Drying characteristics of yam slices (Dioscorea rotundata) in a convective hot air dryer: application of ANFIS in the prediction of drying kinetics.

John O Ojediran,Nana E George,Abiola J Adeyi,Oladayo Adeyi,Abiola F Olaniran,Clinton E Okonkwo,Adeniyi T Olayanju

doi:10.1016/j.heliyon.2020.e03555

John O Ojediran, Nana E George + Show 5 more

Open Access

https://doi.org/10.1016/j.heliyon.2020.e03555

Copy DOI

Abstract

This study applied Adaptive Neuro-Fuzzy Inference System (ANFIS) to predict the moisture ratio (MR) during the drying process of yam slices (Dioscorea rotundata) in a hot air convective dryer. Also the effective diffusivity, activation energy, and rehydration ratio were calculated. The experiments were carried out at three (3) drying air temperatures (50, 60, and 70 °C), air velocities (0.5, 1, and 1.5 m/s), and slice thickness (3, 6, and 9 mm), and the obtained experimental data were used to check the usefulness of ANFIS in the yam drying process. The result showed efficient applicability of ANFIS in predicting the MR at any time of the drying process with a correlation value (R2) of 0.98226 and root mean square error value (RMSE) of 0.01702 for the testing stage. The effective diffusivity increased with an increase in air velocity, air temperature, and thickness and the values (6.382E -09 to 1.641E -07 m2/s). The activation energy increased with an increase in air velocity, but fluctuate within the air temperatures and thickness used (10.59–54.93 KJ/mol). Rehydration ratio was highest at air velocity×air temperature×thickness (1.5 m/s×70 °C × 3 mm), and lowest at air velocity × air temperature×thickness (0.5 m/s×70 °C × 3 mm). The result showed that the drying kinetics of Dioscorea rotundata existed in the falling rate period. The drying time decreased with increased temperature, air velocity, and decreased slice thickness. These established results are applicable in process and equipment design, analysis and prediction of hot air convective drying of yam (Dioscorea rotundata) slices.

Highlights

Yam (Dioscorea spp.) has been identified as one of the most important food crops for a wide range of tropical countries including Nigeria, Ghana, Togo, Burkina Faso, Cote d’Ivoire, with over 600 species, in which only a few are cultivated for food purpose (Olatoye and Arueya, 2019)
The partitioning of the experimental drying data into various data sets in this study was represented in Figure 4, where the training data set are represented with an 'o' sign and checking data set are represented with a 'þ' sign
This spread was important so that Adaptive Neuro-Fuzzy Inference System (ANFIS) can understand the dynamics that exist in data during training and testing operations

Summary

Introduction

Yam (Dioscorea spp.) has been identified as one of the most important food crops for a wide range of tropical countries including Nigeria, Ghana, Togo, Burkina Faso, Cote d’Ivoire, with over 600 species, in which only a few are cultivated for food purpose (Olatoye and Arueya, 2019). Drying has been regarded by humans as probably the most important and oldest food preservation method and it entails a complex thermal process in which simultaneous heat and mass transfer occur (Ojediran and Raji, 2010; Doymaz, 2011) It is a process of moisture reduction in agricultural products to extend its shelf life (Abbaspour-Gilandeh et al, 2019). The applicability of the black-box modeling method called soft computing technique is becoming popular partly because of their high accuracies and ease of use They are the best fit for the situation where exact mathematical models or information is difficult to establish for the dynamics of a system. This is a result of difficulties associated with the interpretation of artificial intelligence language for the realistic needs of drying community or probably due to the availability of simpler but less accurate alternatives such as Proportional Integral Derivative controller

Methods

Results

Conclusion