Long-term energy load forecasting using Auto-Regressive and approximating Support Vector Regression

摘要: Several techniques can be retrieved in literature, which cope with the problem of energy load forecasting short-term framework, that is hours up to few days ahead. However, order properly schedule operative conditions including purchasing and generation, fuel supply, infrastructure development maintenance, a long-term prediction crucial importance. In this generalization more complex usually characterized by scarce performance. paper, we present an innovative method, exploits Savitzky-Golay filter Support Vector Regression algorithm reliably predict consumption framework: extrapolation ability our proposal validated on real-world problem.