16 elo SoreX – Service oriented automation for efficient rock and ore exploration
Name: Service oriented automation for efficient rock and ore exploration
Duration: 1.1.2012 – 31.8.2014
Total costs (1,6 M€) /Tekes support (1,1 M€)
Leading research organization partner: Aalto University, Department of Electrical Engineering and Automation
Contact persons: Kai Zenger (Aalto University, kai.zenger(at)aalto.fi), Matti Vilkko (Tampere University of Technology, matti.vilkko(at)tut.fi)
Research organization partners: Aalto University, Tampere University of Technology, University of Jyväskylä, Geological Survey of Finland
Company partners: Outotec Oy, Metso Minerals Oy, Pyhäsalmi Mine Oy, FQM Kevitsa Mining Oy, Agnico-Eagle Finland Oy, Mine Oli-Line Service Oy, Ima Engineering Ltd Oy.
International partners: Julius Kruttschnitt Mineral Research Centre (The University of Queensland, Australia)
Number of reviewed publications, incl. submitted manuscripts: 18
Number of thesis: Master 4, Bachelor 5
Contents and key results:
Need and motivation of the project (Exploration):
Sorex project concentrated on three main application areas, which cover the whole mineral processing work chain: exploration, crushing and mineral concentration. In the exploration phase the focus was on developing and improving the analysis methods for ore exploration and mining operations. The tendency of new mining projects is that the ore bodies are smaller and contain low-grade ore, which increases the challenges in locating the ore resources as accurately as possible. The demand of the mining and exploration companies is on faster and more versatile analysis of ore samples. Improved ore analysis shortens the exploration time and reduces the costs since the fast analysis results can be used as a feedback for controlling the exploration work. More advanced measurement methods and analytical techniques are required to achieve those goals. Therefore, the exploration phase of Sorex project focused on developing rapid and on-site measurement techniques which can be applied on the drill core samples in the mines and before sending the samples to the laboratories. Such developments greatly assist the geologists for logging and selecting the necessary drill cores rather than sending large volumes to the laboratories.
Information about mineral contents not only gives broad knowledge about source, history and type of the rock bodies, but also today is necessary for better process control and more efficient use of energy and raw materials. However, the majority of the online and on-stream analytical techniques are based on elemental analysis or performed off-line when applied to mineral identification. Developing new measurement setups only for the purpose of mineral identification is costly, needs proper space for installation in the plant and requires advanced data analyzing to extract mineralogical knowledge. Therefore, part of this research was focused on qualitative mineral identification of the samples accurately from the available elemental data.
Main set targets:
- Characterizing the drill core samples by combining suitable contactless and non-destructive techniques directly from the drill cores in the boxes. It is cost-effective, needs less manual work for sample preparation, less energy consumption and thus less environmental impact.
- Reviewing advanced data handling and fusion methods and developing new techniques for combining data from different sources. Special focus is on the management of large data sets which will be created by the various ore and drill core measurements from a wide range of ore types and mines. One aspect in this task will be the clear and illustrative visualization of the different data sets.
- Developing and implementing new data analysis techniques in order to extract mineralogical information from the available elemental data.
- Laser-induced breakdown spectroscopy (LIBS) was studied as a potential rapid on-line method for automated elemental analysis of drill core. It was shown that the LIBS method can produce similar elemental concentrations as the laboratory measurements.
- LIBS technique was applied as a complementary technique for hyperspectral imaging. Based on elemental contents, some minerals were identified where hyperspectral imaging did not measure them
- Laser-induced fluorescence imaging (LIFI) was studied and developed as a potential rapid analysis method for drill core analysis. The technique was combined with Raman spectroscopy so that LIFI was first calibrated with Raman technique and the combined techniques were used for estimation of the minerals abundance.
- Combined LIFI and Raman spectroscopy was applied to generate a map of the fluorescent minerals from the surface of the drill core samples rapidly. The results were compatible with laboratory XRD results
- A statistical analysis technique based on singular value decomposition was proposed for mineral identification from elemental results of LIBS technique. The mineralogical information achieved by LIBS technique were used as complementary information for LIFI technique
- Raman spectroscopy was applied on the drill core samples. The main results found were the Non-negative least squares (NNLS) method for studying the spectral data yielded by Raman analysis. The results were preliminary, however, but showing the potential of mathematical methods in the study of exploration phase of mining.
- Concentration of different minerals was successfully measured using spectral analysis of Raman results
- Data Envelopment Analysis (DEA) was introduced as a data fusion method for analysis. DEA was applied as a tool for measuring the sustainability in the exploration phase of mining
- DEA was used as a tool for studying the breakage characteristics of rocks by selecting the related variables
Need and motivation of the project (Crushing)
The crushing phase focused on sensor development and automatic control of crushing processes, which are traditionally operated without automatic control. The topic is important since almost all of the crushing processes are currently operated manually, without proper feedback from the size reduction process performance. It is therefore almost impossible to improve the existing operation and affect e.g. energy- and material efficiency of the circuit before proper process measurements and size reduction control algorithms have been developed.
Main set targets:
- To develop crushing process performance indices for quantitative analysis the circuit performance
- To develop self-calibrating and cost effective mass flow measurement sensor network for feedback control and circuit performance optimization purposes
- To develop advanced control schemes for size reduction control of cone crushers
- Developed sensor fusion-based data reconciliation and online calibration techniques for cost effective replacement of belt scales using inclined conveyor power measurements.
- Developed automatic size reduction control schemes that enable energy- and material efficient crushing, while maximizing the plant net profit.
Need and motivation of the project (Mineral concentration)
The utilization of a dynamic process simulator was investigated as a tool to assist the process operator. A key research target was the development of an on-line adaptation algorithm for the process model parameters. That is necessary for an accurate simulator, which can continuously follow the state of the process, based on the measurement data obtained from the process. The goal was to make it possible to use the simulator as a tool for operator training, to make fast “what if” analyses of the process possible when doing control actions, etc.
Main set targets:
- To develop an online adaptive parameter tuning algorithm for an existing process simulator of the mining process. To verify its performance in practical operation of the mine and in operator use.
- The targets were met at a sufficient level. However, the existing process simulator was not accurate enough and fast enough to fully meet the requirements set for this subproject.
- Itävuo, M. Vilkko, A. Jaatinen and K. Viilo. (2013). ‘Dynamic modeling and simulation of cone crushing circuits’, Miner. Eng., 43–44, pp. 29–35.
- Kauppinen, T., Khajehzadeh, N., Haavisto, O. (2014). Laser-induced fluorescence images and raman spectroscopy studies on rapid scanning of rock drillcore samples. International Journal of Mineral Processing 132 (10), 26-33.
- Itävuo, M. Vilkko, A. Jaatinen and K. Viilo. (2012). ‘Dynamic modeling and simulation of cone crushing circuits’, In Proc. of Comminution ’12, Cape Town, South Africa.
- Itävuo, M. Vilkko and A. Jaatinen. (2012). ‘Principles for optimal operating point selection in cone crushers’, In Proc. of the 9th International Mineral Processing Seminar (Procemin), Santiago, Chile, pp. 36–44.
- Itävuo, M. Vilkko and A. Jaatinen. (2012). ‘Specific energy consumption-based cone crusher control’, In Proc. of IFAC Workshop on Automation in the Mining, Mineral and Metal Industries, Gifu, Japan, pp. 78–83.
- Väyrynen, P. Itävuo, M. Vilkko, A. Jaatinen and M. Peltonen. (2013). ‘Mass-flow estimation in mineral-processing applications’, In Proc. of IFAC MMM, San Diego, USA, 2013.
- Itävuo, M. Vilkko and A. Jaatinen. (2013). ‘Indirect particle size distribution control in cone crushers’, In Proc. of IFAC MMM, San Diego, USA, 2013.
- Kaartinen, J. Pietilä, A. Remes, S. Torttila (2013). ’Using a Virtual Flotation Process to Track a Real Flotation Circuit’, In Proc. of IFAC MMM, San Diego, USA, 2013.
- Pietilä, J. Kaartinen, A-M. Reinsalo (2013). ’Parameter estimation for a flotation process tracking simulator’, In Proc. of IFAC MMM, San Diego, USA, 2013.
- Kauppinen, O. Haavisto, H. Häkkänen (2013). ’Optimisation algorithms in the case of mineral detection using Raman Analysis’, In Proc. of IFAC MMM, San Diego, USA, 2013.
- Haavisto, T. Kauppinen, H. Häkkänen (2013). ’Laser-Induced Breakdown Spectroscopy for Rapid Elemental Analysis of Drillcore’, In Proc. of IFAC MMM, San Diego, USA, 2013.
- Kauppinen, J. Jackson, E. Manlapig, J. Lay (2014). ’Data fusion of rock breakage using Data Envelopment Analysis’. In Emrouznejad, A., R. Banker R., S. M. Doraisamy and B. Arabi (2014), Recent Developments in Data Envelopment Analysis and its Applications, Proceedings of the 12th International Conference of DEA, April 2014, Kuala Lumpur, Malaysia, ISBN: 978 1 85449 487 0
- Itävuo, T. Väyrynen , M. Vilkko and A. Jaatinen. (2014). ‘Strategies for size reduction control in cone crushers’, In Proc. of Comminution ’14, Cape Town, South Africa, 2014.
- Väyrynen, P. Itävuo, A. Jaatinen , M. Vilkko and M. Peltonen. (2014). ‘Adaptive mass flow sensor calibration method for crushing circuits’, In Proc. of Comminution ’14, Cape Town, South Africa, 2014.
- Itävuo, T. Väyrynen , M. Vilkko and A. Jaatinen. (2014). ‘Tight feed-hopper level control in cone crushers’, In Proc. International Minerals Processing Congress, Chile, 20.-24.10.2014
- Kauppinen, O. Haavisto, N. Khajehzadeh, H. Häkkänen (2014). ’Raman analysis: detection of minerals and measurement of concentration using reference spectra.’ In Proc. International Minerals Processing Congress, Chile, 20.-24.10.2014
- Khajehzadeh, T. Kauppinen, H. Häkkänen (2014). ’Laser-induced fluorescence imaging: detection of fluorescent minerals and estimation of abundance.’ In Proc. International Minerals Processing Congress, Chile, 20.-24.10.2014
- Pietilä, A. Remes, J. Kaartinen, S. Torttila, J. Huuskonen, S. Lähteenmäki, K. Zenger (2015). ’On-line Flotation Simulator at Pyhäsalmi Mine’, 47th Canadian Mineral Processors Conference, Ottawa, Canada, 20-22 Jan 2015
- Navid Khajehzadeh and Kai Zenger, ”Rapid mineral detection using elemental Laser-Induced Breakdown Spectroscopy,” in Proc. Automaatio XXI 2015 seminaari, Helsinki, March17-18,2015, 2015
- Olli Haavisto, Tommi Kauppinen and Navid Khajehzadeh, ”Rapid rock drillcore analysis,” in Automation and Systems without Borders – beyond Future, Matti Vilkko (ed.), Helsinki: Suomen Automaatioseura ry, 2013, 6pp.
- Tommi Kauppinen and Olli Haavisto, ”Raman analysis on rock drillcore,” in Automation and Systems without Borders – beyond Future, Matti Vilkko (ed.), Helsinki: Suomen Automaatioseura ry, 2013, 6 pp.
- Teemu Väyrynen: Mass flow estimation in mineral processing applications. Diplomityö, Tampereen teknillinen yliopisto, 2013
- Ari-Matti Reinsalo: Continuous parameter estimation for online process simulator. Diplomityö, Aalto-yliopisto, 2013
- Sampo Torttila: Adaptoituva vaahdotusprosessin simulaattori. Diplomityö, Aalto-yliopisto, 2013
- Mikael Siirtola: Magnetic susceptibility measurement of drillcore. Diplomityö, Aalto-yliopisto, 2013
- Juho Leinonen: Automaattinen parametrien viritys vaahdotusprosessissa. Kandidaatintyö, Aalto-yliopisto, 2012
- Severi Haverila: Konenäkösovellukset mineraalirakeiden tunnistamisessa. Kandidaatintyö, Aalto-yliopisto, 2012
- Juuso Meriläinen: Mineraalien rikastusprosessien simulointi. Kandidaatintyö, Aalto-yliopisto, 2013
- Eero Siivola: Virtuaalimittaukset mineraalirikastuksessa. Kandidaatintyö, Aalto-yliopisto, 2014
- Antti Paloposki: Raman analyysi mineraalien tunnistuksessa. Kandidaatintyö, Aalto-yliopisto, 2014