Due to decreasing iron ore qualities and reduced availability of lump ore, iron ore pellets are becoming increasingly important as valuable agglomerates for blast furnaces and DRI production. Pellets fit in any burden at any quantity and add value to the steelmaking process. The production of pellets requires frequent monitoring of its mineralogical, chemical and physical properties. X-ray diffraction (XRD) was tested as an alternative to a) quantify and monitor mineral composition, with special focus on the magnetite content of the pellets and b) correlate metallurgical parameters such as swelling or compression directly with the XRD measurement. Recently statistical methods like partial least square regression (PLSR) in combination with XRD raw data have become more and more popular to handle large amount of data and to correlate process relevant parameters directly with an XRD measurement. Analysis within minutes compared to time and labour intense traditional physical and chemical methods as well as the additional information about phase composition enable efficient manufacturing of iron ores pellets. Results show that XRD is a useful tool to determine the magnetite content in green and fired pellets. Preliminary results suggest swelling and compression can be predicted without any additional sampling and measurement.
CITATION:Knig, U, 2019. Process monitoring for iron ore pelletizing XRD in combination with PLSR, in Proceedings Iron Ore 2019, pp 121126 (The Australasian Institute of Mining and Metallurgy: Melbourne).