Journal of Food, Agriculture and Environment

Prediction of draft force and energy of subsoiling operation using ANN model

Author(s):

Reza Alimardani ^1*, Yousef Abbaspour-Gilandeh ², Ahmad Khalilian ³, Alireza Keyhani ², Seyed Hossein Sadati ⁴

Recieved Date: 2009-07-20, Accepted Date: 2009-10-11

Abstract:

Development and application of Artificial Neural Network (ANN) in modeling of the physical and dynamical soil properties and study of the possible relation of this technique with finite element method now a day is increasing. It is expected the success of ANN application particularly in simulating the operation of tillage practices and dynamic behavior of agricultural soil. One of the reasons in which leads us to use this method is its powerful estimation and the other one is that there is no specific material relation between the dependent and independent variables. In this research, parameters such as travel speed, tillage depth and other soil parameters such as cone index, moisture content, electrical conductivity and percentage of clay and sand are used to develop a prediction model. Experiments were conducted on three different types of coastal plain soils in the south eastern of United States for collection of required data. The model designed in this study was a multilayer network back propagation. Three algorithms of gradient descent with momentum, Lavenberg-Marquardt and scaled conjugated gradient algorithm were used to train the network. The best algorithm, used to train the network was chosen based on the high accuracy of prediction (95.8%) and more accurate simulation (97.6%), was the Lavenberg-Marquardt algorithm which recognized as the best algorithm (with two mean layers which include 12 neurons in the second layer) in comparison with other algorithms. The obtained scatter charts showed a correlation coefficient (R²) of 0.996 in network training and correlation coefficient of 0.987 in network testing between actual data and obtained data from ANN. Also these ANN models were compared with regression models based on data presented by ASAE and Garner’s model to evaluate the model. The comparison of the results of model and regression models to predict the draft force needed in subsoiling showed that ANN data are more close to actual data than the regression models.

Keywords:

Tillage energy, draft force, subsoiler, Artificial Neural Network, Lavenberg-Marquardt algorithm

---

Journal: Journal of Food, Agriculture and Environment
Year: 2009
Volume: 7
Issue: 3&4
Category: Agriculture
Pages: 537-542

---

Full text for Subscribers

---

Information:

Note to users

The requested document is freely available only to subscribers/registered users with an online subscription to the Journal of Food, Agriculture & Environment. If you have set up a personal subscription to this title please enter your user name and password. All abstracts are available for free.

Article purchasing

If you like to purchase this specific document such as article, review or this journal issue, contact us. Specify the title of the article or review, issue, number, volume and date of the publication. Software and compilation, Science & Technology, all rights reserved. Your use of this website details or service is governed by terms of use. Authors are invited to check from time to time news or information.

---

Purchase this Article:   20 Purchase PDF Order Reprints for 15

---

Share this article :