Numerical and Experimental Investigation of Throughput Capacity in Cassava Peeling Processes

Ogunnigbo Charles Olawale; Lodewyk Willem Beneke; Christiaan Coenrad Oosthuizen

doi:10.37934/araset.56.5.97109

Authors

Ogunnigbo Charles Olawale Tshwane University of Technology, Faculty of Engineering and Built Environment, Department of Mechanical and Mechatronics Engineering, Pretoria, South Africa
Lodewyk Willem Beneke Tshwane University of Technology, Faculty of Engineering and Built Environment, Department of Mechanical and Mechatronics Engineering, Pretoria, South Africa
Christiaan Coenrad Oosthuizen Tshwane University of Technology, Faculty of Engineering and Built Environment, Department of Mechanical and Mechatronics Engineering, Pretoria, South Africa

DOI:

https://doi.org/10.37934/araset.56.5.97109

Keywords:

Cassava Peeling, throughput capacity, dimensional analysis, Buckingham’s π theorem, predictive modelling, empirical validation

Abstract

This study aims to develop a theoretical model for predicting the throughput capacity of cassava peeling machines utilizing dimensional analysis based on Buckingham’s π theorem. Six governing variables—machine throughput capacity, tuber angular speed, peeling tool speed, tuber mass, peel penetration force, and peeling time—were identified and expressed in terms of three fundamental dimensions (M, L, T). Application of the Buckingham π theorem therefore yielded three independent dimensionless groups, derived using tuber angular speed, peel penetration force, and peeling tool speed as repeating variables. These π terms formed the basis of the predictive throughput model, which was subsequently calibrated and validated against experimental data. The established model demonstrated a high coefficient of determination (R² = 0.9047), indicating a strong correlation between predicted and measured throughput capacities. Model performance was further evaluated using error-based metrics, with the selected model yielding low RMSE (0.00065 kg/s) and MAE (0.00049 kg/s), confirming its strong predictive accuracy.