Vol 8 Issue 1 March 2021-August 2021
Uwem Ekwere Inyang, Jock Asanja Alexander, Nkwocha Promise Chibuzo
Abstract: This paper presents a model developed using artificial neural network (ANN) to predict the drying parameters of cocoyam (Xanthosoma sagittifolium L.) slices dried using an oven. To obtain data for developing the model, 311 drying experiments were performed on cocoyam slices of thickness ranging from 2 mm to 8 mm subjected to varying temperatures ranging from 40°C – 70°C in 10°C increments. The input variables for the model include: cocoyam slice thickness, drying time and the drying air temperature while the outputs were the moisture content and drying rate. After series of simulation runs, it was found that the ANN architecture of 3-8-2 gave the optimum prediction performance. In assessing the performance of the developed model using statistical error metrics, it was found that the model had a mean squared error of 9.2 x 10-4, a root mean square error of 0.0303 and a correlation coefficient (R) of 0.99645. Overall, the results from the model’s prediction were found to be consistent with experimental data. To the best of the authors’ knowledge, this study is one of the few available studies on cocoyam drying parameters modelling using ANN. This indicates that ANN can effectively describe the drying process of cocoyam owing to its good accuracy and applicability to a wide range of situations.
Keywords: Artificial Neural Network, Drying, Cocoyam, Modeling, Moisture content.
Title: PREDICTION OF COCOYAM DRYING PARAMETERS USING ARTIFICIAL NEURAL NETWORK
Author: Uwem Ekwere Inyang, Jock Asanja Alexander, Nkwocha Promise Chibuzo
International Journal of Novel Research in Engineering and Science