Prediction of Energy Requirement of a Tillage Tool in a Soil Bin using Artificial Neural Network

Abstract

A Multilayer feed-forward neural network with back propagation (BP) learning algorithm was used to build neural network model to predict energy requirement (N) of a tillage tool from the laboratory data. The neural network model was trained and tested with soil moisture content, plowing depths, and forward operating speeds as input parameters. The measured energy requirement (N) for a tillage tool in silty clay loam soil was used as output parameter. The architecture of the neural networks consisted of one hidden layer with 7 nodes. The hidden and output layer has a sigmoid transfer functions in-neural network. Lavenberg-Marquardt learning rule was used to train the network. The results showed that the variation of measured and predicted energy requirement (N) was small and the correlation coefficient was 0.9991 and mean squared error, root mean squared error and mean arithmetic error between measured and predicted energy requirement (N) were 6.1, 2.47 and 1.81 respectively. Such encouraging results indicate that the developed ANN model for energy requirement (N) prediction could be considered as an alternative and practical tool for predicting energy requirement (N) of a tillage tool under the selected experimental conditions. Further work is required to demonstrate the generalised value of this ANN in other soil conditions.