Allocation of charging stations in Kota city using data analysis/machine learning

ABSTRACT

This paper proposes the allocation of charging stations of optimal capacity at appropriate locations within Kota city in Rajasthan state of India. The location of charging stations has been identified by considering various parameters including population density, family income, distribution network performance, number of electric vehicles and transformer capacity. The proposed approach of allocation of charging stations has been modeled mathematically and also by machine learning algorithms such as data mining, clustering with all the possible scenarios and constraints. In identifying suitable places for positioning charging stations in the city by observing different parameters and aspects, i.e. geo-statistical approaches, minimum traveling time and travel distance for charging stations has been included in the proposed approach. The proposed approach has been tested in MATLAB/Simulink environment.

KEYWORDS:

• EV
• EVCS
• power supply
• charging time
• distribution

Disclosure statement

No potential conflict of interest was reported by the author(s).

Nomenclature

EV	=	Electric Vehicles
CS	=	Charging Station
EVCS	=	Electric Vehicles Charging Station
PM10	=	Particulate Matter
$\mathit{Dp}$	=	density of population
$\mathit{N}$	=	total population as a number of people
$\mathit{A}$	=	the land area covered by that population
$k_n$	=	for a certain period of time the total number of vehicles per charge
$P_D$	=	power distribution function
C	=	total charging capacity of the charging station
$\mathit{\delta }$	=	arrival rate
$\mathit{\mu }$	=	service rate of one server
$P_i^{\prime }(),U_i^{\prime }()$	=	flow from bus-I into the system in terms of voltage magnitudes and angles
$P_{ni}^{\prime },P_{ci}^{\prime }$	=	generations at the bus
$U_{ni}^{\prime }$$U_{ci}^{\prime }$	=	Demand at the bus
$\mathit{f}\left(l_n\right)$	=	function define for real variable
$\mathit{f}\prime \left(l_n\right)$	=	function’s derivative for real variable
$\mathit{n}$	=	number of iterations
$l_n$	=	root at value nth iteration.
$R_i$	=	real power for any bus i
$\mathit{V}{m}_i$	=	voltage magnitude
$e_i$	=	real components of the bus voltage vi
$f_i$	=	imaginary components of the bus voltage vi
$A_{\mathit{ik}}$	=	conductance
$D_{\mathit{ik}}$	=	susceptance
$\mathit{fTr}$	=	Transformer’s
$\mathit{fPc}$	=	power analysis
$\mathit{fAr}$	=	various locality Area
$\mathit{fPd}$	=	population density
$\mathit{fEc}$	=	electricity consumption
$\mathit{fnv}$	=	Vehicle counts