- Title
- Investigation into the effects of material variability in the performance of bulk solids handling systems
- Creator
- Grassi Freire, Priscilla
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2021
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Material build-up, blockage and wear are amongst the most costly handling issues in materials handling systems. Efficient design must ensure flow and sufficiently long wear liner life, which often result in competing design criteria. Furthermore, current approaches for the design of bulk solids handling facilities are typically deterministic and lack due consideration of natural variability in material properties. The main objective of this thesis is to analyse flow and abrasive wear performance of materials handling systems, in particular for mass-flow bins and transfer chutes. A secondary aim is to investigate the influence of material property variability on flow and abrasive wear performance using stochastic modelling. For bins, the current standard for design of mass flow bins was used to determine the critical opening dimension to prevent arching and ensure reliable flow. Determining the critical opening dimension is a straightforward way of analysing performance from a flow perspective: should the material handled result in larger opening dimensions than the critical one, arching is likely to occur. To implement material variability to the existing procedure, Monte-Carlo simulations were used to generate random moisture contents off a distribution, and for each moisture content the material properties of relevance were determined. The critical arching dimension was calculated for each moisture, and the probability of the system failing was calculated by comparing the critical opening dimension of a base-case with the critical opening dimension required for flow of each random case. A second step in the model involved the application of an existing abrasive wear model to estimate the probability of failure by wear based on the variability imposed on the system. A combined performance analysis was then conducted which included both flow and wear to allow for the selection of bin design parameters that can optimise performance. Current guidelines for bin design suggest adding a safety factor of 20% to the calculated opening dimension. When variability in material properties was considered in the calculations, however, this factor has shown to be potentially unnecessarily conservative, especially when designing for the worst-case moisture content. This value is, however, conditional to the moisture distribution and material variability, and a stochastic approach is more recommended than a single safety factor applied to any design. For transfer chutes, the existing continuum mechanics approach was used to calculate flow velocities on flat plate transfers. Using laboratory experiments to measure the velocity after impact on a flat plate chute allowed for the conclusion that the current approach is inadequate to represent flow of cohesive materials as the velocities predicted differed greatly from those measured. Modifications to the method were then proposed to address such differences: a novel method to incorporate cohesiveness was presented, which builds upon the current frictional only approach. Another improvement to the existing continuum mechanics approach included the consideration of restitution in the calculation of velocity after impact, which has commonly been neglected. With these proposed changes, the velocity after impact predicted was used to feed an existing abrasive wear model. Performance in terms of wear was assessed using laboratory testing data to estimate liner life under the predicted operating conditions. Performance from a flow point of view was assessed in terms of free cross-sectional area, which builds off existing design criteria for accelerated flow, however with the added consideration of build up height. An overall performance curve was obtained using reliability analysis and considering the transfer as a system in series. This performance curve was then used to conduct a cost analysis. A stochastic component was added to account for the effects of variability in material properties. Similarly to the work conducted for bins, this stochastic component used Monte Carlo simulations to generate random moistures off a specified distribution. Material properties such as internal shear friction, bulk density and wall friction were estimated for each moisture and used in the calculation of design parameters for varying simulation conditions, and, using defined failure thresholds, the probability of failure was calculated. This research has showed that the existing continuum mechanics approach for transfer chute design is limited in representing the flow of cohesive materials. The velocities measured for an example of wet and sticky iron ore differed greatly from the velocities predicted with the existing approach. Such difference has impacts not only on the determination of trajectory but also on predicting liner life from an abrasive wear point of view. In order to improve velocity prediction, an improved consideration of cohesion is required, as well as adequate consideration of restitution. Finally, the frameworks presented for performance analysis of both bins and transfer chutes were incorporated in two case-studies applying the Life Cycle Cost analysis, in which the costs associated with failure from a flow/build-up and abrasive wear were analysed.
- Subject
- transfer chute; variability; life cycle cost analysis; mass-flow bin; flow properties
- Identifier
- http://hdl.handle.net/1959.13/1507359
- Identifier
- uon:56013
- Rights
- Copyright 2021 Priscilla Grassi Freire
- Language
- eng
- Full Text
- Hits: 15
- Visitors: 16
- Downloads: 1
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 5 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 370 KB | Adobe Acrobat PDF | View Details Download |