Estimation on River Water Flow Rate Using Long Short-term Memory (LSTM) Neural Network and Adaptive Neuro-fuzzy Inference Systems (ANFIS)

Yıl 2025, Cilt: 41 Sayı: 1, 20 - 42, 30.04.2025

Akın İlhan

Öz

The forecasting of river water flow rate (RWFR) plays a prominent role in planning and constructing of new hydraulic dams, or running the ones that were formerly built. This study suggests algorithms of machine learning to predict future water flow rate values for river flow. Namely, estimation models were advanced according to the past time-series RWFR to obtain future RWRF values. Accordingly, long short-term memory (LSTM) neural network, adaptive neuro-fuzzy inference system (ANFIS) with fuzzy c-means (FCM), were advanced for the aim of RWFR predictions. A measurement station (MS), settled at the border of Türkiye and Bulgaria, named as Svilengrad MS was selected on the Maritsa River, as the study region. Accordingly, it was concluded that FCM algorithm of ANFIS have generated better results compared with respect to the LSTM algorithm. The comparisons of the data estimations according to the real observed water flow values were accomplished depending on the statistical error values including mean absolute error (MAE), root mean square error (RMSE), else the correlation coefficient (R). Eventually, it was concluded and shown that the superior model of FCM have generated those statistical accuracy values, respectively to correspond 3.13 m3/s MAE, 4.90 m3/s RMSE, and 0.9978 R, among the total of 49 tested models using FCM and LSTM.

Anahtar Kelimeler

Fuzzy C-means (FCM), Long Short-term Memory (LSTM), Adaptive Neuro Fuzzy Inference System (ANFIS)

Uzun Kısa Süreli Bellek (LSTM) Sinir Ağı ve Uyarlanabilir Nöro Bulanık Çıkarım Sistemleri (ANFIS) Kullanılarak Nehir Suyu Akış Hızının Tahmini

Öz

Nehir suyu akış hızının (RWFR) tahmini, yeni su barajlarının planlanması ve inşa edilmesinde veya daha önce inşa edilmiş olanların işletilmesinde çok önemli bir rol oynar. Bu çalışma, nehir akışının gelecekteki su akış hızı değerlerini tahmin etmek için makine öğrenme algoritmaları önermektedir. Yani tahmin modelleri, gelecekteki RWRF değerlerini elde etmek için geçmiş zaman serisi RWFR'ye dayalı olarak geliştirildi. Buna göre, uzun kısa süreli bellek (LSTM) sinir ağı, bulanık c-ortalamalara (FCM) sahip uyarlanabilir nöro-bulanık çıkarım sistemi (ANFIS), RWFR tahminleri amacıyla geliştirildi. Çalışma bölgesi olarak Meriç Nehri üzerinde Türkiye-Bulgaristan sınırında yer alan ve Svilengrad MS olarak adlandırılan ölçüm istasyonu (MS) seçilmiştir. Buna göre, ANFIS'in FCM algoritmasının LSTM algoritmasına kıyasla daha iyi sonuçlar ürettiği sonucuna varılmıştır. Veri tahminlerinin gerçek gözlemlenen su akış değerlerine göre karşılaştırılması, ortalama mutlak hata (MAE), ortalama karekök hata (RMSE) ve korelasyon katsayısı (R) dahil olmak üzere istatistiksel hata değerlerine bağlı olarak gerçekleştirildi. Neticede, FCM'nin en iyi modelinin, FCM ve LSTM kullanan toplam 49 test modeli arasında, sırasıyla, 3,13 m3/s MAE, 4,90 m3/s RMSE ve 0,9978 R'ye karşılık gelen istatistiksel doğruluk değerlerini ürettiği sonucuna varıldı ve gösterildi.

Anahtar Kelimeler

Uyarlanabilir Nöro Bulanık Çıkarım Sistemi (ANFIS), Bulanık C-ortalamaları (FCM), Uzun Kısa Süreli Bellek (LSTM)

Kaynakça

• Lurwan, S. M., Mariun, N., Hizam, H., Radzi, M. A. M., Zakaria, A. 2014. Predicting Power Output of Photovoltaic Systems with Solar Radiation Model. 2014 IEEE International Conference on Power and Energy (PECon), 1-3 December, Kuching, Malaysia, 304-308.
• Oruç, E. N. 2021. Yapay sinir ağları kullanılarak kısa süreli güneş enerjisi tahmini. İstanbul Teknik Üniversitesi, Lisansüstü Eğitim Enstitüsü, Yüksek Lisans Tezi, İstanbul.
• Theocharides, S., Makrides, G., Georghiou, G. E., Kyprianou, A. 2018. Machine Learning Algorithms for Photovoltaic System Power Output Prediction. IEEE International Energy Conference (ENERGYCON), 3-7 June, Limassol, Cyprus, 1-6.
• Scott, C., Ahsan, M., Albarbar, A. 2023. Machine Learning for Forecasting a Photovoltaic (PV) Generation System. Energy, 278, 127807.
• Gaboitaolelwe, J., Zungeru, A. M., Yahya, A., Lebekwe, C. K., Vinod, D. N., Salau, A. O. 2023. Machine Learning Based Solar Photovoltaic Power Forecasting: A Review and Comparison. IEEE Access, 11, 40820-40845.
• Liu, H., Tian, H., Li, Y. 2015. An EMD-recursive ARIMA Method to Predict Wind Speed for Railway Strong Wind Warning System. Journal of Wind Engineering and Industrial Aerodynamics, 141, 27-38.
• Kavasseri, R. G., Seetharaman, K. 2009. Day-ahead Wind Speed Forecasting Using f-ARIMA Models. Renewable Energy, 34(5), 1388-1393.
• Zhang, J., Yan, J., Infield, D., Liu, Y., Lien, F. 2019. Short-term Forecasting and Uncertainty Analysis of Wind Turbine Power Based on Long Short-term Memory Network and Gaussian Mixture Model. Applied Energy, 241, 229-244.
• Zaytar, M. A., Amrani, C. E. 2016. Sequence to Sequence Weather Forecasting with Long Short-term Memory Recurrent Neural Networks. International Journal of Computer Applications, 143(11), 7-11.
• Prabha, P. P., Vanitha, V., Resmi, R. 2019. Wind Speed Forecasting Using Long Short Term Memory Networks. 2nd International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT), 5-6 July, Kannur, Kerala, India, 1310-1314.
• Benmouiza, K., Cheknane, A. 2019. Clustered ANFIS Network Using Fuzzy C-means, Subtractive Clustering, and Grid Partitioning for Hourly Solar Radiation Forecasting. Theoretical and Applied Climatology, 137, 31-43.
• Tumse, S., Bilgili, M., Sekertekin, A., Unal, S., Sahin, B. 2023. Comparison and Evaluation of Machine Learning Approaches for Estimating Heat Index Map in Türkiye. Neural Computing and Applications, 35(21), 15721-15742.
• Tumse, S., Bilgili, M., Sahin, B. 2022. Estimation of Aerodynamic Coefficients of a Non-slender Delta Wing Under Ground Effect Using Artificial Intelligence Techniques. Neural Computing and Applications, 34(13), 10823-10844.
• Tumse, S., Ilhan, A., Bilgili, M. 2022. Estimation of Wind Turbine Output Power Using Soft Computing Models. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 44(2), 3757-3786.
• Hewett, R., Leuchner, J., Carvalho, M. 2001. From Climate History to Prediction of Regional Water Flows with Machine Learning. Institute of Electrical and Electronics Engineers (IEEE), 292-297.
• Tayfur, G., Singh, V. P., Moramarco, T., Barbetta, S. 2018. Flood Hydrograph Prediction Using Machine Learning Methods. Water (Multidisciplinary Digital Publishing Institute), 986, 1-13.
• Flake, J. T. 2007. Application of the relevance vector machine to canal flow prediction in the Sevier River basin. Utah State University, Electrical and Computer Engineering, Master of Science Thesis, Utah.
• Farhadi, H., Zahiri, A., Hashemi, M. R., Esmaili, K. 2019. Incorporating a Machine Learning Technique to Improve Open-channel Flow Computations. Neural Computing and Applications, 31, 909-921.
• Sahraei, A., Chamorro, A., Kraft, P., Breuer, L. 2021. Application of Machine Learning Models to Predict Maximum Event Water Fractions in Streamflow. Front Water, 3, 1-21.
• Il Kim, H., Kim, B. H. 2020. Flood Hazard Rating Prediction for Urban Areas Using Random Forest and LSTM. Water Resources and Hydrologic Engineering, 24, 3884-3896.
• Yaseen, Z. M., Jaafar, O., Deo, R. C., Kisi, O., Adamowski, J., Quilty, J., El-Shafie, A. 2016. Stream-flow Forecasting Using Extreme Learning Machines: A Case Study in a Semi-arid Region in Iraq. Journal of Hydrology, 542, 603-614.
• Xiao, L., Zhong, M., Zha, D. 2022. Runoff Forecasting Using Machine-learning Methods: Case Study in the Middle Reaches of Xijiang River. Front Big Data, 4, 1-11.
• Mathworks, 2020a. Fuzzy C-means Clustering. https://www.mathworks.com/help/deeplearning/ug/long-short-term-memory-networks.html (Access Date: 05.05.2024).
• Mathworks, 2020b. Long Short-term Memory Networks. https://www.mathworks.com/help/deeplearning/ug/long-short-term-memory-networks.html (Access Date: 07.06.2024).
• Jang, J. R. 1993. ANFIS: Adaptive-network-based Fuzzy Inference System. IEEE Transactions on Systems, Man, and Cybernetics, 23(3), 665-685.
• Chandy, K. M., Taylor, S. 1992. An Introduction to Parallel Programming. MA: Jones and Bartlett, Boston.
• Liu, R., Liu, L. 2019. Predicting Housing Price in China Based on Long Short-term Memory Incorporating Modified Genetic Algorithm. Soft Computing, 23, 11829-11838.
• Ilhan, A. 2023. Forecasting of Volumetric Flow Rate of Ergene River Using Machine Learning. Engineering Applications of Artificial Intelligence, 105983.
• Devlet Su Isleri (DSI). 2024. 11th regional directorate of DSI of Türkiye. http://edirnenehir.dsi.gov.tr/ (Access Date: 25.09.2024).