Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Modelling of mass transfer operations with artificial neural networks
Thesis (MEng) -- StellenboschUniversity, 1994.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Stellenbosch : Stellenbosch University
2012
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613979063353345 |
|---|---|
| access_status_str | Open Access |
| author | Giles, Andre Egerton |
| author2 | Van Deventer, J. S. J. |
| author_browse | Giles, Andre Egerton Van Deventer, J. S. J. |
| author_facet | Van Deventer, J. S. J. Giles, Andre Egerton |
| author_sort | Giles, Andre Egerton |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (MEng) -- StellenboschUniversity, 1994. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/58461 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:44:44.746Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2012 |
| publishDateRange | 2012 |
| publishDateSort | 2012 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/58461 Modelling of mass transfer operations with artificial neural networks Giles, Andre Egerton Van Deventer, J. S. J. Aldrich, C. Stellenbosch University. Faculty of Engineering. Dept. of Chemical Engineering. Mass transfer -- Computer simulation Artificial intelligence Chemical processes -- Computer simulation Metallurgy -- Computer simulation Rare earth metals -- Metallurgy -- Computer simulation Solvent extraction -- Computer simulation Dissertations -- Chemical engineering Thesis (MEng) -- StellenboschUniversity, 1994. Mass transfer processes are fundamental to the process industries, ranging from simple diffusion to large scale processes involving multiple modes of mass transfer. The majority of industrial mass transfer processes occur in heterogeneous or multi‑phase solutions and are consequently complex. Some chemical and metallurgical processes, such as solvent extraction or mass transfer in stirred tanks, involve multi‑component systems with mixtures containing numerous components and are well‑known examples of processes that are difficult to model from first principles or even empirically, owing to a wide range of possible interactions between the chemical species. Often the behaviour of industrial mass transfer systems is such that existing modelling techniques cannot account for the complexity of the reactions and interactions contributing to the behaviour of the system. This especially comes to the fore in the inability of most of the existing modelling methods aimed at generalizing the behaviour of the subsystems of large mass transfer systems. It is therefore not surprising that notwithstanding the considerable research on chemical and metallurgical processes, the process industries are abound with ill‑defined processes that can only be modelled partially from first principles if at all. It is in this application that neural network modelling offers a new opportunity to the process engineer, for it allows the rapid development of models of complex systems that would normally be very time consuming to develop using conventional modelling techniques. Neural networks are suitable for systems that can be modelled via a general approach (i.e. does not require explicit modelling) for it has the ability to capture hidden process characteristics. In this study the neural network modelling of complex mass transfer operations pertaining to the solvent extraction of rare earth metals and the solid‑liquid mass transfer in agitated vessels was investigated. These neural network models were evaluated against existing regression models typically used to represent these mass transfer systems. The first chapter of this thesis is devoted to a brief overview of the nature and use of artificial neural network modelling techniques for the simulation and modelling of processes relevant to this thesis. In the second chapter the current knowledge relating to rare earth solvent extraction and the solid‑liquid mass transfer in agitated vessels is reviewed and the models developed are critically evaluated. In the third chapter the basic principles of artificial neural networks and the potential advantages that neural network modelling holds for practical applications are discussed. Recent advances in neural network technology, as well as some industrial applications of neural networks are highlighted. In the fourth chapter the use of artificial neural networks in the analysis and modelling of the complex behaviour found in rare earth solvent extraction systems is investigated. Some of the more popular extraction systems were investigated experimentally, confirming trends that have been documented for these systems. It is shown that neural network models of the rare earth solvent extraction systems considered in this investigation were more able to generalize their behaviour than traditional regression models. In the penultimate chapter, neural networks are used to model solid‑liquid mass transfer in agitated vessels. It is shown that caution has to be exercised in the use of dimensionless variables as this can result in less accurate representations of the mass transfer behaviour in certain systems. Masters 2012-08-27T11:38:59Z 2012-08-27T11:38:59Z 1994 Thesis http://hdl.handle.net/10019.1/58461 en Stellenbosch University 140 pages : ill. application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Mass transfer -- Computer simulation Artificial intelligence Chemical processes -- Computer simulation Metallurgy -- Computer simulation Rare earth metals -- Metallurgy -- Computer simulation Solvent extraction -- Computer simulation Dissertations -- Chemical engineering Giles, Andre Egerton Modelling of mass transfer operations with artificial neural networks |
| title | Modelling of mass transfer operations with artificial neural networks |
| title_full | Modelling of mass transfer operations with artificial neural networks |
| title_fullStr | Modelling of mass transfer operations with artificial neural networks |
| title_full_unstemmed | Modelling of mass transfer operations with artificial neural networks |
| title_short | Modelling of mass transfer operations with artificial neural networks |
| title_sort | modelling of mass transfer operations with artificial neural networks |
| topic | Mass transfer -- Computer simulation Artificial intelligence Chemical processes -- Computer simulation Metallurgy -- Computer simulation Rare earth metals -- Metallurgy -- Computer simulation Solvent extraction -- Computer simulation Dissertations -- Chemical engineering |
| url | http://hdl.handle.net/10019.1/58461 |
| work_keys_str_mv | AT gilesandreegerton modellingofmasstransferoperationswithartificialneuralnetworks |