DEM analysis of a conveyor transfer chute
Date
2020-12
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: n the past, conveyor systems were designed and built based on trial-and-error,scale model prototyping, or using fundamental engineering mechanics analyses.However, these approaches can be inaccurate, time-consuming and costly. The analytical models used for the prediction of the flow behaviour of materials has become a widely accepted design tool. However, there are some variables which these models ignore, limiting the accuracy and application of this approach. The discrete element method (DEM), introduced by Peter A. Cundall in 1971, has developed into a popular design tool for the improvement of bulk material handling equipment and processes. DEM can model the complex interaction between discrete particles and particles-and-walls (structures) and designers can use it to quantify the performance of their designs. However, the main concern when using DEM, is the calibration and validation of the material properties. Inaccurate calibration of DEM models can lead to inaccurate results and consequently in inadequate designs.
The purpose of this study was to investigate the capability of, and accuracy with which DEM (PFC3D) can predict the bulk material flow of a cohesion less granular material through conveyor transfer points. The test material used was corn grains, having various non-spherical shapes and sizes. The material properties of the corn grains were determined by implementing several tests. The particle shape was modelled as spheres and clumps (multi-sphere particles) and further calibration tests were performed to obtain an accurate calibrated set of parameter values.The inclined conveyor belts used in the experimental tests had a chevron pat-tern to provide more control over the mass flow rate, especially at higher belt speeds. The chevron pattern caused less material to roll back to the feeding zone and its inclusion in the DEM model was first investigated and validated. The different components of a transfer point that were investigated included an impact plate, hood, rock box and a chute. The predicted material flow characteristics(impact forces, velocities and flow patterns) were verified by comparing the results to particle image velocimetry (PIV ) analysis of high-speed videos. Simulation strategies such as particle scaling, simplification of the shape model and decreasing the contact stiffness were also investigated to decrease computation time. DEM accurately predicted the impact force (|error|≤8.57 %), velocity (|error|≤5.31 %), outflow velocity (|error|≤19.90 %) and other investigated flow patternson a vertical impact plate, as well as the flow patterns in and from a rock box,through a hood and the mass flow rate causing blockage (|error|≤5.00 %). The maximum particle scale factor to accurately predict the flow against an impact plate was 7.0, while the particles could only be scaled by a factor 1.6 for the rocbox and chute build-up analyses. The analytical model, however, provided less accurate results in predicting the impact force (|error|≤14.17 %), velocity (|error|≤6.14 %) and outflow velocity (|error|≤19.38 %) on a vertical impact plate. It is therefore concluded that DEM accurately predicted the material flow through a conveyor transfer system and can be used with confidence in industrial applications if an appropriate set of calibrated parameters is used.
AFRIKAANSE OPSOMMING: Raadpleeg teks vir opsomming
AFRIKAANSE OPSOMMING: Raadpleeg teks vir opsomming
Description
Thesis (MEng)--Stellenbosch University, 2020.
Keywords
Conveyor transfer chute, Particle scaling, Discrete element method, Calibration, UCTD, Conveying machinery