Results (
Thai) 1:
[Copy]Copied!
The recent advent of smart meters has led to large micro-level datasets. For thefirst time, the electricity consumption at individual sites is available on a near real-time basis.Efficient management of energy resources, electric utilities, and transmission grids, can begreatly facilitated by harnessing the potential of this data. The aim of this study is to generateprobability density estimates for consumption recorded by individual smart meters. Suchestimates can assist decision making by helping consumers identify and minimize theirexcess electricity usage, especially during peak times. For suppliers, these estimates can beused to devise innovative time-of-use pricing strategies aimed at their target consumers. Weconsider methods based on conditional kernel density (CKD) estimation with theincorporation of a decay parameter. The methods capture the seasonality in consumption, andenable a nonparametric estimation of its conditional density. Using eight months of halfhourlydata for one thousand meters, we evaluate point and density forecasts, for lead timesranging from one half-hour up to a week ahead. We find that the kernel-based methodsoutperform a simple benchmark method that does not account for seasonality, and comparewell with an exponential smoothing method that we use as a sophisticated benchmark. Togauge the financial impact, we use density estimates of consumption to derive predictionintervals of electricity cost for different time-of-use tariffs. We show that a simple strategy ofswitching between different tariffs, based on a comparison of cost densities, deliverssignificant cost savings for the great majority of consumers
Being translated, please wait..
