Abstract : This paper presents a methodology for short-term load forecasting based on genetic algorithm feature selection and artificial neural network modeling. A feedforward artificial neural network is used to model the 24-h ahead load based on past consumption, weather and stock index data. A genetic algorithm is used in order to find the best subset of variables for modeling. Three datasets of different geographical locations, encompassing areas of different dimensions with distinct load profiles are used in order to evaluate the methodology. The developed approach was found to generate models achieving a minimum mean average percentage error under 2 %. The feature selection algorithm was able to significantly reduce the number of used features and increase the accuracy of the models.
Luis M. Camarinha-Matos; António J. Falcão; Nazanin Vafaei; Shirin Najdi. 7th Doctoral Conference on Computing, Electrical and Industrial Systems (DoCEIS), Apr 2016, Costa de Caparica, Portugal. IFIP Advances in Information and Communication Technology, AICT-470, pp.485-493, 2016, Technological Innovation for Cyber-Physical Systems. 〈10.1007/978-3-319-31165-4_45〉
https://hal.inria.fr/hal-01438275
Contributeur : Hal Ifip
<>
Soumis le : mardi 17 janvier 2017 - 15:38:43
Dernière modification le : mardi 17 janvier 2017 - 15:49:43
Document(s) archivé(s) le : mardi 18 avril 2017 - 15:04:25
Joaquim Viegas, Susana Vieira, Rui Melício, Victor Mendes, João Sousa. GA-ANN Short-Term Electricity Load Forecasting. Luis M. Camarinha-Matos; António J. Falcão; Nazanin Vafaei; Shirin Najdi. 7th Doctoral Conference on Computing, Electrical and Industrial Systems (DoCEIS), Apr 2016, Costa de Caparica, Portugal. IFIP Advances in Information and Communication Technology, AICT-470, pp.485-493, 2016, Technological Innovation for Cyber-Physical Systems. 〈10.1007/978-3-319-31165-4_45〉. 〈hal-01438275〉
