Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant
Thesis (M. Ing.) -- University of Stellenbosch, 1995.
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_ | 1867613974545039360 |
|---|---|
| access_status_str | Open Access |
| author | Eksteen, Jacobus Johannes |
| author2 | Van Deventer, J. S. J. |
| author_browse | Eksteen, Jacobus Johannes Van Deventer, J. S. J. |
| author_facet | Van Deventer, J. S. J. Eksteen, Jacobus Johannes |
| author_sort | Eksteen, Jacobus Johannes |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (M. Ing.) -- University of Stellenbosch, 1995. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/58724 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:44:39.798Z |
| 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:10019.1/58724 The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant Eksteen, Jacobus Johannes Van Deventer, J. S. J. Aldrich, C. Stellenbosch University. Faculty of Engineering. Dept. of Industrial Engineering. Flotation -- Automation Separation (Technology) -- Automation Computer vision -- Industrial applications Dissertations -- Engineering Thesis (M. Ing.) -- University of Stellenbosch, 1995. The overall objective of this thesis was to develop a knowledge base for the development and implementation of an industrial vision based system for flotation control, which would be economically feasible, accurate and robust. The research was based on an extensive literature review and test work that was done at the concentrator of Eastern Platinum Limited. The test work consisted of the image analysis of video images of flotation froths which were mineralised with platinum group minerals and base metal sulphides. It was found that all the froth properties that operators can distinguish visually can also be accurately characterised by digital image analysis. Five predominant froth characteristics were identified by which all froth types may be characterised. These characteristics were found to be: the texture, bubble size, stability, horizontal mobility, and the degree of mineralisation of the froth. Image analysis algorithms have been implemented which could distinguish different classes for each of the five characteristics. Two image processing algorithms have been used to characterise froth texture: the Spatial Grey Level Dependence Matrix (SGLDM) and the Neighbouring Grey Level Dependence Matrix (NGLDM). Both texture characterisation algorithms generate vectors of textural features which are then classified by a Learning Vector Quantisation neural network. In all the cases studied, the froth textures could be classified with accuracies above 90 %. Horizontal froth mobilities were classified with 100% accuracy. Bubble size was found to be well correlated (>90%) with two of the SGLDM features. Bubble stability was evaluated by the use of a difference image (difference between two consecutive frames) from which a stability feature was calculated which followed the trend in the stability that was observed for the real froths. Self-organising map neural networks provided a powerful basis to analyse the process by following the behaviour of flotation plants which could be represented by a well-defined trajectory in a topological feature map. This thesis takes into account the important role that technology transfer plays in the successful implementation of new intelligent control technologies, and it identified the barriers against change that result from automation and it presented ways to surmount those barriers. Furthermore, it proposes a control strategy, evaluates the commercially available hardware for image acquisition, evaluates the role of ore mineralogy and plant structure on flotation characteristics (using Eastern Platinum as a case-study), and it includes an in-depth operability study of the proposed machine vision system. Finally, it was concluded that the proposed vision based control system would be commercially viable, subjected to industrially set performance and robustness criteria. Masters 2012-08-27T11:39:06Z 2012-08-27T11:39:06Z 1995 Thesis http://hdl.handle.net/10019.1/58724 en Stellenbosch University 240 pages : ill. application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Flotation -- Automation Separation (Technology) -- Automation Computer vision -- Industrial applications Dissertations -- Engineering Eksteen, Jacobus Johannes The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant |
| title | The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant |
| title_full | The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant |
| title_fullStr | The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant |
| title_full_unstemmed | The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant |
| title_short | The development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant |
| title_sort | development and technology transfer of an industrial machine vision system for the control of a platinum flotation plant |
| topic | Flotation -- Automation Separation (Technology) -- Automation Computer vision -- Industrial applications Dissertations -- Engineering |
| url | http://hdl.handle.net/10019.1/58724 |
| work_keys_str_mv | AT eksteenjacobusjohannes thedevelopmentandtechnologytransferofanindustrialmachinevisionsystemforthecontrolofaplatinumflotationplant AT eksteenjacobusjohannes developmentandtechnologytransferofanindustrialmachinevisionsystemforthecontrolofaplatinumflotationplant |