Method and apparatus for on-line estimation of soil parameters during excavation

Time-varying forces from soil–machine interactions cause stresses in the components of earthmoving machinery, which may cause damage to the machine. It is not always possible to know all the characteristics of a soil sample prior to excavation; however, by estimating necessary soil parameters, it is possible to predict the soil–machine interaction forces in a practical manner. This article presents the development of a simple apparatus and method for estimating the soil parameters from the cutting force measured by the novel bench-scale excavating tool, validation of the soil model, and comparison with other available techniques. The apparatus used to collect data of soil forces on a tool consists of an instrumented crank-slider mechanism equipped with a thin plate to fragment the soil, which is contained in a sample box. Using the Mohr-Coulomb earth pressure model to predict failure force during the interaction, two methods are used to minimize the error between the predicted and measured failure force, that allows to estimate soil parameters: First, the Newton–Raphson Method (NRM) is used to minimize the error, which allows estimation of two soil parameters (interface friction angles) on non-cohesive soil samples. Additionally, a new estimation scheme based on the NRM is presented, that uses an auxiliary equation, and allows estimation of up to three soil parameters, including interface friction angles and cohesion. Comparing the results obtained from the presented apparatus, it is confirmed that the friction angles are successfully estimated for two non-cohesive particulate materials. Additionally, it is shown that the new scheme demonstrates smaller error in estimating soil parameters for cohesive and non-cohesive soil samples than previously reported methods. The parameter estimation method is subsequently applied to determine the properties of highly cohesive oil sand, and delivers promising results.     

The mechanics of soil cutting by a rotating wire

The cutting of soil by a rotating wire analogous to the tip of a rotary tiller blade while cutting a two-dimensional soil slice over a range of ‘fetch-ratios’ (bite length/depth-ratios) in a quasi-static condition is presented. A theoretical models based on Mohr-Coloumb soil mechanics has been proposed to predict forces on the wire (tip). The model is dependent upon observed passive general shear failure of the soil slice towards the curved free surface of a previous cut and the lateral local shear failure towards the undeformed soil. The predicted forces in a frictional soil and in a pure cohesive medium (artificial clay) agreed well with experimental results.     

Simulation on mechanical behavior of cohesive soil by Distinct Element Method

Analysis in the dynamic mechanical behavior of cohesive soils subjected to external forces is very important in designing and optimizing terrain machines. Distinct Element Method (DEM) is an ideal method to analyze large discontinuous deformations of soil, but the conventional DEM model is difficult in simulating the complex behavior of cohesive soil. In order to simulate and analyze the behavior of cohesive soil accurately, the DEM mechanical model of cohesive soil with parallel bonds between particles was established by considering the capillary and the dynamic viscous forces induced by the presence of water between soil particles. During the excavation process by a bulldozing plate, the dynamic behavior of cohesive soil was simulated by DEM software PFC2D. The phenomena that the discrete particles were bonded into clusters initially, and the clusters were broken into smaller clusters or discrete particles during the excavation process, are consistent with the ruptures and separations of the actual cohesive soils subjected to external forces.     

Dislocation approach to study interaction of adhesively bonded weak zones

The problem of interaction between crack-like interface defects subjected to a remote tensile stresses in an elastic bi-plane is considered using a model of an adhesively bonded asymmetric weak zone. In this model, the opening displacements are prescribed by a basis function which contains free parameters and automatically accounts for the asymmetry and the “true” stress–strain field behavior near the tips. The corresponding adhesive forces which can be very different by physical origin, are determined a posteriori. The limiting situations: transformation of one of the defects to the nucleus of a cohesive crack or the rupture of an obstacle between the weak zones are analytically described.     

Constitutive behaviour of mixed mode loaded adhesive layer

Mixed mode testing of adhesive layer is performed with the Mixed mode double Cantilever Beam specimen. During the experiments, the specimens are loaded by transversal and/or shear forces; seven different mode mixities are tested. The J-integral is used to evaluate the energy dissipation in the failure process zone. The constitutive behaviour of the adhesive layer is obtained by a so called inverse method and fitting an existing mixed mode cohesive model, which uses a coupled formulation to describe a mode dependent constitutive behaviour. The cohesive parameters are determined by optimizing the parameters of the cohesive model to the experimental data. A comparison is made with the results of two fitting procedures. It is concluded that the constitutive properties are coupled, i.e. the peel and shear stress depend on both the peel and shear deformations. Moreover, the experiments show that the critical deformation in the peel direction is virtually independent of the mode mixity.     

Improved sinkage algorithms for powered and unpowered wheeled vehicles operating on sand

Modeling and simulation of vehicles in sand is critical for characterizing off-road mobility in arid and coastal regions. This paper presents improved algorithms for calculating sinkage (z) of wheeled vehicles operating on loose dry sand. The algorithms are developed based on 2737 tests conducted on sand with 23 different wheel configurations. The test results were collected from Database Records for Off-road Vehicle Environments (DROVE), a recently developed database of tests conducted with wheeled vehicles operating in loose dry sand. The study considers tire diameters from 36 to 124 cm with wheel loads of 0.19–36.12 kN. The proposed algorithms present a simple form of sinkage relationships, which only require the ratio of the wheel ground contact pressure and soil strength represented by cone index. The proposed models are compared against existing closed form solutions defined in the Vehicle Terrain Interface (VTI) model. Comparisons suggest that incorporating the proposed models into the VTI model can provide comparable predictive accuracy with simpler algorithms. In addition to simplicity, it is believed that the relationship between cone index (representing soil shear strength) and the contact pressure (representing the applied pressure to tire-soil interface) can better capture the physics of the problem being evaluated.     

Mobility algorithm evaluation using a consolidated database developed for wheeled vehicles operating on dry sands

A substantial number of laboratory and field tests have been conducted to assess performance of various wheel designs in loose soils. However, there is no consolidated database which includes data from several sources. In this study, a consolidated database was created on tests conducted with wheeled vehicles operating in loose dry sand to evaluate existing soil mobility algorithms. The database included wheels of different diameters, widths, heights, and inflation pressures, operating under varying loading conditions. Nine technical reports were identified containing 5253 records, based on existing archives of laboratory and field tests of wheels operating in loose soils. The database structure was assembled to include traction performance parameters such as drawbar pull, torque, traction, motion resistance, sinkage, and wheel slip. Once developed, the database was used to evaluate and support validation of closed form solutions for these variables in the Vehicle Terrain Interface (VTI) model. The correlation between predicted and measured traction performance parameters was evaluated. Comparison of the predicted versus measured performance parameters suggests that the closed form solutions within the VTI model are functional but can be further improved to provide more accurate predictions for off-road vehicle performance.     

Sinkage tests for mobility study, modelling and experimental validation

The sinkage of the bearing tracks or wheels of a vehicle in soil induces a resistance to travel motion. Usually it is determined with methods based on the modelling of soil pressure-sinkage curves. This article presents a new method for modelling soil penetration tests as a result of experimental study of three standard soils. These soils have been chosen to represent the mechanical properties of a range of soils: a sand for frictional soils, a silt for cohesive soils and a silty sand for cohesive frictional soils. The models take into account the mechanical behaviour of soils where a small vertical sinkage can be assumed analogous to elastic behaviour, while for large sinkage, the analogy is with plastic behaviour. A New Model of Mobility (N2M) is proposed. A new equation relating the pressure p and the sinkage z is governed by four parameters which are constant for a specific soil in a given physical state. These parameters can be calculated with two sinkage tests made with two different plate diameters and are particularly stable: a small change of one of them involves a small change of the modelling. They are independent of the size of the sinkage plate and hence could pave the way for the extrapolation to the scale of full size vehicles. For the tested soils, comparison of the model results with experimental tests is very promising.     

Experimental study on wheel-soil interaction mechanics using in-wheel sensor and particle image velocimetry part II. Analysis and modeling of shear stress of lightweight wheeled vehicle

Wheeled vehicle mobility on loose sand is highly subject to shear deformation of sand around the wheel because the shear stress generates traction force of the wheel. The main contribution of this paper is to improve a shear stress model for a lightweight wheeled vehicle on dry sand. This work exploits two experimental approaches, an in-wheel sensor and a particle image velocimetry that precisely measure the shear stress and shear deformation generated at the interaction boundary. Further, the paper improves a shear stress model. The model proposed in this paper considers a force chain generated inside the granular media, boundary friction between the wheel surface and sand, and velocity dependency of the friction. The proposed model is experimentally validated, and its usefulness is confirmed through numerical simulation of the wheel traction force. The simulation result confirmed that the proposed model calculated the traction force with an accuracy about 70%, whereas the conventional one overestimated the force, and its accuracy was 13% at the best.     

Adhesion between elastic cylinders based on the double-Hertz model

A cohesive zone model for two-dimensional adhesive contact between elastic cylinders is developed by extending the double-Hertz model of Greenwood and Johnson (1998). In this model, the adhesive force within the cohesive zone is described by the difference between two Hertzian pressure distributions of different contact widths. Closed-form analytical solutions are obtained for the interfacial traction, deformation field and the equilibrium relation among applied load, contact half-width and the size of cohesive zone. Based on these results, a complete transition between the JKR and the Hertz type contact models is captured by defining a dimensionless transition parameter μ, which governs the range of applicability of different models. The proposed model and the corresponding analytical results can serve as an alternative cohesive zone solution to the two-dimensional adhesive cylindrical contact.     

The Bekker theory of rolling resistance amended to take account of skid and deep sinkage

The Bekker theory of rolling resistance of free rolling, towed, rigid wheels is amended to take account of both skid and deep sinkage without leading to excessive complexity in the predictive equations. Theoretical relationships between skid and sinkage are derived for a free rolling, towed, rigid wheel on a purely cohesive soil ( = 0) and on a purely frictional soil (c = 0) with a sinkage exponent of unity. Generally, good agreement is found between predicted and measured values of rolling resistance and sinkage at a given vertical load, on both sand and clay soils, at shallow and deep sinkage.     

Force transmitted below the soil surface by human gait

A force platform, which can provide three dimensional forces and moments on its top surface, was used to study force transmitted by human gait below the soil surface in order to understand detonation of antipersonnel landmines. Soils of varying depth were packed on the top surface of the platform to measure the forces transferred from the soil surface. Experimental variables included subjects (people), soil depth, soil type, moisture content, and compaction level. Soils used in this study were sand and sandy loam. There were medium and high two compaction levels for each soil. Sandy loam soil included two moisture contents; sand tested involved two moisture contents and dry sand. Soil depth varies from 0 (bare platform) to 200 mm. Five subjects with different weights were selected and used in this study.The subsoil force and its duration were measured for different subjects at a depth up to 200 mm. The impulse in subsoil was then calculated and used in evaluating the effect of different subjects on the force transfer in soil. The results indicated that loose soil can transfer larger force to subsoil than dense soil; test results showed that heavier subjects also created larger subsoil forces than lighter ones. Whether the effect of soil depth on subsoil impulse was significant was depended on the soil conditions. For the sand with 5.5% moisture content and bulk density of 1800 kg/m³, soil depth significantly affected subsoil impulses. For the sandy loam soil, the mass of subject increased from 50 to 100 kg resulted in 100% increase in subsoil impulses at all four depths; for the sand, the mass of subject increased from 55 to 100 kg approximately. This resulted in 80% increase in subsoil impulses under all four depths regardless of moisture content and bulk density. The results of this study will helpful for designing new equipment and evaluating existing machines for neutralizing landmines.     

Analytical models and laboratory measurements of the soil-tool interaction force to push a narrow tool through JSC-1A lunar simulant and Ottawa sand at different cutting depths

Excavation equipment for developing NASA’s lunar outpost must be carefully designed to reduce launch cost, minimize operation cost, and enhance reliability. Excavation equipment requires knowledge of the stresses and strains in the equipment caused by the forces experienced during excavation. The types of excavation anticipated indicate that blade tools would move the most material. There are several analytical models available to predict forces from blade tools interacting with soil; however, it is not clear which if any, can predict lunar excavation forces precisely enough. Consequently, we measured the forces to push narrow (2.5-cm wide) square and round rods through a control material, Ottawa sand, and JSC-1A lunar mare regolith simulant at different cut depths in a controlled laboratory setting. The measurement results were compared with the forces predicted by eight analytical models. The Zeng, Luth and Wismer, and the Qinsen and Suren models fit the measurements best, considering that our study was limited to pushing stimulant and sand with small rods. The results show that depth of cut has a dramatic effect on the soil-tool interaction forces. Consequently, lunar missions should use a series of shallow cuts to reduce equipment size and power requirements.     

Static liquefaction of a saturated loose sand stratum

The ideal case of an infinite horizontal, homogeneous oedometrically consolidated saturated loose sand stratum, sheared in displacement controlled conditions is numerically analysed. The phenomenon of the stratum liquefaction, which is due to both the material mechanical instability and the water presence within pores, is discussed. The numerical results have been obtained by means of a spatial one dimension finite difference numerical code within which both the local and the non-local versions of the same elasto-viscoplastic constitutive model are implemented. Both pore water and shear strain wave propagation within the soil layer is described.     

Modeling of Wormhole Growth in Cold Production

Cold production is a non-thermal process in which sand is intentionally produced with the oil in order to enhance oil recovery. Two experiments were performed to investigate the effect of producing large quantities of sand on the overall permeability of a formation. A large high porosity channel (wormhole) was observed in both experiments. A model of wormhole growth was successfully tested in two sand production experiments simulating the growth of a wormhole from a perforation in a vertical well. The produced volumes of oil, sand and gas, the pressure distribution along the pack and the final length of the wormholes were well predicted. The two sand packs had significantly different cohesive strengths. The strength of the sand did not have a significant effect on the growth of the wormholes. The formation of tensile failure bands at the wormhole surface, as observed in the experiments, may weaken the sand and allow it to be fluidized more easily. This weakening effect would explain the lower pressure gradients calculated at the surface of the wormhole while it developed compared to the critical pressure gradient for sand production predicted by Bratli and Risnes (1981).     

On the determination of the cohesive zone properties of an adhesive layer from the analysis of the wedge-peel test

An extensive numerical study of the mechanics of the “wedge-peel test” is performed in order to analyze the mode I steady state debonding of a sandwich structure made of two thin plastically deforming metallic plates bonded with an adhesive. The constitutive response of the metallic plates is modeled by J₂ flow theory, and the behavior of the adhesive layer is represented with a cohesive zone model characterized by a maximum separation stress and the fracture energy. A steady-state finite element code accounting for finite rotation has been developed for the analysis of this problem. Calculations performed with the steady-state formulation are shown to be much faster than simulations involving both crack initiation and propagation within a standard, non-steady-state code. The goal of this study is to relate the measurable parameters of the test to the corresponding fracture process zone characteristics for a representative range of adherent properties and test conditions. An improved beam bending model for the energy release rate is assessed by comparison with the numerical results. Two procedures are proposed for identifying the cohesive zone parameters from experimental measurements.     

Measurement and modeling for two-dimensional normal stress distribution of wheel on loose soil

On the Moon or Mars, typical target environments for exploration rovers are covered with fine sand, so their wheels easily slip on such weak ground. When wheel slippage occurs, it is hard for the rover to follow its desired route. In the worst case, the rover gets stuck in loose soil and cannot move anymore. To reduce the risk of the rover getting stuck, analysis of the contact mechanics between the soil and wheel is important. Various normal stress distribution models for under the wheel surface have been proposed so far. However, classical models assume a uniform stress distribution in the wheel’s width direction. In this study, we measured the two-dimensional normal stress distribution of a wheel in experiments. The results clarified that the stress distribution in the wheel’s width direction is a mountain-shape curve with a peak located at the center of the wheel. Based on the results, we constructed a stress distribution model for the wheel’s width direction. In this paper, we report our measurements for the two-dimensional stress distribution of a wheel on loose soil and introduce our stress distribution model for the wheel’s width direction based on our experimental results.     

A mechanical model for the adhesive contact with local sliding induced by a tangential force

Adhesion has been demonstrated to play an important role in contact and friction between objects at small scales. While various models have been established for adhesive contact under normal forces, studies on the adhesive contact under tangential force have been far fewer, which if any, are mostly confined to the non-slipping situations. In the present work, a model has been proposed for adhesive contact with local sliding under tangential forces. Herein, the onset of local sliding in adhesive contact has been addressed by assuming the nucleation of dislocations. By analogy with the emission of dislocations at a crack tip, the critical tangential force for the onset of sliding has been determined, and its effect on the evolution of contact size has also been studied. Comparison with relevant experiments has verified the validity of the present model.     

Developing an excavation torque model for dense cohesionless soils

Torque encountered during the rotary excavation of soils (e.g., when using the DJM method for deep soft ground improvement) poses a serious detrimental effect not only to the excavating machines but also to the viability of a project as a whole. Consequently, this research investigates ways and means of realizing the reduction of torque encountered during the excavation of cohesionless soils. In this paper, the development of a torque model for a rotary excavation of cohesionless soils is proposed. Whereas in most of the soil tillage theories (i) the cutting tool is usually partially exposed at the surface, and (ii) excavation is generally longitudinal, this model is significant because; (i) the excavation process is radial, and (ii) the blade is completely immersed in the excavated medium. Various theories for the prediction of forces acting during the interaction of cutting tools and soils in conjunction with localized modeling of all the other forces, applied and adopted to suit this excavation geometry, have been applied in the development of the torque model. Experimental data was obtained from excavation experiments performed on compacted completely saturated sand samples. Within the experimental and theoretical limitations, the results showed that this model represented the excavation process.     

Development of in-wheel sensor system for accurate measurement of wheel terrain interaction characteristics

Planetary rovers need high mobility on a rough terrain such as sandy soil, because such a terrain often impedes the rover mobility and causes significant wheel slip. Therefore, the accurate estimation of wheel soil interaction characteristics is an important issue. Recent studies related to wheel soil interaction mechanics have revealed that the classical wheel model has not adequately addressed the actual interaction characteristics observed through experiments. This article proposes an in-wheel sensor system equipped with two sensory devices on the wheel surface: force sensors that directly measure the force distribution between the wheel and soil and light sensors that accurately detect the wheel soil surface boundary line. This sensor design enables the accurate measurement of wheel terrain interaction characteristics such as wheel force distribution, wheel–soil contact angles, and wheel sinkage when the powered wheel runs on loose sand. In this article, the development of the in-wheel sensor system is introduced along with its system diagram and sensor modules. The usefulness of the in-wheel sensor system is then experimentally evaluated via a single wheel test bench. The experimental results confirm that explicit differences can be observed between the classical wheel model and practical data measured by the in-wheel sensor system.