Avoiding overages by deferred aggregate demand for PEV charging on the smart grid

Abstract

Abstract-In this paper, we model the aggregate overnight demand for electricity by a large community of (possibly hybrid) plug-in electric vehicles (PEVs) each of whose power demand follows a prescribed profile and is interruptible. The community is serviced by a regional electrical utility which is assumed to purchase electricity from a state/national distribution grid according to a flat-rate φ per kilowatt-unit-time up to a threshold L, and thereafter overage charges π > φ are leveed per kilowatt-unit-time. Rather than a spot-price system for household consumers (which would necessarily need to be operated by automated means overnight when most people sleep), the "grid" (regional utility) is "smart" in that it monitors its total load and, when overages threaten, can reduce load by signaling certain consumers to defer their load by one unit of time. In this paper, we model undeferred load by a Gaussian process which we justify by means of a functional central limit theorem. This limiting Gaussian process is the arrival process of a discrete-time queue which is used to model the deferred load over a finite timehorizon. We can then compute the mean amount of overage at the end of this time horizon (say at 6 AM when charging is to be completed ahead of the morning commute).

Event stream processing for improved situational awareness in the smart grid

“…In other words, to extract meaningful information from PMU data, evolving data stream mining algorithms should be applied. That is different from other statistical distributions that deal with static and limited samples of power systems' data (Kashyap & Callaway, 2010;Pang, Kesidis, & Konstantopoulos, 2012;Schellenberg, Rosehart, & Aguado, 2005) 3.2.2.1. Hoeffding window trees using ADWIN.…”

Stochastic Information Management in Smart Grid

“…Taking advantage of wireless communications, the charging demand of an EV which is physically connected to a charger can be deferred to reduce the peak demand of the grid, if the available charging sockets are fully occupied [149]. A GI/D/1 queueing model is used in [150], where a general arrival process (specifically, a Gaussian process) is used to capture different power consumption profiles of different EVs. A controllable deterministic service process is used to model the threshold of aggregated charing power specified by the utility.…”

Automated scheduling of deferrable PEV/PHEV load by power-profile unevenness

“…Their objective is a long-term average cost function of instantaneous power consumed. Noninterruptible demand is also considered in [13], but there Poisson arrivals are due to an assumed mechanism to desynchronize demand (not the subject of policy optimization), the power profile is assumed unimodal, again consistent with that of the batteries of certain PEV /PHEV batteries [8]-[ 1 0], and all demand was assumed "available" at the start of a finite (e. g. , overnight) charging interval; demand was incrementally deferred to avoid overages. Specifically which consumers were deferred at a given time could be ascertained using a heuristic scoring system as for example given in [11].…”

“…In [13], we studied PEVIPHEV battery power-charging profiles of unimodal type [8]- [10], an idealization of which is shown in Fig …”