|
Predicting Unit Draft of Tillage Implements Using Statistical Models and Neural Net Works
|
|
Abstract: A unit draft of tillage implements was predicted using statistical and neural networks models. The neural network was a multilayer feedforward network with 11 input and 1 output neurons. The input variables were chisel plow, moldboard plow, disc plow, soil texture index, plowing depth, rated plow width, forward speed, initial soil moisture content, initial soil bulk density, rated tractor power, and the number of plow passes over the soil. The neural network was trained using backpropagation learning algorithm. The overall performance of the neural network was quite sufficient. It could be used to predict the unit draft of tillage implement trends of the measured data for all plows. The standard deviations of the errors were 9.38, 6.57,and 8.45 kN/m2 for moldboard, chisel, and disc plows, respectively. Also, the coefficients of the linear correlation between the measured and the predicted values were higher than 0.95 for both the neural network and statistical models.
|
Publication year |
2004
|
Pages |
249 - 239
|
Availability location |
مكتبة كلية الزراعة جامعة الازهر
|
Availability number |
|
Organization Name |
|
Country |
Egypt
|
City |
القاهرة
|
Publisher |
Name:
الجمعية المصرية للهندسة الزراعية
|
serial title |
بمؤتمر الجمعية المصرية للهندسة الزراعية الثانى عشر مع قسم الهندسة الزراعية بكلية الزراعة جامعة الأسكندرية فى 4-5 أكتوبر 2004
|
Department |
Agriculture Power and Energy
|
Author(s) from ARC |
|
External authors (outside ARC) |
|
Agris Categories |
Agricultural machinery and equipment
|
AGROVOC TERMS |
Models.
Tillage.
Tillage equipment.
|
Proposed Agrovoc |
draft force;
|
Publication Type |
Conference/Workshop
|