Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/B4A5A2ED-BD87-404D-AFED-827F4E111C16 | ||
| Year | 2024 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Nikiforos Bouros, "Forecasting natural gas demand in Greece using the adaptive neuro-fuzzy inference system - ANFIS", Diploma Work, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2024 https://doi.org/10.26233/heallink.tuc.100543 | ||
| Appears in Collections |
The energy sector is the driving force of economies worldwide, as it is not only limited to fueling industrial development, thus enabling technological evolution and innovation, but at the same time, it sustains the needs of daily life as a whole. For this reason, forecasts in the energy sector are deemed necessary, with the present paper focusing on the forecasting of natural gas demand. In order to develop an accurate natural gas forecasting model for Greece, the application of the adaptive neuro-fuzzy inference system (ANFIS), a flexible and easy-to-use technique that integrates neural networks with fuzzy logic principles, is considered. Initially, similar studies are reviewed and the mechanisms of creation, search, extraction, processing and distribution of natural gas are presented, with special reference to the Greek market and its main players. The theoretical background of forecasting models follows, with reference to time series, fuzzy logic and neuro-fuzzy systems, with an emphasis on ANFIS method and its architecture. Using data of the daily import of natural gas from 2008 and over fifteen years in Greece, from the database of DEPA Commercial, the analysis is carried out using the MATLAB software. Sequential parameter testing and method iteration with the ANFIS model shows low root mean square error (RMSE) and lower errors than the Autoregressive (AR) and ARMA (Autoregressive Moving Average) models, thus highlighting ANFIS as a satisfactorily effective tool in natural gas demand forecasting.
Copyright © DIAS 2013
