首页> 外文OA文献 >Development of accurate contact force models for use with Discrete Element Method (DEM) modelling of bulk fruit handling processes
【2h】

Development of accurate contact force models for use with Discrete Element Method (DEM) modelling of bulk fruit handling processes

机译:开发精确的接触力模型,以用于散装水果处理过程的离散元方法(DEM)建模

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务,本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文,但由于OA文献来源多样且变更频繁,仍可能出现获取不到、文献不完整或与标题不符等情况,如果获取不到我们将提供退款服务。请知悉。

摘要

This thesis is primarily concerned with the development of accurate, simplified and validated contact force models for the discrete element modelling (DEM) of fruit bulk handling systems. The DEM is essentially a numerical technique to model a system of particles interacting with one another and with the system boundaries through collisions. The specific area of application envisaged is in postharvest agriculture, where DEM could be used in simulation of many unit operations with bulk fruit, such as harvesting, transportation, sorting, storage and processing. Such simulations would provide crucial information to aid equipment and process designs to optimise the systems or to limit physical damage to produce during machine handling, for instance. In a DEM model, the material behaviour at collision is captured by a contact force model relating the local deformation in the contact region to the contact force. Contact force models represent one of the crucial elements of a DEM model. They have a direct bearing on the accuracy with which the material behaviour can be captured by the model.In carrying out the main objective of this project, theoretical, experimental as well as numerical approaches were used in order to, among other things, identify, derive, validate and simplify the relevant contact models proposed. Using the elastic contact force models as bases, mathematical contact force models for viscoelastic spheres are derived, for normal contact, tangential contact as well as torsional contact. The derived model for torsional contact was experimentally validated using a rheometer setup. Still using the rheometer setup, parameters of the derived tangential contact force model were determined. These parameters are: the friction coefficient, the viscous constant and the static shear modulus.The finite element method was used to further investigate the normal and tangential contact force models for elastic spheres. Regarding normal contact, the validity of the Hertz contact model under conditions of frictional contact, non-Hertzian contact surfaces and finite strains was investigated. The effect of friction to the force-deformation behaviour of elastic spheres was found to be negligible. The same is the case for the effects from the non-flat contact surfaces that result when stiffer objects come into contact with the more compliant ones. On the other hand, the study of the effect of finite strains revealed a strong influence of the extent of deformation on the Hertz model. The error arising was found to increase quadratically with increase in the normal deflection.Regarding the tangential contact, because the relevant models currently available in literature are generally either too inaccurate or computationally too costly to be implemented in DEM simulations involving large numbers of particles, an alternative model was derived herein and proposed. The model, it is argued, is simple to implement and can have the same accuracy as the Mindlin-Deresiewicz model. Because the proposed model does not include many different conditions and loading scenarios, it is argued that it will be much more convenient and less costly, computationally.By numerically filtering out all the vibration modes in the transient dynamic analyses of normal collisions of elastic spheres using FEA, an estimation of the relative quantities of the kinetic energy involved in the propagation of elastic waves in the colliding bodies could be made. The proportions varied according to the impact velocity from very low (less 4 %) to substantial proportions of up to 40 %. The relationship with the impact velocity was estimated to be proportional to a power 1/3 of the impact velocity. While this observation does not prescribe the amount of energy dissipation that can occur, it is extremely important because it provides an indication of the quantities of energy that can be available as elastic waves energy during collisions, and which potentially can be lost under some conditions. A significant amount of translational energy loss was found to occur for purely elastic collisions through absorption of energy as elastic waves, depending on the material elastic properties. The restitution coefficients for the purely elastic collisions depend on the impact velocity.Eigenvalue analysis of apple fruit models (based on Jonagold apple properties) yielded frequency modes in the ranges from 87.3 Hz to 849.4 Hz. These results were in good agreement with the ranges of values found in literature. The transient analysis of the viscoelastic collisions of apples, using the time dependent properties obtained from literature revealed no viscous response by the fruits. This is explained by the fact that the time constants reported on, and obtained from stress relaxation experiments are much higher than the typical collision times of the fruits. Because pendulum experiments on apple fruits have revealed the existence of viscous response in apples for these kinds of impacts, it can be concluded that the stress relaxation or creep experiments are not suitable for obtaining viscoelastic data for instantaneous processes such as collisions. From further dynamic transient analysis of FE models of the collisions of viscoelastic objects, restitution coefficients due to viscous energy dissipation were found to be dependent in a nonlinear manner to the impact velocities. A method to determine the effective viscous coefficient in the case of two viscoelastic objects colliding was investigated and between the two investigated choices - sum of inverses of the individual coefficients and arithmetic sum of the individual coefficients, the former seemed to yield more accurate estimation of the effective coefficient.
机译:本文主要涉及为水果散装处理系统的离散元建模(DEM)开发精确,简化和经过验证的接触力模型。 DEM本质上是一种数值技术,用于对通过碰撞相互影响的粒子系统以及系统边界进行建模的系统。设想的特定应用领域是在收获后农业中,其中DEM可用于模拟许多具有散装水果的单位操作,例如收获,运输,分类,存储和加工。例如,此类模拟将提供关键信息,以帮助设备和过程设计优化系统或限制在机器处理过程中产生的物理损坏。在DEM模型中,碰撞时的材料行为由将接触区域中的局部变形与接触力相关的接触力模型捕获。接触力模型代表了DEM模型的关键要素之一。它们直接关系到模型可以捕获材料行为的准确性。在执行该项目的主要目标时,使用了理论,实验以及数值方法来识别,推导,验证和简化建议的相关联系模型。以弹性接触力模型为基础,导出了粘弹性球体的数学接触力模型,用于法向接触,切向接触以及扭转接触。使用流变仪设置对实验得出的扭转接触模型进行了验证。仍使用流变仪设置,确定导出的切向接触力模型的参数。这些参数是:摩擦系数,粘滞常数和静态剪切模量。使用有限元方法进一步研究了弹性球体的法向和切向接触力模型。关于正常接触,研究了在摩擦接触,非赫兹接触表面和有限应变条件下赫兹接触模型的有效性。发现摩擦对弹性球的力-变形行为的影响可以忽略。当较硬的物体与较柔顺的物体接触时,非平坦接触表面会产生同样的效果。另一方面,对有限应变影响的研究表明,变形程度对Hertz模型有很大的影响。发现所产生的误差随法向挠度的增加而平方增加。关于切向接触,因为目前文献中可用的相关模型通常太不精确或计算量太大,以至于无法在涉及大量粒子的DEM模拟中实现。本文提出并提出了替代模型。有人认为该模型易于实现,并且可以与Mindlin-Deresiewicz模型具有相同的准确性。由于所提出的模型没有包括许多不同的条件和载荷情况,因此认为它将在计算上更加方便和廉价。通过数值过滤,在弹性球体法向碰撞瞬态动力学分析中,使用振动过滤所有振动模式有限元分析,可以估算与弹性波在碰撞体中传播有关的动能的相对量。比例根据冲击速度而变化,从非常低的比例(小于4%)到高达40%的比例。估计与冲击速度的关系与冲击速度的幂1/3成正比。尽管此观察结果并未规定可能发生的能量耗散量,但它非常重要,因为它提供了在碰撞过程中可以作为弹性波能量使用的能量数量的指示,并且在某些情况下可能会丢失。对于纯弹性碰撞,通过吸收能量作为弹性波,会发生大量的平移能量损失,具体取决于材料的弹性特性。纯弹性碰撞的恢复系数取决于撞击速度。苹果果实模型的特征值分析(基于乔纳金苹果的特性)产生的频率模式为87.3 Hz至849.4 Hz。这些结果与文献中发现的数值范围非常吻合。苹果的粘弹性碰撞的瞬态分析,使用从文献中获得的时间依赖性,发现水果没有粘性响应。时间常数报告于从应力松弛实验获得的,比典型的水果碰撞时间要高得多。由于对苹果果实进行的摆锤实验已经揭示了苹果对这些冲击的粘性响应的存在,因此可以得出结论,应力松弛或蠕变实验不适用于获取瞬时过程(如碰撞)的粘弹性数据。通过对粘弹性物体碰撞的有限元模型进行进一步的动态瞬态分析,发现由于粘性能量耗散导致的恢复系数以非线性方式依赖于冲击速度。研究了在两个粘弹性物体碰撞的情况下确定有效粘滞系数的方法,并且研究了两个选择的选择之间的关系:单个系数的逆之和和各个系数的算术和,前者似乎可以更准确地估算粘滞系数。有效系数。

著录项

  • 作者

    Dintwa Edward;

  • 作者单位
  • 年度 2006
  • 总页数
  • 原文格式 PDF
  • 正文语种 en
  • 中图分类

相似文献

  • 外文文献
  • 中文文献
  • 专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号