Prediction Of Soil Compaction Research Articles

Prediction of soil degree of compaction based on machine learning: a case study of two fine-grained soils

PurposeThe purpose of this paper is to develop an appropriate machine learning model for predicting soil compaction degree while also examining the contribution rates of three influential factors: moisture content, electrical conductivity and temperature, towards the prediction of soil compaction degree.Design/methodology/approachTaking fine-grained soil A and B as the research object, this paper utilized the laboratory test data, including compaction parameter (moisture content), electrical parameter (electrical conductivity) and temperature, to predict soil degree of compaction based on five types of commonly used machine learning models (19 models in total). According to the prediction results, these models were preliminarily compared and further evaluated.FindingsThe Gaussian process regression model has a good effect on the prediction of degree of compaction of the two kinds of soils: the error rates of the prediction of degree of compaction for fine-grained soil A and B are within 6 and 8%, respectively. As per the order, the contribution rates manifest as: moisture content > electrical conductivity >> temperature.Originality/valueBy using moisture content, electrical conductivity, temperature to predict the compaction degree directly, the predicted value of the compaction degree can be obtained with higher accuracy and the detection efficiency of the compaction degree can be improved.

Prediction of Soil Compaction using Conventional Tillage Systems under Different Operating Conditions

The field experiment was conducted in the College of Agriculture - University of Basra in Karma Ali located in the south of Basra city. The soil was of a clay-silty texture. The study aims to evaluate the negative impact of primary plowing equipment (moldboard plow, disk plow, and chisel plow) on the physical properties of the soil by studying soil compaction indicators such as soil resistance to penetration, soil bulk density, and total porosity. Three plowing depths (15, 20, and 25) cm and three forward speeds of the tractor (0.45, 0.75, and 1.05) m s-1 were used. Intelligent computing was introduced to predict results and simulate soil compaction through the use of Design Expert software. The results showed that there is an effect of all studied factors on the variables of soil physical properties. The effect on the soil penetration resistance characteristic of the type of plow was greater than the depth, and the depth was greater than the forward speed, as the soil penetration resistance was (4.53, 3.84, and 3.23) MPa for the moldboard plow, disc plow, and chisel plow, respectively. As for the effect of increasing the depth, it was 27%, while the effect of increasing the forward speed was 24%. As for the bulk density of the soil, the effect of the type of plow was also greater than the effect of the increase in depth, and the increase in depth was more effective than the increase in the forward speed, as the bulk density was (1.68, 1.52, and 1.48) g.cm-3 for the moldboard, disc, and chisel, respectively. As for the effect of depth, it reached 7%, while the effect of speed reached 5%. As for the total porosity of the soil, the effect of the plow type was greater than the depth, and the greater the depth than the forward speed, as the total porosity reached (43, 41, and 35) % for the moldboard, disc, and chisel plow, respectively. The increase in depth led to a decrease of 7%, and the increase in velocity led to a decrease of 4% in the index of total soil porosity. As for the predicted results, the amount of convergence between the field calculated and predicted results was (96, 95, 95) % for soil penetration resistance, bulk density, and total soil porosity, respectively.

Quantification of traffic-induced compaction based on soil and agricultural implement parameters

Vehicle-induced soil compaction occurs when agricultural machinery is working in the fields. The accumulated soil compaction could destroy soil structure and inhibit crop growth. The low degree of visualization of soil compaction has always been an important reason for restricting the development of compaction alleviation technology. Therefore, the main objective of this study was to predict soil compaction based on soil and agricultural implement parameters. The component of soil compaction prediction includes traffic-induced stress transmission evaluation and the quantitative relationship between soil stress and bulk density. The modified FRIDA model was used to elucidate the soil stress propagation, which has been validated by previous studies. The Bailey formula was used to establish the intrinsic relationship between soil stress and bulk density. The soil uniaxial compression test was applied to obtain the parameters of the Bailey formula, and soil samples were prepared with three different levels of water content. After fitting with the Bailey formula, under the condition that the soil moisture contents were 16%, 20%, and 24%, the fitting coefficients of soil bulk density were respectively 0.980, 0.959, and 0.975, which were close to 1. The results indicated that the Bailey formula could be used to calculate soil bulk density based on the stress conditions of the soil. To verify the practicality of the soil compaction prediction model, a field experiment was carried out in Zhuozhou City, Hebei Province, China. The treatment was set for 1, 3, 5, 7, and 9 times compaction with two different loads of compaction equipment. The results showed that the fit coefficient between the predicted and measured values of soil bulk density was greater than 0.641. The slope of the equation was greater than 0.782, proving that the soil bulk density prediction model based on agricultural implements and soil parameters has a good predictive effect on soil bulk density. The soil compaction evaluation model can provide a theoretical basis to further understand the soil compaction mechanism, allowing rational measures of soil compaction alleviation to be made. Keywords: soil traffic-induced compaction, agricultural implement, soil bulk density, no-tillage, soil structure, quantification DOI: 10.25165/j.ijabe.20201305.5480 Citation: Wang X L, Zhang X C, Lin X N, Sha L M, Yang H Y, Guo Z Y, et al. Quantification of traffic-induced compaction based on soil and agricultural implement parameters. Int J Agric & Biol Eng, 2020; 13(5): 134–140.

Finite element modeling of interaction between non-pneumatic mechanical elastic wheel and soil

The tire/soil interaction is an important and complex research topic in vehicle-terrain mechanics, which has a great influence on vehicle mobility and soil compaction. The purpose of this work was to develop a detailed nonlinear finite element model for parametric analysis of the interaction between mechanical elastic wheel and the soil. The three-dimensional finite element model of the mechanical elastic wheel, which considers material nonlinearity, geometric nonlinearity, as well as large contact deformation between the mechanical elastic wheel and soil, was developed based on the physical model. The reliability and accuracy of the finite element model were validated by comparing the predicted wheel static loading characteristics in rigid plane with those of experiments. The finite element model of the soil is modeled as a two-layer system and the Extended Drucker–Prager model was used to describe the nature soil properties. The deformation and stress of the soil and mechanical elastic wheel under the static loading condition were studied in detail. Moreover, the effects of the axial load on mechanical elastic wheel/soil interaction were also analyzed. The research results could offer reference for an optimum structural design of a mechanical elastic wheel and a prediction of soil compaction caused by non-pneumatic tires.

Study and simulation of the effective factors on soil compaction by tractors wheels using the finite element method

Soil is a nonrenewable source that needs considerable management to prevent physical deteriorationby erosion and compaction. Compacted soil causes low fertility and yield. The purpose of this study isto investigate the effect of viscoelastic properties of soil and to determine important factors oncompaction. Furthermore, stress distribution, prediction of soil compaction and simulation of its effectunder tractor wheels using ANSYS software were also studied. Predicted results using ANSYSsoftware are compared with laboratory and field results. Simulations were carried out by changing andmeasuring effective factors on soil compaction. These factors consist of wheel parameters whichinclude: number of wheel passes, speed and load; and the soil parameters such as soil bulk density andYoung’s modulus. The predicted results indicated that maximum soil compaction in the first trafficwith 512 mm was induced by viscoelastic properties of soil and the minimum soil compaction in thesixth traffic was 8 mm caused by soil elasticity properties. Variation in soil bulk density wasnegligible. Also at each wheel pass, e maximum stress was in the soil surface and this decreased withincrease in depth. The maximum vertical stress on the soil in the sixth traffic was 120.477 kPa at 2.52km/h and the minimum was 117.46 kPa at 5 km/h.

Prediction of soil compaction under pneumatic tires a using fuzzy logic approach

Artificial intelligence systems are widely accepted as a technology offering an alternative way to tackle complex and ill-defined problems. They can learn from examples, are fault tolerant in the sense that they are able to handle noisy and incomplete data, are able to deal with non-linear problems, and once trained can perform prediction and generalization at high speed. Compared with traditional approaches, fuzzy logic is more efficient in linking the multiple inputs to a single output in a non-linear domain. The purpose of this study was to investigate the relationship between tire working parameters and soil compaction characteristics, and to illustrate how Fuzzy expert system might play an important role in prediction of soil. All experimental values were collected from soil bin. The trials were conducted in different tire types, vertical loads, inflation pressures and forvard velocities. In this paper, a sophisticated intelligent model, based on Mamdani approach fuzzy modeling principles, was developed to predict the changes in penetration resistance, final pressure and bulk density of soil due to wheel traffic. The verification of the proposed model is achieved via various numerical error criteria. For all parameters, the relative error of predicted values was found to be less than the acceptable limits (10%).

Prediction of soil compaction under pneumatic tyres

A computer program for prediction of the influence of traffic on compaction of agricultural soils is presented. Stress distribution in soil under rubber tyres with different inflation pressures and carrying different loads is calculated by use of a simple equation, accounting for uneven stress distribution over the tyre-soil contact area. The resulting soil compression and wheel sinkage depend on a number of factors including: the initial soil density, soil water content, time since the previous soil disturbance, the number of passes; and on possible horizontal soil displacement, resulting in critical state conditions. The short duration of wheel loads results in incomplete soil compression, as compared with the usual long-duration laboratory compression tests. An empirical function to account for this lag in the compaction process is included in the model.