Prediction of Energy Consumption for Variable Customer Portfolios Including Aleatoric Uncertainty Estimation

TL;DR

This paper presents a deep learning approach that predicts hourly energy consumption and estimates aleatoric uncertainty using lognormal distributions, enabling probabilistic forecasts for individual customers and portfolios.

Contribution

It introduces a novel method combining deep neural networks with lognormal probabilistic layers to estimate both energy consumption and its aleatoric uncertainty.

Findings

Accurate probabilistic forecasts of hourly energy consumption.

Effective aggregation of individual uncertainties into portfolio predictions.

Demonstrated applicability to real-world smart meter data.

Abstract

Using hourly energy consumption data recorded by smart meters, retailers can estimate the day-ahead energy consumption of their customer portfolio. Deep neural networks are especially suited for this task as a huge amount of historical consumption data is available from smart meter recordings to be used for model training. Probabilistic layers further enable the estimation of the uncertainty of the consumption forecasts. Here, we propose a method to calculate hourly day-ahead energy consumption forecasts which include an estimation of the aleatoric uncertainty. To consider the statistical properties of energy consumption values, the aleatoric uncertainty is modeled using lognormal distributions whose parameters are calculated by deep neural networks. As a result, predictions of the hourly day-ahead energy consumption of single customers are represented by random variables drawn from lognormal distributions obtained as output from the neural network. We further demonstrate, how these random variables corresponding to single customers can be aggregated to probabilistic forecasts of customer portfolios of arbitrary composition.