Research on the Quality Improvement and Consumption Reduction of Iron Ore Agglomeration Based on Optimization

Mingshun Zhou, Desheng Zhao, Jiangning Zhang, Guang Yang, Enjian Hou, Mingxu Liu, Hui Zhang, Xin Jiang, Kai Fan, Fengman ShenMetals 2023 · DOI

TL;DR

This paper presents an intelligent blending method and process optimization strategies for iron ore agglomeration at Ansteel, aiming to boost production quality and efficiency while cutting energy consumption. By employing statistical modeling and a novel drum granulator, they achieved significant improvements in sinter yield, stability, and reduced fuel usage, demonstrating a path towards more sustainable and cost-effective ironmaking.

What Problem Does This Paper Address?

Modern blast furnaces demand high-quality sinter for efficient operation, but fluctuating raw material quality and scarcity pose significant challenges to sinter plants. This instability can lead to increased fuel consumption, reduced hot metal productivity, and higher operational costs. The paper addresses the critical need for optimizing the iron ore agglomeration process, specifically focusing on improving sinter quality and reducing energy consumption, which are key to the economic viability and environmental performance of blast furnace ironmaking. The authors highlight that traditional blending methods, often dictated by logistical constraints rather than optimal material properties, are no longer sufficient.

Methods

The research employs a multi-pronged approach to optimize iron ore agglomeration:

  • Statistical Modeling and Optimal Experimental Design: A statistical model for ore blending was developed using optimal experimental design principles. This allowed for the analysis of single-factor influences of different iron ores on sinter productivity and quality, leading to the derivation of optimal blending ratios.
  • Process Transformation (Drum Granulator): Recognizing the limitations of traditional disc pelletizers in achieving efficient mixing, Ansteel implemented a massive technical transformation by replacing 16 disc pelletizers with a single, large-scale drum granulator. This aimed to improve the uniformity and stability of the sinter mix.
  • Single Lattice Method for Fuel Optimization: To further enhance sinter quality and reduce fuel consumption, a single lattice method was used to optimize the particle-size distribution of coke breeze. This involved systematically testing different size fractions and their combinations.
  • Deep Bed Sintering: The study also touches upon achieving increased bed depths (1050 mm or more) in sintering, aligning with international trends for higher productivity.

Datasets

The study utilizes data from Ansteel's Sinter Plant 3 and laboratory experiments. Specific datasets include:

  • Sinter Pot Test Data: Used to evaluate the performance of different coke breeze size distributions and the impact of the drum granulator.
  • Raw Material Chemical Composition Data: Provided for various iron ore types (Ore A, B, C, D) and their basicity.
  • Historical Sinter Grade Data: Presented as a variation of sinter grade in Ansteel since 2000 (Figure 1).
  • Experimental Data from Optimization Studies: Used to build regression models for various sintering parameters.

No new benchmark datasets were introduced; rather, existing plant and laboratory data were leveraged for analysis and optimization.

Key Insights and Ideas

  • Ore Blending is Crucial and Quantifiable: The research demonstrates that a statistically optimized blend of specific iron ores can significantly improve sinter yield. The finding that a precise ratio of Ore A and Ore C (47.2% and 52.8%) yields the best sinter yield (72.44%) underscores the importance of precise ore selection.
  • Basicity is Key for Reducibility: Beyond yield, sinter reducibility is vital. The study reveals a critical threshold for sinter basicity (not less than 1.98) when a significant proportion of Ore A is used, highlighting a trade-off between yield and reducibility that needs careful management.
  • Drum Granulators Revolutionize Mixing: The shift from disc pelletizers to a drum granulator dramatically improved the uniformity and stability of the sinter mix. This is evidenced by a substantial reduction in the standard deviation of key parameters like total Fe and basicity, leading to more consistent sinter quality.
  • Coke Breeze Size Matters (A Lot): Optimizing the particle size distribution of coke breeze, not just its total amount, is a powerful lever for improving sinter tumble strength. The identified optimal distribution (e.g., 57.20% <1 mm) suggests that fine-tuning fuel particle size can lead to better sinter mechanical properties.
  • Holistic Optimization Pays Off: The paper emphasizes that achieving optimal sinter performance requires considering multiple factors simultaneously – ore blend, granulation technology, and fuel particle size – rather than optimizing them in isolation.

Key Results

  • Optimized Ore Blend: A blend of 47.2% Ore A and 52.8% Ore C achieved the best sinter yield of 72.44%.
  • Drum Granulator Impact:
    • Standard deviation for total Fe (T.Fe) decreased from 1.517 to 0.7332.
  • Standard deviation for basicity decreased from 0.146 to 0.0956.
  • Compared to disc pelletizers, the drum granulator showed lower standard deviations for T.Fe (0.6926 vs. 0.8902) and basicity (0.05449 vs. 0.2033).
  • Productivity and Fuel Savings: After adopting the optimized sinter blend, Sinter Plant No 2 saw a productivity increase of 0.07 t/m²h and a reduction in fuel rate by 0.9 kg/t.
  • Optimal Coke Breeze Distribution: For maximum sinter tumble strength, the coke breeze should ideally consist of approximately 57.20% (<1 mm), 25.63% (1–3 mm), 11.17% (3–5 mm), and 6.00% (>5 mm) particles.
  • Deep Bed Sintering Achieved: Ansteel successfully achieved bed heights of 1050 mm or more, typically with blends containing 75% concentrates.

Limitations and Future Work

  • Limitations:
    • The statistical model's accuracy is within a defined statistical range.
  • The paper acknowledges that in actual production, ore availability, bin capacity, and plant layout are crucial constraints that influence blending decisions beyond theoretical optimization.
  • While the drum granulator is presented as superior, the paper doesn't delve into the specific energy consumption differences between the two granulation methods.
  • Future Work:
    • Development of a comprehensive evaluation system for iron ore sinter fines, integrating sinter characteristics, cost information, and blend optimization capabilities.
  • Further exploration of new sintering technologies, such as oxygen-enriched sintering and gas-fuel-injected sintering.
  • Continued research into advanced techniques like double-layer sintering for even deeper beds and higher productivity.
  • The paper suggests that the optimization of coke breeze size distribution can be tailored to maximize other sintering parameters beyond just tumble strength, indicating potential for further fine-tuning.

Summary updated

Abstract

In order to increase agglomeration production and improve its quality while reducing energy consumption, a new intelligent blending method has been established at Ansteel to optimize its sinter blends. The statistical model of the ore blending results shows that (1) a blending ratio of 47.2% Ore A and 52.8% Ore C corresponded to the best sinter yield of 72.44%. (2) From the viewpoint of sinter reducibility, sinter basicity should not be less than 1.98 when the proportion of Ore A in the blend is more than 35%. Due to the low mixing efficiency of disc pelletizers, Ansteel has therefore gone through a massive technical transformation at Sinter Plant 3 to replace its 16 existing disc pelletizers with one drum granulator. Since the installation of the drum granulator, the standard deviation has decreased from 1.517 to 0.7332 for total Fe (T.Fe) and from 0.146 to 0.0956 for basicity. In the case of the drum granulator, the standard deviation for sinter T.Fe and basicity were 0.6926 and 0.05449, respectively, as compared to 0.8902 and 0.2033 for the disc pelletizers. In addition, a single lattice method is proposed to optimize the particle-size distribution of the coke breeze to further improve sinter quality and reduce fuel consumption. The lattice method indicated that the optimum coke breeze to achieve maximum sinter tumble strength should consist of approximately 57.20%: −1 mm, 25.63%: 1–3 mm, 11.17%: 3–5 mm, and 6.00%: &gt;5 mm particles. Given the international trend of increasing bed depth, Ansteel has successfully achieved a bed height of 1050 mm or more under its blend conditions, which typically contain 75% concentrates. Finally, some new iron ore agglomerations research is discussed.
BibTeX
@inproceedings{zhou2023research,
  author = {Mingshun Zhou and Desheng Zhao and Jiangning Zhang and Guang Yang and Enjian Hou and Mingxu Liu and Hui Zhang and Xin Jiang and Kai Fan and Fengman Shen},
  title = {{Research on the Quality Improvement and Consumption Reduction of Iron Ore Agglomeration Based on Optimization}},
  booktitle = {Metals},
  year = {2023},
  doi = {10.3390/met13030480},
}
Figure