arXiv subjects

Raffi Sevlian

Publications and source records attributed to Raffi Sevlian.

6 recordsLinked to original sources

arXiv2017

Short Term Electricity Load Forecasting on Varying Levels of Aggregation

We propose a simple empirical scaling law that describes load forecasting accuracy at different levels of aggregation. The model is justified based on a simple decomposition of individual consumption patterns. We show that for different forecasting methods and horizons, aggregating more customers improves the relative forecasting performance up to specific point. Beyond this point, no more improvement in relative performance can be obtained.

stat.AP↗

arXiv2017

Pricing Residential Electricity Based on Individual Consumption Behaviors

The conventional practice of retail electric utilities is to aggregate customers geographically. The utility purchases electricity for its customers via bulk transactions on the wholesale market, and it passes these costs along to its customers, the end consumers, through their rate plan. Typically, all residential consumers are offered the same per unit rate plan, which leads to cost sharing. Some consumers use their electricity at peak hours, when it is more expensive on the wholesale market, and others consume mostly at off peak hours, when it is cheaper, but they all enjoy the same per unit rate through their utility. This paper proposed a method for the utility to segment a population of consumers on the basis of their individual consumption patterns. An optimal recruitment algorithm was developed to aggregate consumers into groups with a relatively low per unit cost of electricity on the wholesale market. It was then proposed that the utility should group together enough consumers to ensure an adequately low forecast error, which is related to risks it faces in wholesale market transactions. Finally, it was shown that by repeated application of this process, the utility could segment the entire population into groups and offer them differentiated rate plans based on their actual consumption behavior. These groupings are stable in the sense that no one consumer can unilaterally improve her outcome.

math.OC↗

arXiv2017

Actively Calibrated Line Mountable Capacitive Voltage Transducer For Power Systems Applications

A class of Actively Calibrated Line Mounted Capacitive Voltage Transducers (LMCVT) are introduced as a viable line mountable instrumentation option for deploying large numbers of voltage transducers onto the medium and high voltage systems. Active Calibration is shown to reduce the error of line mounted voltage measurements by an order of magnitude from previously published techniques. The instrument physics and sensing method is presented and the performance is evaluated in a laboratory setting. Finally, a roadmap to a fully deployable prototype is shown.

physics.ins-det↗

arXiv2016

Distribution System Outage Detection using Consumer Load and Line Flow Measurements

An outage detection framework for power distribution networks is proposed. Given the tree structure of the distribution system, a method is developed combining the use of real-time power flow measurements on edges of the tree with load forecasts at the nodes of the tree. A maximum a posteriori detector {\color{black} (MAP)} is formulated for arbitrary number and location of outages on trees which is shown to have an efficient detector. A framework relying on the maximum missed detection probability is used for optimal sensor placement and is solved for tree networks. Finally, a set of case studies is considered using feeder data from the Pacific Northwest National Laboratories. We show that a 10\% loss in mean detection reliability network wide reduces the required sensor density by 60 \% for a typical feeder if efficient use of measurements is performed.

eess.SY↗

arXiv2014

Distribution System Load and Forecast Model

This short document provides experimental evidence for the set of assumptions on the mean load and forecast errors made in \cite{Sevlian2014A_Outage} and \cite{Sevlian2014B_Outage}. We show that the mean load at any given node is distributed normally, where we compute the mean and variance. We then present an aggregation-error curve for a single day ahead forecaster. Residual analysis shows that beyond 500 customers, gaussian residuals is a reasonable model. We then show the forecaster has uncorrelated errors.

stat.AP↗

arXiv2010

Travel Time Estimation Using Floating Car Data

This report explores the use of machine learning techniques to accurately predict travel times in city streets and highways using floating car data (location information of user vehicles on a road network). The aim of this report is twofold, first we present a general architecture of solving this problem, then present and evaluate few techniques on real floating car data gathered over a month on a 5 Km highway in New Delhi.

cs.LG↗