The Estimating of Rambutan's Peel Thickness by Using the Artificial Neural Network for Designing of the Rambutan Peeling Machine

摘要

The artificial neural network (ANN) models were used for estimating the peel thickness of rambutan by measuring its maximum diameter for designing of the rambutan peeling machine. This paper proposed the suitable pair of the input and output which provided the best accuracy by the ANN method. The analytical result showed that the best accuracy occurred in the case of the maximum diameter in Z-axis was an input and the peel thickness in Z-axis was an output and the optimal ANN model of this case consisted of one hidden layer with six neurons in hidden layer, was able to produce the peel thickness of rambutan values with the R2 value of curve between the values which predicted by optimal ANN and the actual values was 0.9401.