Machine Learning Approaches for One-Day Ahead Soil Temperature Forecasting

Authors : Mehmet BİLGİLİ, Şaban ÜNAL, Aliihsan ŞEKERTEKİN, Cahit GÜRLEK

Pages : 221-238

Doi:10.15832/ankutbd.997567

View : 23 | Download : 8

Publication Date : 2023-01-31

Article Type : Research Paper

Abstract :Present study investigates the capabilities of six distinct machine learning techniques such as ANFIS network with fuzzy c-means insert ignore into journalissuearticles values(ANFIS-FCM);, grid partition insert ignore into journalissuearticles values(ANFIS-GP);, subtractive clustering insert ignore into journalissuearticles values(ANFIS-SC);, feed-forward neural network insert ignore into journalissuearticles values(FNN);, Elman neural network insert ignore into journalissuearticles values(ENN);, and long short-term memory insert ignore into journalissuearticles values(LSTM); neural network in one-day ahead soil temperature insert ignore into journalissuearticles values(ST); forecasting. For this aim, daily ST data gathered at three different depths of 5 cm, 50 cm, and 100 cm from the Sivas meteorological observation station in the Central Anatolia Region of Turkey was used as training and testing datasets. Forecasting values of the machine learning models were compared with actual data by assessing with respect to four statistic metrics such as the mean absolute error, root mean square error insert ignore into journalissuearticles values(RMSE);, Nash−Sutcliffe efficiency coefficient, and correlation coefficient insert ignore into journalissuearticles values(R);. The results showed that the ANFIS-FCM, ANFIS-GP, ANFIS-SC, ENN, FNN and LSTM models presented satisfactory performance in modeling daily ST at all depths, with RMSE values ranging 0.0637-1.3276, 0.0634-1.3809, 0.0643-1.3280, 0.0635-1.3186, 0.0635-1.3281, and 0.0983-1.3256 °C, and R values ranging 0.9910-0.9999, 0.9903-0.9999, 0.9910-0.9999, 0.9911-0.9999, 0.9910-0.9999 and 0.9910-0.9998 °C, respectively.
Keywords : ANFIS, Daily soil temperature, LSTM, Elman neural network ENN, , Feed forward neural network