Evaluation of optimization techniques in predicting optimum moisture content reduction in drying potato slices

• Main Authors: Chijioke Elijah Onu, Philomena K. Igbokwe, Joseph T. Nwabanne, Charles O. Nwajinka, Paschal E. Ohale
• Format: Article
• Language: English
• Published: KeAi Communications Co., Ltd. 2020-01-01
• Series: Artificial Intelligence in Agriculture
• Subjects: Moisture content; Potato; Adaptive neuro fuzzy inference systems; Artificial neural network; Response surface methodology
• Online Access: http://www.sciencedirect.com/science/article/pii/S2589721720300118

The use of artificial intelligence models in predicting the moisture content reduction in the drying of potato (Ipomoea batata) slices was the focus of this work. The models used were adaptive neuro fuzzy inference systems (ANFIS), artificial neural network (ANN) and response surface methodology (RSM). The parameters considered were drying time, drying air speed and temperature. The capability and sensitivity analysis of the three models were evaluated using the correlation coefficient (R2) and some statistical error functions such as the average relative error (ARE), root mean square error (RMSE), Hybrid Fractional Error Function (HYBRID) and absolute average relative error (AARE). The result showed that the three models demonstrated significant predictive behaviour with R2 of 0.998, 0.997 and 0.998 for ANFIS, ANN and RSM respectively. The calculated error functions of ARE (RSM = 1.778, ANFIS = 1.665 and ANN = 4.282) and RMSE (RSM = 0.0273, ANFIS = 0.0282 and ANN = 0.1178) suggested good harmony between the experimental and predicted values. It was concluded that though the three models gave adequate predictions that were in good agreement with the experimental data, the RSM and ANFIS gave better model prediction than ANN.