Estimation of concentrate grade in platinum flotation based on froth image analysis
| dc.contributor.advisor | Aldrich, C. | en_ZA |
| dc.contributor.author | Marais, Corne | en_ZA |
| dc.contributor.other | University of Stellenbosch. Faculty of Engineering. Dept. of Process Engineering. | |
| dc.date.accessioned | 2010-11-19T10:00:43Z | en_ZA |
| dc.date.accessioned | 2010-12-15T10:36:17Z | |
| dc.date.available | 2010-11-19T10:00:43Z | en_ZA |
| dc.date.available | 2010-12-15T10:36:17Z | |
| dc.date.issued | 2010-12 | en_ZA |
| dc.description | Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2010. | en_ZA |
| dc.description | Thesis presented in partial fulfilment of the requirements for the degree MASTER OF SCIENCE IN ENGINEERING (EXTRACTIVE METALLURGICAL ENGINEERING) in the Department of Process Engineering at the University of Stellenbosch | en_ZA |
| dc.description.abstract | ENGLISH ABSTRACT: Flotation is an important processing step in the mineral processing industry wherein valuable minerals are extracted. Flotation is a difficult process to control due to its complexity, meaning that the reversal of series of changes will not necessarily bring the process back to its original state. Expert knowledge is incorporated in flotation control through operator experience and intervention, which is subject to many challenges, creating the need for improvement in control. The performance of a flotation cell is often determined by evaluating froth appearance. The application of image analysis to capture, evaluate and monitor froth appearance poses multiple benefits such as consistent and reliable froth appearance evaluation. The objective for this study was to conduct a laboratory study for the collection of froth images with the purpose of evaluating the feasibility of using image information to predict platinum froth grade. Laboratory test work was performed according to a fractional factorial experimental design. Six variables were considered: air flowrate, pulp level and collector, activator, frother and depressant dosages. The laboratory study results were quantified by assay analysis. Analysis of variance only revealed the significant effect of pulp height and collector addition on flotation performance. Data pre-processing revealed information regarding feature correlations and variance contributions. Data analysis from captured images achieved reliable froth grade predictions using random forest classification and artificial neural network (ANN) regression techniques. Random forest classification accuracies of 86.8% and 75.5% were achieved for the following respective datasets: image data of each individual experiment (average of all experiments) and all image data. The applied ANN models achieved R2 values 0.943 and 0.828 for the same 2 datasets. An industrial case study was done wherein a series of step changes in air flowrate was made on a specific flotation cell. The limited industrial case study results supported laboratory study results. Multiple linear regression performed very well, reaching RÂȘ values up to 0.964. Neural networks achieved slightly better with R2 values of up to 0.997. Based on the findings, the following main conclusions were drawn from this study: - Reliable predictions using classification and regression models on image data were proved possible in concept by the laboratory study, and supported by results from an industrial case study on a narrow system. The following main recommendations were made for further investigation: - Research over a larger range of operating conditions is needed to find a more comprehensive solution. - Investigations should be conducted to determine hardware requirements and specifications in terms of minimum resolution, lighting requirements, sampling frequency and data storage. Software requirements, specifications and maintenance challenges should also be investigated for implementation purposes once a more comprehensive solution has been found. - Strategies in terms of camera placement and model building will need to follow, giving special attention to a strategy to handle ore composition change. | en_ZA |
| dc.description.abstract | AFRIKAANSE OPSOMMING: Flotasie is ân belangrike proses in die mineraal proseseringsbedryf vermoeid met die ontginning van waardevolle minerale. Die proses is moelik om te beheer vanweĂ« sy kompleksiteit, wat verwys na die onvermoĂ« om die proses terug te bring na sy oorspronklike toestand deur ân reeks veranderinge om te keer. In die algemeen word spesialis kennis deel van prosesbeheer deur die toepassing van operateurs se ervaring en ingryping, wat opsigself verskeie uitdagings bied wat die behoefte aan verbeterde beheertoestelle en strategieĂ« daarstel. Die werkverrigting van flotasieselle word gereeld beoordeel op grond van die voorkoms van die skuim. Die gebruik van beeldverwerking om diĂ© inligting vas te vang vir monitering en evaluering doeleindes hou verskeie voordele in, bv. konsikwente en betroubare evaluasie van die skuimvoorkoms. Die doelwitte vir hierdie studie was om ân laboratorium studie te loods vir die opname van skuimbeelde, met die doel om die bruikbaarheid van beeldinligting vir die voorspelling van die flotasieprodukkwaliteit, te ondersoek. Die laboratorium gevallestudie is uitgevoer aan die hand van ân fraksionele faktoriale eksperimentele ontwerp. Ses veranderlikes was ondersoek naamlik, lugvloeitempo, pulphoogte en versamelaar aktiveerder en depressant toevoeging. Die studie se resultate is gekwantifiseer deur die analise van die skuim inhoud. ân Analise van variansie het slegs die invloed van pulphoogte en versamelaartoevoeging op die flotasievertoning uitgelig. Data voorverwerking het inligting uitgelig rondom die veranderlikes se verhouding met mekaar. Data analise metodes, naamlik lukrake klassifiseringswoude en neurale netwerk regressie, is toegepas op die versamelde beelddata en het belowende resultate gelewer. Lukrake klassifiseringswoude het klasse gedentifiseer met akkuraathede van 86.8% en 75.5% vir die volgende onderskeie datastelle: individuele eksperimente se beeld data (gemiddeld oor alle eksperimentele lopies), alle beelddata as een stel. Die neurale netwerke het RÂȘ waardes van 0.943 rn 0.828 gelewer vir dieselfde 2 datastelle. Die beperkte nywerheidsgevallestudie het verandering in lugvloeitempo toegelaat vir ân enkele flotasie sel. Die resultate het die bevindinge van die laboratorium gevallestudie gesteun. Veelvoudige lineere regressie het RÂȘ waardes van tot en met 0.964 gelewer. Neurale netwerke het daarop verbeter met waardes tot en met 0.997. Die volgende hoof gevolgtrekkinge was duidelik vanuit die resultate: - Betroubare voorspellings was moontlik met die toepassing van klassifikasie en regressie modelle op die laboratorium studie data. Die resultate is ondersteun deur soortgelyke resultate van die beperkte nywerheidsgevallestudie. Die volgende hoof aanbevelings was gemaak vir verdere navorsing: - Navorsing oor ân wyer reeks proseskondisies is nodig om ân meer omvattende oplossing te vind. - ân Ondersoek moet geloods word om die hardeware vereistes en spesifikasies in terme van die minimum beeld resolusie, beligting vereistes, monsterneming tempo en die berging van data te bepaal. Sagteware vereistes, spesifikasies en instandhouding uitdagings moet ook ondersoek word vir implementasie doeleindes sodra ân meer omvattende oplossing gevind is. - StrategieĂ« in verband met die plasing van kamers en die ontwikkeling van modelle is nodig, waarin spesiale aandag gegee moet word om die probleem van veranderende ertssamestelling op te los. | af |
| dc.format.extent | 80 p. : ill. | |
| dc.identifier.uri | http://hdl.handle.net/10019.1/5346 | |
| dc.language.iso | en | |
| dc.publisher | Stellenbosch : University of Stellenbosch | |
| dc.rights.holder | University of Stellenbosch | |
| dc.subject | Flotation | en_ZA |
| dc.subject | Dissertations -- Process engineering | en |
| dc.subject | Theses -- Process engineering | en |
| dc.subject | Image Analysis | en |
| dc.subject | Machine vision | en |
| dc.subject | Platinum | en |
| dc.title | Estimation of concentrate grade in platinum flotation based on froth image analysis | en_ZA |
| dc.type | Thesis |