A New Strategy of Load Forecasting Technique for Smart Grids.

Smart grids, or intelligent electricity grids which employ modern IT/communication/control technologies is becoming a global trend nowadays. Load forecasting is a vital part for the power system planning as well as operation to provide intelligence to energy management within smart grids. There are several methods that are used to improve the load forecasting accuracy. Those techniques differ in the mathematical formulation as well as the features used in each formulation. Electric loads may have bad data called outliers' data. As well as, those loads are affected by several features such as weather, season, and events; with some of them having more impact than the others. Therefore, the filtration process is an important stage before forecasting process in order to achieve a faster and more accurate forecasting process than others. This paper intends review different widely used load forecasting techniques and provides a comparative study to show the effectiveness of each of them with simulations in MATLAB. Experimental results have shown that modify/extended logistic model provides the best performance as it gives the minimal error. In split of this advantage, this method doesn't achieve fast prediction as well as optimal minimization of error value. Therefore, a new prediction strategy using data mining techniques will be introduced to give more accurate and fast predicted load value. This strategy is presented as two stages; filtration stage as well as forecasting stage using advanced data mining techniques. Keywords—Smart grids; load forecasting; time series methods; econometric methods; data mining techniques.

[1]  Sunita Beniwal,et al.  Classification and Feature Selection Techniques in Data Mining , 2012 .

[2]  W. Marsden.  I and J , 2012 .

[3]  Nima Amjady,et al.  Short-term hourly load forecasting using time-series modeling with peak load estimation capability , 2001 .

[4]  Clara Pizzuti,et al.  Fast Outlier Detection in High Dimensional Spaces , 2002, PKDD.

[5]  S. M. El-Debeiky,et al.  Long-Term Load Forecasting for Fast-Developing Utility Using a Knowledge-Based Expert System , 2002, IEEE Power Engineering Review.

[6]  Ahmet Serdar Yilmaz,et al.  Long Term Energy Consumption Forecasting Using Genetic Programming , 2008 .

[7]  Hassan Soltan,et al.  A methodology for Electric Power Load Forecasting , 2011 .

[8]  M. S. Kandil,et al.  Overview and comparison of long-term forecasting techniques for a fast developing utility: part I , 2001 .

[9]  Mohsen Kalantar,et al.  Different Methods of Long-Term Electric Load Demand Forecasting a Comprehensive Review , 2011 .

[10]  M. Kalantar,et al.  Methods for long-term electric load demand forecasting; a comprehensive investigation , 2008, 2008 IEEE International Conference on Industrial Technology.

[11]  Milena Stankovic,et al.  Mid-term load forecasting using recursive time series prediction strategy with support vector machines , 2010 .

[12]  Arunesh Kumar Singh,et al.  An Overview of Electricity Demand Forecasting Techniques , 2013 .

[13]  Engr,et al.  Comparison of Conventional and Modern Load Forecasting Techniques Based on Artificial Intelligence and Expert Systems , 2011 .