Forecasting Uncertainty in Electricity Smart Meter Data by Boosting Additive Quantile Regression

Smart electricity meters are currently deployed in millions of households to collect detailed individual electricity consumption data. Compared with traditional electricity data based on aggregated consumption, smart meter data are much more volatile and less predictable. There is a need within the energy industry for probabilistic forecasts of household electricity consumption to quantify the uncertainty of future electricity demand in order to undertake appropriate planning of generation and distribution. We propose to estimate an additive quantile regression model for a set of quantiles of the future distribution using a boosting procedure. By doing so, we can benefit from flexible and interpretable models, which include an automatic variable selection. We compare our approach with three benchmark methods on both aggregated and disaggregated scales using a smart meter data set collected from 3639 households in Ireland at 30-min intervals over a period of 1.5 years. The empirical results demonstrate that our approach based on quantile regression provides better forecast accuracy for disaggregated demand, while the traditional approach based on a normality assumption (possibly after an appropriate Box-Cox transformation) is a better approximation for aggregated demand. These results are particularly useful since more energy data will become available at the disaggregated level in the future.

[1]  P. Pinson,et al.  Very‐short‐term probabilistic forecasting of wind power with generalized logit–normal distributions , 2012 .

[2]  J. A. Carta,et al.  A review of wind speed probability distributions used in wind energy analysis: Case studies in the Canary Islands , 2009 .

[3]  Tao Hong,et al.  Short Term Electric Load Forecasting , 2012 .

[4]  P. Bühlmann,et al.  Boosting with the L2-loss: regression and classification , 2001 .

[5]  Rob J. Hyndman,et al.  Boosting multi-step autoregressive forecasts , 2014, ICML.

[6]  Tao Hong,et al.  Probabilistic electric load forecasting: A tutorial review , 2016 .

[7]  R. Koenker Additive models for quantile regression: Model selection and confidence bandaids , 2010 .

[8]  Patrick E. McSharry,et al.  Quantile Forecasting of Wind Power Using Variability Indices , 2013 .

[9]  A. Raftery,et al.  Strictly Proper Scoring Rules, Prediction, and Estimation , 2007 .

[10]  Yannig Goude,et al.  Massive-Scale Simulation of Electrical Load in Smart Grids Using Generalized Additive Models , 2015 .

[11]  J. Friedman Greedy function approximation: A gradient boosting machine. , 2001 .

[12]  R. Koenker Quantile Regression: Name Index , 2005 .

[13]  Ram Rajagopal,et al.  Distribution System Load and Forecast Model , 2014 .

[14]  A. Raftery,et al.  Probabilistic forecasts, calibration and sharpness , 2007 .

[15]  Souhaib Ben Taieb,et al.  A gradient boosting approach to the Kaggle load forecasting competition , 2014 .

[16]  R. Ramanathan,et al.  Probabilistic methods in forecasting hourly loads , 1993 .

[17]  Torsten Hothorn,et al.  Model-based Boosting 2.0 , 2010, J. Mach. Learn. Res..

[18]  Ram Rajagopal,et al.  Short Term Electricity Load Forecasting on Varying Levels of Aggregation , 2014, ArXiv.

[19]  Silvia Santini,et al.  Revealing Household Characteristics from Smart Meter Data , 2014 .

[20]  Jianxue Wang,et al.  Review on probabilistic forecasting of wind power generation , 2014 .

[21]  R. Tibshirani,et al.  Generalized Additive Models , 1991 .

[22]  Heng Huang,et al.  Using Smart Meter Data to Improve the Accuracy of Intraday Load Forecasting Considering Customer Behavior Similarities , 2015, IEEE Transactions on Smart Grid.

[23]  Gerhard Tutz,et al.  Variable Selection and Model Choice in Geoadditive Regression Models , 2009, Biometrics.

[24]  H. Madsen,et al.  Predictive Densities for Day-Ahead Electricity Prices Using Time-Adaptive Quantile Regression , 2014 .

[25]  Siddharth Arora,et al.  Forecasting electricity smart meter data using conditional kernel density estimation , 2014, 1409.2856.

[26]  Rob J. Hyndman,et al.  Bandwidth selection for kernel conditional density estimation , 2001 .

[27]  M. Genton,et al.  Short‐Term Wind Speed Forecasting for Power System Operations , 2012 .

[28]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[29]  Tri Kurniawan Wijaya,et al.  Forecasting Uncertainty in Electricity Demand , 2015, AAAI Workshop: Computational Sustainability.

[30]  Tao Hong,et al.  Probabilistic Load Forecasting via Quantile Regression Averaging on Sister Forecasts , 2017, IEEE Transactions on Smart Grid.

[31]  Henrik Madsen,et al.  Online short-term solar power forecasting , 2009 .

[32]  Tin Kam Ho,et al.  Demand forecasting in smart grids , 2014, Bell Labs Technical Journal.

[33]  Rob J Hyndman,et al.  Short-Term Load Forecasting Based on a Semi-Parametric Additive Model , 2012, IEEE Transactions on Power Systems.

[34]  Yoav Freund,et al.  Boosting: Foundations and Algorithms , 2012 .

[35]  P. Bühlmann,et al.  Boosting With the L2 Loss , 2003 .

[36]  V. Chernozhukov,et al.  QUANTILE AND PROBABILITY CURVES WITHOUT CROSSING , 2007, 0704.3649.

[37]  Torsten Hothorn,et al.  Boosting additive models using component-wise P-Splines , 2008, Comput. Stat. Data Anal..

[38]  Torsten Hothorn,et al.  Prediction intervals for future BMI values of individual children - a non-parametric approach by quantile boosting , 2012, BMC Medical Research Methodology.

[39]  Pin T. Ng,et al.  Quantile smoothing splines , 1994 .

[40]  Yu-Hsiang Hsiao,et al.  Household Electricity Demand Forecast Based on Context Information and User Daily Schedule Analysis From Meter Data , 2015, IEEE Transactions on Industrial Informatics.

[41]  T. Gneiting Quantiles as optimal point forecasts , 2011 .

[42]  D. Ruppert Selecting the Number of Knots for Penalized Splines , 2002 .