Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/A1AEB9B5-88B0-48F4-924C-09DFD53C2767 | ||
| Year | 2024 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Vasileios Petikas, "Neuro-fuzzy techniques for energy production forecasting", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2024 https://doi.org/10.26233/heallink.tuc.100406 | ||
| Appears in Collections |
The increased demand for clean and sustainable energy sources, combined with concerns about climate change and the reduction of fossil fuels, has led to the promotion of renewable energy sources (RES) as a key objective of global energy policy. The shift towards renewable sources, such as solar, wind and hydroelectric power, is considered necessary to reduce greenhouse gas emissions, enhance energy security and achieve sustainable development.In this context, this thesis examines the application of neuro-fuzzy techniques, in particular the Adaptive Neuro-Fuzzy Inference System (ANFIS), for the estimation of energy production from Photovoltaics.The objective of the study is to address the challenges posed by the increasing integration of RES in the power grid, which requires accurate and reliable estimation methods to ensure grid stability and efficiency. Forecasting the performance of RES is critical due to their volatile and unpredictable nature. Power grids require stable and reliable energy supplies, and accurate estimation can reduce the risks and costs associated with energy storage or replenishment.The methodology involves the collection and processing of historical energy data from PV systems, followed by the development and training of the ANFIS model. This model is compared to traditional statistical methods, such as Autoregressive (AR) and Autoregressive Moving Average (ARMA) models, as well as an Artificial Neural Network (ANN), and a Type-2 Fuzzy Inference System optimized using Particle Swarm Optimization (PSO).The results show that ANFIS outperforms the selected models in terms of prediction accuracy and adaptability to changing environmental conditions. The study concludes that integrating intelligent systems such as ANFIS can significantly improve the accuracy and reliability of RES generation estimates, thus facilitating their efficient and stable integration into the energy grid.This research contributes to the energy sector and provides a comprehensive comparative analysis of different estimation models. Furthermore, it highlights the advantages of neuro-fuzzy techniques in managing the variability and uncertainty associated with renewable energy sources. The findings demonstrate the potential of ANFIS as a powerful tool for energy production estimation, thus promoting the development and adoption of renewable energy-related technologies.
Copyright © DIAS 2013
