Pappas et al. / J Zhejiang Univ Sci A 2008 9(12):1724-1730 1724 Adaptive load forecasting of the Hellenic electric grid S. Sp. PAPPAS 1 , L. EKONOMOU †‡2 , V. C. MOUSSAS 3 , P. KARAMPELAS 2 , S. K. KATSIKAS 4 ( 1 Department of Information and Communication Systems Engineering, University of the Aegean, Samos 83200, Greece) ( 2 Information Technology Faculty, Hellenic American University, Athens 10680, Greece) ( 3 School of Technological Applications, Technological Educational Institute of Athens, Egaleo 12210, Greece) ( 4 Department of Technology Education and Digital Systems, University of Piraeus, Piraeus 18532, Greece) † E-mail: leekonom@gmail.com Received Jan. 16, 2008; revision accepted May 26, 2008; CrossCheck deposited Nov. 10, 2008 Abstract: Designers are required to plan for future expansion and also to estimate the grid’s future utilization. This means that an effective modeling and forecasting technique, which will use efficiently the information contained in the available data, is required, so that important data properties can be extracted and projected into the future. This study proposes an adaptive method based on the multi-model partitioning algorithm (MMPA), for short-term electricity load forecasting using real data. The grid’s utilization is initially modeled using a multiplicative seasonal ARIMA (autoregressive integrated moving average) model. The proposed method uses past data to learn and model the normal periodic behavior of the electric grid. Either ARMA (autoregressive moving average) or state-space models can be used for the load pattern modeling. Load anomalies such as unexpected peaks that may appear during the summer or unexpected faults (blackouts) are also modeled. If the load pattern does not match the normal be- havior of the load, an anomaly is detected and, furthermore, when the pattern matches a known case of anomaly, the type of anomaly is identified. Real data were used and real cases were tested based on the measurement loads of the Hellenic Public Power Cooperation S.A., Athens, Greece. The applied adaptive multi-model filtering algorithm identifies successfully both normal periodic behavior and any unusual activity of the electric grid. The performance of the proposed method is also compared to that produced by the ARIMA model. Key words: Adaptive multi-model filtering, ARIMA, Load forecasting, Measurements, Kalman filter, Order selection, Seasonal variation, Parameter estimation doi:10.1631/jzus.A0820042 Document code: A CLC number: TM714 INTRODUCTION Accurate and fast forecasting of the electricity load is an important factor in various fields, ranging from power system real-time control, which ensures system safety, to reliable and economical operation. One day ahead load prediction and one week ahead load prediction, which are also referred to as short term predictions, are very useful in the preparation of the individual daily-schedule plan and weekly- schedule plan. This procedure includes the decision as to which units are going to contribute to the power generation, the coordination between hydroelectric power and electricity generated from any heat source, power interchange liaison, the economical distribu- tion of load power, equipment inspection, etc. (Rajesh, 1997). The problem of load forecasting has been studied extensively during recent decades. Some of the pro- posed techniques make use of time series analysis using ARMA or ARIMA models (Nowicka-Zagrajek and Weron, 2002; Contreras et al., 2003; Huang and Shih, 2003; Zhou et al., 2004; Espinoza et al., 2005; di Caprio et al., 2006; Sisworahardjo et al., 2006; Ediger and Akar, 2007). Other algorithms achieve load forecasting by adopting piecewise linear meth- ods in order to relate the electricity load to weather variables (Haida and Muto, 1994; Charytoniuk et al., 1999). Artificial neural networks (ANNs) have been extensively used either alone or combined with Journal of Zhejiang University SCIENCE A ISSN 1673-565X (Print); ISSN 1862-1775 (Online) www.zju.edu.cn/jzus; www.springerlink.com E-mail: jzus@zju.edu.cn ‡ Corresponding author