Forecasting for demand response in smart grids: An analysis on use of anthropologic and structural data and short term multiple loads forecasting Fahad Javed a,⇑ , Naveed Arshad a , Fredrik Wallin b , Iana Vassileva b , Erik Dahlquist b a Department of Computer Science, LUMS School of Science and Engineering, Lahore, Pakistan b Department of Public Technology, Malardalen University, Vasteras, Sweden article info Article history: Received 17 August 2011 Received in revised form 9 January 2012 Accepted 12 February 2012 Available online xxxx Keywords: Smart grids Demand response Load forecasting Short term multiple loads forecasting abstract The electric grid is changing. With the smart grid the demand response (DR) programs will hopefully make the grid more resilient and cost efficient. However, a scheme where consumers can directly partic- ipate in demand management requires new efforts for forecasting the electric loads of individual con- sumers. In this paper we try to find answers to two main questions for forecasting loads for individual consumers: First, can current short term load forecasting (STLF) models work efficiently for forecasting individual households? Second, do the anthropologic and structural variables enhance the forecasting accuracy of individual consumer loads? Our analysis show that a single multi-dimensional model fore- casting for all houses using anthropologic and structural data variables is more efficient than a forecast based on traditional global measures. We have provided an extensive empirical evidence to support our claims. Ó 2012 Elsevier Ltd. All rights reserved. 1. Introduction The electric grid is going through a major change. Smart grid initiatives around the world are pushing the grid into a more ro- bust, dynamic and open system which will bring consumers and their devices directly into the management realm of the grid. This integration of IT provides a great opportunity for improving and enhancing DSM and DR programs. Such programs can be improved with intelligence, pervasive device management or renewable integration for increased throughput. Various DSM planning strat- egies have been proposed for smart grids but to implement such planning methods the knowledge of the amount of energy demand at house level is a must. This requires a short term load forecast for houses, and in some cases even devices. To this end in this paper we propose two unique concepts for short term load forecasting of houses through which accuracy for forecasting loads of houses can increase by as much as 50%. This provides an important cog in our proposed smart grid architecture for demand side manage- ment discussed in Appendix A. Forecasting for larger loads such as city or the entire grid has been achieved with relatively high accuracy [1]. But for smaller populations such as a building, or a micro-grid the dynamics change so drastically that standard STLF tools require certain re- adjustments [4]. For even smaller consumer group, such as individ- ual houses, the volatility in dynamics is even more pronounced as can be seen from discussion in Section 2. To forecast for such sys- tem we need to look at the STLF modeling, tools, and data. There are two pertinent questions to engineer these re-adjustments for STLF for individual houses in a system that we answer in this pa- per. First is that can we forecast energy load using the existing short term load forecasting model? Second question is that is the knowledge used for existing forecasting models sufficient? Kim and Shcherbakova point out at the lack of data about user as one of the major reason failure for DSM and DR programs [20]. But our initial results showed that simple correlation between house load and house characteristics is weak as shown in Fig. 1. The strongest influence on demand is weather. This was observed on anthropologic and structural data collected from 205 houses in Eskistuna, Sweden. However, we observed a subtle relationship be- tween user characteristics and consumption. Our contributions use this subtle relationship between house statistics and consumption. The relationship between energy use and occupant and building characteristic is such that on a single house level it is insignificant. This can be observed from results 0306-2619/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2012.02.027 Abbreviations: STLF, short term load forecasting; STMLF, short term multiple loads forecasting; ANNs, artificial neural networks; DR, demand response; MSE, mean squared error; MLR, multiple linear regression; SVM, support vector machines; AMR, automatic meter reading; Cid, consumer id; AI, artificial intelligence; ARIMA, autoregressive integrated moving average; GARCH, generalized autoregressive conditional heteroske dasticity; Var, variance; Acc, accuracy. ⇑ Corresponding author. E-mail addresses: fahadjaved@lums.edu.pk (F. Javed), fredrik.wallin@mdh.se (F. Wallin), iana.vassileva@mdh.se (I. Vassileva), erik.dahlquist@mdh.se (E. Dahlquist). Applied Energy xxx (2012) xxx–xxx Contents lists available at SciVerse ScienceDirect Applied Energy journal homepage: www.elsevier.com/locate/apenergy Please cite this article in press as: Javed F et al. Forecasting for demand response in smart grids: An analysis on use of anthropologic and structural data and short term multiple loads forecasting. Appl Energy (2012), http://dx.doi.org/10.1016/j.apenergy.2012.02.027