http://hdl.handle.net/123456789/2088 Search the Channel search http://repository.ksu.edu.sa/jspui/simple-search http://hdl.handle.net/123456789/7875 Title: Prediction of specific draft of different tillage implements:Using neural networks<br/><br/>Authors: Al-Janobi, Abduirahman A.; Aboukatima, Abduhvahed M.; Ahmed, Khaled A.<br/><br/>Abstract: A Multilayer Perception with error backpropagalion learningalgorithm was used to build neural network model to predict specific draft(kN/m) of different tillage implements from the field data. The neuralnetwork model was trained and tested with different sites, tillageimplements, plowing depths, and forward operating speeds as inputparameters and the measured specific draft as output parameter. Thearchitecture of the neural networks consisted of two hidden layers with 24nodes in the first hidden layer and 12 nodes in the second layer. The hiddenand output layers have a sigmoid transfer functions in-neural networksmodel and the learning rule was momentum with 0.9 and step size 1.0. Thebest result was achieved at 65000 training runs that gave: minimum meansquared error equals to 0.0004 during training process. The results showedthat the variation of measured and predicted specific draft was small andthe correlation coefficient was 0.987 and mean squared error betweenmeasured and predicted specific draft was 0.1445. Sun, 01 Jul 2001 00:00:00 GMT