The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
This paper presents a highly adaptable and robust short-term load forecasting algorithm developed using hybrid modeling techniques. Adaptive general exponential smoothing augmented with power spectrum analysis is proposed to account for the changing base load component. The algorithm includes an adaptive autoregressive modeling technique enhanced with partial autocorrelation analysis to model the random component of the load. The Akaike information criterion is employed to guarantee model parsimony. The weighted recursive least square estimate algorithm with variable forgetting factors is applied to estimate the model parameters. A nonlinear weather-sensitive model is used to represent the influence of weather changes on energy consumption. Simulations performed using historical load data from two large utilities revealed that the proposed approach produces highly accurate forecasting and is especially attractive for online applications with little human intervention. Details of the approach and test results are included in the paper.