge " @. o9 E; q. P6 n% C
algorithm formulated as a finite Markov Decision Process. We
% o4 z! V7 B+ N* L+ l7 W, \3 n* Odefine multiple edge energy states, transitions and possible actions
+ \' P4 @& P t X% _: T- Fof edge nodes in dynamic complex network environments which 6 m: E5 k' G( m9 F3 C8 @2 Q) x
are adaptively resolved by multilayer real-time multidimensional % R: `6 \7 y# a/ J v* D |0 p
predictive analytics. This allows SmartCharge edge nodes to more
# E3 z% R; n) [0 o, s& taccurately capture, predict and adapt to dynamic spatial-temporal
* S2 n4 S" r3 D5 ~# oenergy supply and demand as well as mobility patterns when
- _& ~9 M$ {$ h7 m: A$ Renergy peaks are expected. More specifically, SmartCharge edge
' W6 v4 h/ b6 B% rnodes are able to autonomously and collaboratively understand ( z* W7 ]3 i1 d2 k* g- i
when (how soon) and where the geo-temporal peaks are expected
4 n5 W" W5 K5 |to happen, thus enable better local prediction and more accurate : e8 O3 U2 [4 j# }4 T
global distribution of energy resources. We provide multi-criteria $ K. c E" T" k) ~
evaluation of SmartCharge against competitive protocols over & G1 d3 p X, I. G4 J& |
real-world San Francisco Cab mobility traces and in the presence & O, c+ H3 a3 O% Y
of real-world users’ energy interest traces driven by Foursquare
4 N3 Y2 a- d# }1 ISan Francisco dataset. We show that SmartCharge successfully
; I5 Q! P2 g6 E' {% Mpredicts and mitigates congestion in peak charging hours, reduces
0 N. v' M. i3 L; o. f3 X1 j& o& L+ qthe waiting time between vehicles sending energy demand requests ; A, L f7 X' X/ C$ ]- ?
and being successfully charged as well as significantly reduces the
# h+ m3 ]$ @- j- O# Z& ~+ qtotal number of vehicles in need of energy.
2 i. F. [7 h" ?+ m- t: {' C3 N# |/ X' {( S R6 Z& g
: t$ D6 d: F( N8 @
1 P3 G$ v$ b6 k% d/ x8 L
8 `$ z' q" f" W% }1 n/ P
-
-
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 |