Mezclas eficientes de mineral de hierro destinado a la corrección química del cemento a partir de pilas longitudinales tipo Chevron
The chemical composition of iron ore makes that this raw material is used in the cement industry as a correction agent. In addition to iron, iron ore also includes elements such as silica, aluminum, phosphorous and calcium, among others. The efficiency of an iron ore blending is a bottleneck for bla...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Estado: | Versión actualizada desde la publicación |
| Fecha de publicación: | 2019 |
| País: | Colombia |
| Institución: | Universidad Nacional de Colombia |
| Repositorio: | Repositorio UN |
| Idioma: | español |
| OAI Identifier: | oai:repositorio.unal.edu.co:unal/77588 |
| Acceso en línea: | https://repositorio.unal.edu.co/handle/unal/77588 |
| Access Level: | acceso abierto |
| Palabra clave: | 000 - Ciencias de la computación, información y obras generales::003 - Sistemas Ore blending Chevron piles Cement correction Iron ore Blending planning Mezclas de mineral Pilas Chevron Corrección del cemento Mineral de hierro Planificación de mezclas |
| Sumario: | The chemical composition of iron ore makes that this raw material is used in the cement industry as a correction agent. In addition to iron, iron ore also includes elements such as silica, aluminum, phosphorous and calcium, among others. The efficiency of an iron ore blending is a bottleneck for blast furnace performance quality indices, hence the need to understand the behavior chemical components of this mineral to support blending planning in an operation mining, with the purpose of maintain a stable process in cement plants. Statistical and mathematical tools, such as Linear Regression and Geostatistical Simulation, are often used to analyze the variance and quality of the iron ore blending. Goal Programming, Genetic Algorithms and Multi-Objective Stochastic Programming are used commonly to minimize the standard deviation of each of the chemical components in a blending. Although these tools establish the quality of the blending, they do not relate and optimize the mining sequence, the storage system and the distribution of mineral resources to form an efficient and satisfactory blending for the decision-maker. The proper identification of the resulting chemical composition in the iron ore blending can be accomplished by applying decision analysis strategies such as Operations Research. The current thesis proposes the development of a Mixed Integer Programming Model to analyze the iron ore blending to chemical correction of cement. The analyzed material come from Chevron piles from different mining fronts and different Fe2O3 content. Mixed Integer Programming correlates parameters, restrictions of the mining blocks, work fronts, storage system and iron ore demand under a timing-planning scheme. The model also designs the sequence of extraction of mineral blocks contained in different mining fronts, structures a system that allows to punish or financially reward the resulting blending according to the content of SiO2 and Al2O3; and it establishes hypothetical scenarios that allow make the best decision, in financial terms, according to the chemical composition of the blending. The objective function of the model maximizes the income of the operation from the demand in a time horizon, according to the quality specifications requested by the client. Finally, the model was optimized in the AIMMS software and the respective results were analyzed. As a main conclusion, it is had that, in some cases, the mining sequence produced unintuitive decisions for the decision maker, but decisive for the maximization of the benefits of the mining activity, such as the mining of blocks partially in different periods; decisions that in reality and in a discretionary way, are difficult to foresee. |
|---|