Small Island Developing States (SIDS), like Mauritius, share similar sustainable development challenges inherent to their characteristics. Growth in the global energy demand and fears of energy supply disruptions, have triggered much debate geared towards the necessity for sustainable energy planning. Accurate forecasting of future electricity demand is an essential input to this process. Such forecasts are also important in regional or national power system strategy Management.
Non linearity of the factors adds complexity to the electricity load forecasting process. Statistical learning theory, in the form of Support Vector Machines (SVMs), have been used successfully to tackle nonlinear regression and time series problems. However application to the electricity demand forecasting problem with focus on SIDS’ characteristics is lacking. This paper focuses on the application of SVMs to forecast electricity demand of a SIDS member, Mauritius. A two years ahead forecast, for 2008 and 2009, was derived using monthly time series data from years 1996 to 2007. The inputs considered were historical electricity demand and prices, temperature, humidity, population and GDP.
|Keywords:||Electricity Demand Forecasting, Electricity Load Forecasting, Support Vector Machines, Statistical Learning Theory|
Senior Lecturer, School of Business Informatics and Software Engineering, University of Technology Mauritius, Pointe aux Sables, Mauritius
Research Student, Research Cell, School of Sustainable Development Science, University of Technology Mauritius, Pointe Aux Sables, Mauritius
There are currently no reviews of this product.Write a Review