Modelling weather effects for impact analysis of residential time-of-use electricity pricing

Analyzing the impact of pricing policies such as time-of-use (TOU) is challenging in the presence of confounding factors such as weather. Motivated by a lack of consensus and model selection details in prior work, we present a methodology for modelling the effect of weather on residential electricity demand. The best model is selected according to explanatory power, out-of-sample prediction accuracy, goodness of fit and interpretability. We then evaluate the effect of mandatory TOU pricing in a local distribution company in southwestern Ontario, Canada. We use a smart meter dataset of over 20,000 households which is particularly suited to our analysis: it contains data from the summer before and after the implementation of TOU pricing in November 2011, and all customers transitioned from tiered rates to TOU rates at the same time. We find that during the summer rate season, TOU pricing results in electricity conservation across all price periods. The average demand change during on-peak and mid-peak periods is −2.6% and −2.4% respectively. Changes during off-peak periods are not statistically significant. These TOU pricing effects are less pronounced compared to previous studies, underscoring the need for clear, reproducible impact analyses which include full details about the model selection process.

[1]  Daniela M. Witten,et al.  An Introduction to Statistical Learning: with Applications in R , 2013 .

[2]  Sanem Sergici,et al.  Household Response to Dynamic Pricing of Electricity - A Survey of the Empirical Evidence , 2010 .

[3]  Alexander Bruhns,et al.  A NON-LINEAR REGRESSION MODEL FOR MID-TERM LOAD FORECASTING AND IMPROVEMENTS IN SEASONALITY , 2005 .

[4]  Sanem Sergici,et al.  Dynamic Pricing of Electricity for Residential Customers: The Evidence from Michigan , 2012 .

[5]  Guy R. Newsham,et al.  The effect of utility time-varying pricing and load control strategies on residential summer peak electricity use: A review , 2010 .

[6]  Kenneth Train,et al.  Optional Time-of-Use Prices for Electricity: Econometric Analysis of Surplus and Pareto Impacts , 1994 .

[7]  Rob J Hyndman,et al.  Density Forecasting for Long-Term Peak Electricity Demand , 2010, IEEE Transactions on Power Systems.

[8]  A Gasparrini,et al.  Distributed lag non-linear models , 2010, Statistics in medicine.

[9]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[10]  Dean C. Mountain,et al.  A Disaggregated Nonhomothetic Modeling of Responsiveness to Residential Time-of-Use Electricity Rates , 1992 .

[11]  Helge V. Larsen,et al.  Long-term forecasting of hourly electricity load: Identification of consumption profiles and segmentation of customers , 2013 .

[12]  A. A. Weiss,et al.  Semiparametric estimates of the relation between weather and electricity sales , 1986 .

[13]  D. Ruppert.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[14]  Á. Pardo,et al.  Temperature and seasonality influences on Spanish electricity load , 2002 .

[15]  F. Ramsey,et al.  The statistical sleuth : a course in methods of data analysis , 2002 .

[16]  V. Muggeo.  Estimating regression models with unknown break‐points , 2003, Statistics in medicine.

[17]  Antonio Gasparrini,et al.  Distributed Lag Linear and Non-Linear Models in R: The Package dlnm. , 2011, Journal of statistical software.

[18]  Srinivasan Keshav,et al.  Critiquing Time-of-Use pricing in Ontario , 2013, 2013 IEEE International Conference on Smart Grid Communications (SmartGridComm).

[19]  Sanem Sergici,et al.  Dynamic pricing of electricity for residential customers: the evidence from Michigan , 2013 .

[20]  Alok Bhargava,et al.  Serial Correlation and the Fixed Effects Model , 1982 .

[21]  José Ramón Cancelo,et al.  Forecasting the electricity load from one day to one week ahead for the Spanish system operator , 2008 .

[22]  Peter R. Armstrong,et al.  Mid-term forecasting of urban electricity load to isolate air-conditioning impact , 2014 .

[23]  A. Zeileis.  Econometric Computing with HC and HAC Covariance Matrix Estimators , 2004 .

[24]  M. Muggeo,et al.  segmented: An R package to Fit Regression Models with Broken-Line Relationships , 2008 .

[25]  Allen K. Miedema,et al.  A comparative analysis of time-of-use electricity rate effects: The Arizona experiment , 1981 .

[26]  Julián Moral-Carcedo,et al.  Modelling the non-linear response of Spanish electricity demand to temperature variations , 2005 .

[27]  A. Faruqui,et al.  Household response to dynamic pricing of electricity: a survey of 15 experiments , 2010 .

[28]  Katrina Jessoe,et al.  The Effect of a Mandatory Time-of-Use Pricing Reform on Residential Electricity Use∗ , 2012 .

[29]  M. Gallanti,et al.  Impact of the enforcement of a time-of-use tariff to residential customers in Italy , 2013 .

[30]  Patrick Breheny,et al.  Visualization of Regression Models Using visreg , 2017, R J..

[31]  A. Harvey,et al.  Forecasting Hourly Electricity Demand Using Time-Varying Splines , 1993 .

[32]  Shirley Almon.  The Distributed Lag Between Capital Appropriations and Expenditures , 1965 .