Increasing the capacity of Australian raw sugar factory milling units

This thesis reports on an investigation to identify methods to increase the capacity or throughput of the six-roll roller mills used in Australia to extract sugar from sugarcane.The approach taken was to gain an understanding of the factors affecting mill throughput through the application of the computational milling model, developed in recent years at James Cook University. The computational milling model is based on general equations of force equilibrium and continuity and a general description of sugarcane material behaviour.\ud \ud The development of the throughput model was conducted in stages. Firstly, an experiment was conducted on a laboratory two-roll mill to gain an understanding of the factors affecting throughput on this simple milling geometry. A two-roll computational model was constructed to predict the observed behaviour, accounting for all mechanisms identified from the experimental results. Secondly, a three-roll computational model was constructed which was sufficient to describe the throughput behaviour of the factory six-roll mill. An experiment was conducted on a factory six-roll mill to provide data to validate the model. The three-roll computational model was tested across the range of geometries and operating conditions known to exist in Australian factories and its throughput predictions were tested against throughput measurements.\ud \ud The three-roll computational model was used to identify the main factors affecting mill throughput and was used to construct a data set across a wide range of parameter values. The data set was used in a multiple regression analysis to develop an empirical model that could readily be used to identify conditions for maximum throughput.\ud \ud The computational and empirical models developed during this investigation were shown to predict throughput better than existing models. Conditions for maximum throughput were identified and involved the openings between rolls, the speed of the rolls and the amount of water in the sugarcane material being processed.\ud \ud As part of the investigation, further development of the computational milling model was undertaken in order to advance the model to a sufficient standard for this investigation. A material parameter was introduced to define the hardening rule for the plastic material model following established soil mechanics methodology. Darcy’s law, describing fluid flow through porous media, was shown to adequately describe the flow of water through bagasse for a wide range of fluid velocities. Greater confidence in the measured magnitude of the permeability factor in Darcy’s law was gained through improved experimental and parameter estimation procedures. One of the experimental and parameter estimation procedures was found to significantly reduce the time involved in measuring both the hardening rule for the plastic material model and the permeability for Darcy’s law