A model to predict the lifetime of pneumatic conveyor bends

Hanson, Robert (2001) A model to predict the lifetime of pneumatic conveyor bends. PhD thesis, University of Greenwich.

PDF - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (82MB)

Abstract

Bursts in pneumatic conveyor pipelines in industrial processes are a well-known hindrance to the smooth operation of any plant. Unanticipated stoppages can have serious financial implications for any company using pneumatic conveying technology, and health and safety factors are also paramount. This thesis describes an attempt to enable improved prediction of bend-wear and bend lifetime, so that more cost-effective survey work can be undertaken in anticipation of bursts.

This work delivers a tool that allows bend lifetime prediction to be made according to: the bend geometry and material; the material conveyed and its rate of transportation; and bend wall thickness.

Firstly, a computational model based on the coupling of CFD and particle tracking techniques is created in order to encompass the mechanics of the erosion process. This erosion process is assumed to be dominated by impact damage, and predictions of bend lifetime are made using empirical erosion algorithms gleaned from laboratory experiments, commercial CFD code (PHOENICS, and GENTRA its particle tracking sister code), and custom erosion modelling code that employs a three-dimensional toroidal geometry.

Secondly, matrices of predictions are built up using the mathematical modelling technology mentioned above. These predictions are collated behind a friendly interface to produce a far more accessible piece of PC software that an engineer can employ quickly and easily. More general bend-life predictions are interpolated from this fundamental dataset using behaviours established in the course of this work, according to the particular conveying conditions input by the user.

Predictions in the desktop tool are calibrated to actual bend lives as established by experimentation on a full-scale pneumatic conveyor. This experimental work was an integral part of this EPSRC-funded project, and allows some estimation of error magnitude in the predictive tool.

Actions (login required)