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

- B1 r6 N, P6 o. |7 `, s
Distribution for Sustainable Vehicular Edge and Fog

- N4 w) |3 P! |  z4 c/ u6 \* w
Networks

! t6 }4 Z$ T3 i5 ?
) T; v& N1 Q6 Z6 _- ]7 o* O! L& C$ c2 p- J$ d
The fast-growing popularity of electric vehicles 9 d  ~5 Q" r" R
(EVs) poses complex challenges for the existing power grid
7 i2 V) g% q/ D& \" U0 iinfrastructure to meet the high demands at peak charging hours. # c( h7 v$ U' s+ `+ w
Discovering and transferring energy amongst EVs in mobile
# t" ^4 Z$ a7 [5 t5 Q8 ^/ v3 Wvehicular edges and fogs is expected to be an effective solution for ( D* V" `1 J6 y, F& j" S# w  ]4 S5 e
bringing energy closer to where the demand is and improving the , w6 _* m0 L( \* k: X
scalability and flexibility compared to traditional charging
. E5 T: N8 x9 f& x* h  zsolutions. In this paper, we propose a fully-distributed energy
7 c/ j# J  W5 w4 s. ]aware opportunistic charging approach which enables distributed
3 d' F, B6 x: j3 {. zmulti-layer adaptive edge cloud platform for sustainable mobile
) G! J' L. Q5 S3 E& Aautonomous vehicular edges which host dynamic on-demand 0 x3 ]( _( Q  z% }* ]$ N- D; S0 ]$ t
virtual edge containers of on-demand services. We introduce a
2 N* Q+ E0 F( b- q* A& \novel Reinforcement Learning (Q-learning) based SmartCharge
4 R% f4 }# ~+ |  v1 qalgorithm formulated as a finite Markov Decision Process. We
1 J* R" s: a) W9 y1 j& `; }define multiple edge energy states, transitions and possible actions . V9 l6 ~) m0 e6 V
of edge nodes in dynamic complex network environments which - d  f+ t' N0 y$ N8 t  W
are adaptively resolved by multilayer real-time multidimensional
3 t2 c) N$ x# f0 ]) mpredictive analytics. This allows SmartCharge edge nodes to more + Z5 y. f5 d7 N1 p
accurately capture, predict and adapt to dynamic spatial-temporal 3 u3 P6 L" j" ]: J8 _. ^7 A
energy supply and demand as well as mobility patterns when % @9 C6 y5 r3 }' b* n; k+ o* X, E' y
energy peaks are expected. More specifically, SmartCharge edge 8 d; ]  S( O2 K2 h6 n/ i  [7 e. V
nodes are able to autonomously and collaboratively understand
! q' h3 ?* l" F. Z& |+ ^8 L! xwhen (how soon) and where the geo-temporal peaks are expected
0 C& P: Q1 S7 t  T! F2 ^+ N+ bto happen, thus enable better local prediction and more accurate 0 @4 t2 H  d) `# U
global distribution of energy resources. We provide multi-criteria
9 J! x  X$ B% g' a5 n0 [evaluation of SmartCharge against competitive protocols over
3 j; S# `# y% C$ n  J$ \, yreal-world San Francisco Cab mobility traces and in the presence / _% \7 g2 W( }  v
of real-world users’ energy interest traces driven by Foursquare
4 ?5 S8 u( M2 L# g3 ^, DSan Francisco dataset. We show that SmartCharge successfully 0 z) P* L+ ^2 v2 C4 V
predicts and mitigates congestion in peak charging hours, reduces - A# k0 H+ Z! O0 n
the waiting time between vehicles sending energy demand requests
2 c& {5 S* H" n6 X( G- Tand being successfully charged as well as significantly reduces the : _5 E: v. |1 s) z% T+ R8 K: @
total number of vehicles in need of energy. : \! M! v: p! R$ x  }

- J. K5 y8 z5 e1 v" P. l0 V
% H5 ~" F9 ]0 ~" U4 ^7 ?7 z
% `9 T, _) q7 @! m2 O0 A- |, x( |9 W3 Q: x. H3 E6 W* X% m

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