Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Evaluation and comparison of different machine-learning methods to integrate sparse process data into a spatial model in geometallurgy
Luleå University of Technology, Department of Civil, Environmental and Natural Resources Engineering, Minerals and Metallurgical Engineering.ORCID iD: 0000-0002-9227-2470
Luleå University of Technology, Department of Civil, Environmental and Natural Resources Engineering, Minerals and Metallurgical Engineering.
Luleå University of Technology, Department of Civil, Environmental and Natural Resources Engineering, Minerals and Metallurgical Engineering.ORCID iD: 0000-0002-5228-3888
2019 (English)In: Minerals Engineering, ISSN 0892-6875, E-ISSN 1872-9444, Vol. 134, p. 156-165Article in journal (Refereed) Published
Abstract [en]

A spatial model for process properties allows for improvedproduction planning in mining by considering the process variability ofthe deposit. Hitherto, machine-learning modelling methods have beenunderutilised for spatial modelling in geometallurgy. The goal of thisproject is to find an efficient way to integrate process properties (ironrecovery and mass pull of the Davis tube, iron recovery and mass pull ofthe wet low intensity magnetic separation, liberation of iron oxides, andP_80) for an iron ore case study into a spatial model using machinelearningmethods. The modelling was done in two steps. First, the processproperties were deployed into a geological database by building nonspatialprocess models. Second, the process properties estimated in thegeological database were extracted together with only their coordinates(x, y, z) and iron grades and spatial process models were built.Modelling methods were evaluated and compared in terms of relativestandard deviation (RSD). The lower RSD for decision tree methodssuggests that those methods may be preferential when modelling non-linearprocess properties.

Place, publisher, year, edition, pages
Elsevier, 2019. Vol. 134, p. 156-165
Keywords [en]
Data Integration, Spatial Model, WLIMS, Davis Tube, Iron Ore, Machine-learning, Geometallurgy.
National Category
Mineral and Mine Engineering Metallurgy and Metallic Materials
Research subject
Mineral Processing
Identifiers
URN: urn:nbn:se:ltu:diva-71579DOI: 10.1016/j.mineng.2019.01.032ISI: 000462107200015Scopus ID: 2-s2.0-85060907032OAI: oai:DiVA.org:ltu-71579DiVA, id: diva2:1263184
Note

Validerad;2019;Nivå 2;2019-02-11 (svasva)

Available from: 2018-11-14 Created: 2018-11-14 Last updated: 2019-04-11Bibliographically approved
In thesis
1. Bringing predictability into a geometallurgical program: An iron ore case study
Open this publication in new window or tab >>Bringing predictability into a geometallurgical program: An iron ore case study
2019 (English)Doctoral thesis, comprehensive summary (Other academic)
Alternative title[sv]
Skapande av predikterbarhet i ett geometallurgiskt program : en fallstudie med järbnmalm
Abstract [en]

The risks of starting, operating and closing mining projects have become higher than ever. In order to stay ahead of the competition, mining companies have to manage various risks: technical, environmental, legal, regulatory, political, cyber, financial and social. Some of these can be mitigated with the help of geometallurgy. Geometallurgy aims to link geological variability with responses in the beneficiation process by a wide usage of automated mineralogy, proxy metallurgical tests, and process simulation. However, traditional geometallurgy has neglected the non-technical aspects of mining. This has caused wide-spread discussion among researchers on the benefits of geometallurgy and its place in industry.

In order to improve predictability in geometallurgy, such programs should cover planning, and the testing of hypotheses, and only then should there be an attempt to develop suitable technical tools. Such approach would ensure that those tools would be useful and are needed, not only from the technical point of view, but also from the users’ perspective. Therefore, this thesis introduces methodology on how to decrease uncertainty in the production planning and thus determine how much effort to put into decreasing uncertainty in geometallurgical programs.

The predictability improvement of a geometallurgical program starts at the planning stage. The classification system developed here, through the survey (interviews) and literature review, indicates different ways to link geological information with metallurgical responses, and suggests areas where technical development is called for. The proposed developments can be tested before the start of the geometallurgical program with synthetic data. For the iron ore reference study (Malmberget), it was shown that implementation of geometallurgy is beneficial in terms of net present value (NPV) and internal rate of return (IRR), and building geometallurgical spatial model for the process properties (recovery and total concentrate tonnages), and that it requires fewer samples for making a reliable process prediction than concentrate quality. The new process and proxy for mineralogical characterisation models were developed as part of the geometallurgical program for the iron ore case study (Leveäniemi): an estimator of ore quality (ܺ௅்௎), a model for iron recovery in WLIMS, a model for iron-oxides liberation prediction. Additionally, it was found that DT may be applied only for studying marginal ores. The evaluation of different spatial process modelling methods showed that tree methods can be successfully employed in predicting non-additive variables (recoveries).

Place, publisher, year, edition, pages
Luleå: Luleå University of Technology, 2019
Series
Doctoral thesis / Luleå University of Technology 1 jan 1997 → …, ISSN 1402-1544
Keywords
Additivity, Apatite iron ore, AIO, Block model, Change of support, Classification, Data integration, DT, Feed quality, Geometallurgical program, Geometallurgy, Iron ore, Iron recovery, Leveäniemi, Liberation, Machine learning, Magnetic separation, Malmberget, Mineralogical approach, Mineralogy, Prediction, Proxies, Proxies approach, Sampling, Simulation, Synthetic ore body, Traditional approach, WLIMS
National Category
Mineral and Mine Engineering Metallurgy and Metallic Materials
Research subject
Mineral Processing
Identifiers
urn:nbn:se:ltu:diva-71580 (URN)978-91-7790-266-9 (ISBN)978-91-7790-267-6 (ISBN)
Public defence
2019-02-04, D770, Lulea, 10:00 (English)
Opponent
Supervisors
Available from: 2018-11-14 Created: 2018-11-14 Last updated: 2019-02-07Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records BETA

Lishchuk, ViktorLund, CeciliaGhorbani, Yousef

Search in DiVA

By author/editor
Lishchuk, ViktorLund, CeciliaGhorbani, Yousef
By organisation
Minerals and Metallurgical Engineering
In the same journal
Minerals Engineering
Mineral and Mine EngineeringMetallurgy and Metallic Materials

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 329 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf