Investigation into calibration of discrete element model parameters for scale-up and validation of particle-structure interactions under impact conditions
Discrete elementmethod (DEM) is a popular numerical method used to understand the discontinuous flow ofgranular materials and optimise the design and operation of equipment. The impingement of granularmaterial on to flat surfaces can involve complex flow dynamics and can be difficult to model using analyticaltechniques. There are a vast number of industrial applications where particle-to-structure interactions occursuch as belt conveyor impact plate transfer stations consisting of a large number of particles and complexparticulate behaviour. Classical analytical methods can be used to provide a quantitative description of theflow of granular material through a transfer point in regards to trajectory and velocity distribution but aregenerally limited to 2-D analysis. DEM methodology has been well established but there is a lack of detailedvalidation of DEM models to experimental data and methods to calibrate and scale-up DEM models to attainaccurate predictions and results.This paper presents a detailed comparative analysis between classical analytical methods and DEM to predictthe flow mechanisms associated with the deformation of granular material impacting a flat plate. Results fromDEM simulations and analytical models are compared with experimental results from a variable-geometryconveyor transfer facility to validate and evaluate the numerical methods to solve granular flow problems.The study has focused on evaluating the ability to accurately model material discharge trajectories, thevelocity of impact from the inflowing stream, the velocity of the material stream after impingement and theresultant forces on the impact plate. Methods to effectively calibrate the DEM material interaction parametersare also presented and evaluated to quantify the calibration methodology. A sensitivity analysis has beenconducted to investigate the variation of DEM parameters and contact models on the impact reaction forcesand examine effective techniques to scale parameters to reduce computational time and resources.