高炉煤气流机理与数据建模技术炉况分析策略研究

Blast furnace (BF) ironmaking is a key link in the process of iron and steel smelting. The stable and smooth BF status is a guarantee for the optimal BF operation. The BF gas carry abundant and timely information, which is the important reference for the BF operation. The modeling of blast furnace gas of blast furnace condition is studied in the project. Firstly, the characteristics of gas data and the BF condition are deeply studied from mechanism property, and the relationship between the gas and the BF condition is established. Four kinds of furnace condition are analyzed from the data and mechanism property. Secondly, multi-granularity and fractal dimension method is used in analysis of gas data indexes variation. Based on the grey correlation algorithm and independence test, compute the degree of grey correlation in the indexes and select the ones that have significant effect on gas abnormal BF condition. Later, according to the kinds of data noise, two types of gas flow index including furnace top and bottom are studied. Meanwhile the appropriate structure models are selected and model framework is established by Bayesian method. Finally, on the basis of the above theory research, develop furnace condition analysis and prediction system. Build the experiment platform, and applied to blast furnace for test in the liuzhou iron &steel group.

高炉炼铁是钢铁生产过程的关键环节，长期顺行的炉况是高炉操作的先决条件。高炉煤气流作为高炉操作的重要依据，具有携带信息量丰富、时效性好的优点。本项目通过对高炉煤气建模来对高炉炉况进行研究。首先，从机理角度定性地分析煤气流指标与炉况的关系，并从数据和机理角度分析4种炉况的特征，从而确定煤气流指标与炉况的确切关系。其次，通过多粒度分形策略分析煤气流指标的变化特征，利用灰色关联度分析法剔除相互关联性较大的煤气流指标并对所得到的指标集合进行独立性检验，从而为煤气流指标建模确定最优指标集合。进一步，根据数据的噪声的类型，将所研究的高炉煤气流指标分成炉内和炉顶两种类型，同时选择合适的结构模型，利用贝叶斯技术构建模型框架。最后，开发炉况分析系统并搭建实验平台，并将该系统在柳钢高炉现场测试。

高炉炼铁是钢铁生产过程的关键环节，长期顺行的炉况是高炉操作的先决条件。高炉煤气流作为高炉操作的重要依据，具有携带信息量丰富、时效性好的优点。本项目充分利用煤气流信息的优点，融合机理分析方法和数据建模技术，针对封闭环境下大时滞非线性高炉系统的炉况诊断和预测这一关键科学问题进行深入研究。首先，从机理角度定性地分析煤气流指标与炉况的关系，并从数据和机理角度分析4种炉况的特征，从而确定煤气流指标与炉况的确切关系。其次，通过多粒度分形策略分析煤气流指标的变化特征，利用灰色关联度分析法剔除相互关联性较大的煤气流指标并对所得到的指标集合进行独立性检验，从而为煤气流指标建模确定最优指标集合。进一步，根据数据的噪声的类型，将所研究的高炉煤气流指标分成炉内和炉顶两种类型，同时选择合适的结构模型，利用贝叶斯技术构建模型框架。我们提出了基于贝叶斯技术的T-S模糊模型，对于数据分簇算法得到的大量模糊规则，对其赋予分层稀疏先验，并且考虑了T-S模糊模型的块结构特点，然后利用贝叶斯推理进行模糊规则的稀疏化，得到泛化能力很强的T-S模糊概率模型，并且模型能够给出概率输出。此方法与之前的T-S模糊模型建模方法具有很大优势。最后，开发炉况分析系统并搭建实验平台，并将该系统在柳钢高炉现场测试。本项目的完成将丰富和发展复杂工业过程的建模理论，研究成果有望为高炉炉况的准确诊断提供理论支撑和技术支持。