Application of adaptive neuro-fuzzy inference system to predict draft and energy requirements of a disk plow

Naji Mordi N Al-Dosary,Saad A Al-Hamed,Abdulwahed M Aboukarima

doi:10.25165/j.ijabe.20201302.4077

Abstract

The energy and draft requirements of a disk plow have been recognized as essential factors when attempting to correctly match it with tractor power. This study examines the possible of using an adaptive neuro-fuzzy inference system (ANFIS) approach and its performance compared to a multiple linear regression (MLR) model to determine the energy and draft requirements of a disk plow. A total of 133 data patterns were obtained by conducting experiments in the field and from the literature. Of these 133 data points, 121 were arbitrarily selected and used for training, and the remaining 12 were used for testing the models. The input variables were plowing depth, plowing speed, soil texture index, initial soil moisture content, initial soil bulk density, disk diameter, disk angle, and disk tilt angle, and output variable was draft of the disk plow. Four membership functions were used with ANFIS: a triangular membership function, generalized bell-shaped membership function, trapezoidal membership function, and Gaussian curve membership function. An evaluation of the outcomes of the ANFIS and MLR modeling shows that the triangular membership function performed better than the other functions. When the ANFIS model draft predictions were compared to the measured values, the average relative error was –1.97%. A comparison of the ANFIS model with other approaches showed that the energy and draft requirements of the disk plow could be estimated with satisfactory accuracy. Keywords: ANFIS, MLR, disk plow, draft, tillage DOI: 10.25165/j.ijabe.20201302.4077 Citation: Al-Dosary N M N, Al-Hamed S A, Aboukarima A M. Application of adaptive neuro-fuzzy inference system to predict draft and energy requirements of a disk plow. Int J Agric & Biol Eng, 2020; 13(2): 198–207.

Highlights

Tillage plays a significant role in agricultural crop production
The soil texture index (STI) was determined using the method used by Oskoui and Harvey[33] because it yields single numbers for sand, silt, and clay contents combination
Oskoui and Harvey[33] showed that the STI reflects the effects of all three of the soil fractions

Summary

Introduction

Tillage plays a significant role in agricultural crop production. It produces changes in the soil structure with the goal of improving the tilth[1] and can be performed using primary and secondary tillage implements. One of the primary tillage implements is a disk plow which is used to plow, cut, pulverize, and invert hard soil[2]. Disk plows have inclined disks to the rear and the disk angle is the attachment angle of the disk relative to the direction of travel[3]. The backward slant of the disk from the vertical is commonly referred to as tilt[4]. The disk angle has a range of 35°-55° and the tilt angle has a range of 15°-25°[5]. The disk diameter is commonly 60-70 cm

Methods

Results

Conclusion