АУТОР(И) / AUTHOR(S): Stefan ČUBONOVIĆ, Aleksandar RANKOVIĆ, Marko KRSTIĆ
DOI: 10.46793/EEE24-2.47C
САЖЕТАК / ABSTRACT:
U ovom istraživačkom radu sprovedena je analiza dva tipa veštačkih neuronskih mreža (Artificial Neural Network – ANN) za predikciju snage na izlazu hidroelektrane (HE). Prvi tip je veštačka neuronska mreža sa jednosmernim prostiranjem signala (Feedforward Artificial Neural Networks – FF-ANN), dok je drugi tip rekurentna neuronska mreža (Recurrent Neural Network – RNN). Detaljno su analizirani koraci koji su preduzeti u procesu implementacije neuronskih mreža za ovu svrhu, od prikupljanja i pripreme podataka do treniranja, evaluacije i analize rezultata. Kao ulazni podaci korišćeni su neto pad, protok vode kroz turbinu, kota gornje vode, kota donje vode i temperatura ulazne rashladne vode. Na osnovu koeficijenta korelacije pojedinih veličina iz ulaznog sloja sa izlaznom snagom izvršena implementacija novih ANN. Rezultati dobijeni ovim varijacijama su sistematski analizirani kako bi se postiglo što preciznije modelovanje, sa akcentom na dinamičke promene u protoku vode kroz turbinu.
КЉУЧНЕ РЕЧИ / KEYWORDS:
veštačka neuronska mreža; FF-ANN; RNN; hidroelektrana; izlazna snaga
ЛИТЕРАТУРА / REFERENCES:
- Ranković, A., Maksimović, B.M., Sarić, A.T., Lukič, U. ANN-based correlation of measurements in micro-grid state estimation, International Transactions on Electrical Energy Systems, Vol. 25, No. 10, pp. 2181-2202, 2015. https://doi.org/10.1002/etep.1956
- Warnick, C.C. Hydropower Engineering, Prentice Hall, 1984.
- Thaeer, H.A., Mohd, H.B.S.N., Abdalla, A. Prediction of small hydropower plant power production in Himreen Lake dam (HLD) using artificial neural network, Alexandria Engineering Journal, 57, No. 1, pp. 211-221, 2018. https://doi.org/10.1016/j.aej.2016.12.011
- Artificial Neural Networks: An introduction to ANN theory and practice, in: Braspenning, P.J., Thuijsman, F., Weijters, A.J.M.M. (Ed.), Lecture Notes in Computer Science, 931, Springer-Verlag, Berlin Heidelberg, 1995.
- McGonagle, J., Garcia, J.A., Mollick, S. Feedforward Neural Networks, https://brilliant.org/wiki/feedforward-neural-networks/ [preuzeto 21.11.2023]
- Hagan, M.T., Demuth, H.B., Bale, M. Neural Network Design, PWS Pub. Co., Boston, USA, 1996.
- Lawrence, S., Giles, C.L., Tsoi, A.C. Lessons in neural network training: overfitting may be harder than expected, in Proc. Proceedings of the Fourteenth National Conference on Artificial Intelligence and Ninth Innovative Applications of Artificial Intelligence Conference, Providence, Rhode Island, pp. 540-545, 1997. https://clgiles.ist.psu.edu/papers/AAAI-97.overfitting.hard_to_do.pdf [preuzeto 21.11.2023]
- Hansen, L.K., Salamon, P. Neural network ensembles, IEEE Transactions on Pattern Analysis and Machine Intelligence, 12, No. 10, pp. 993-1001, 1990. https://doi.org/10.1109/34.58871
- Hochreiter, S. The vanishing gradient problem during learning recurrent neural nets and problem solutions, International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, Vol. 06, No. 02, pp. 107-116, 1998. https://doi.org/10.1142/S0218488598000094
- Đurić, M., Đurišić, Ž., Čukarić, A., Ilić, V. Elektrane, AGM, Beograd, 2014.
- Hagan, M.T., Demuth, H.B., Bale, M. Neural Network Design, PWS Pub., Boston, USA, 1996.
- Karayiannis, N.B., Venetsanopoulos, A.N. Artificial neural networks: learning algorithms, performance evaluation, and applications, Springer, USA, 1993.