基于轮廓向量集和遗传算法的高炉炉缸内衬侵蚀预测模型

A Mathematical Model for Estimating the Lining Erosion of Blast Furnace Hearth Based on Profile Vectors and Genetic Algorithm

针对高炉炉缸连续式生产、工况恶劣以及耐火材料内衬实际形貌在生产过程中不断变化的特点,提出了一种利用轮廓向量集表征炉缸内衬热面形状的简易方法,从而将求解未知定温边界几何形状的复杂反问题归结为搜索最佳轮廓向量集合的最优化数学问题.结合数值传热学、有限元法以及遗传算法,建立了能够准确预测高炉炉缸内衬轮廓的传热"反问题"数学模型.在陶瓷杯复合炉缸的基础上,制备典型的非均匀"象脚状"异常侵蚀数值样本,对模型的有效性、稳定性以及计算结果的精确度进行了校验.结果表明,上述模型具有广阔的实际应用前景.

With respect to consecutive drainage, aggressive internal conditions and varying lining profile during production process in blast furnace hearth, a simple method is proposed for representing the lining profile by a set of lining profile vectors. It can convert the solving of a complicated inverse heat transfer problem with unknown geometries at a specified-temperature boundary into searching the optimal set of profile vectors in an optimization problem. Integrating numerical heat transfer theory and finite element method with genetic algorithm, an inverse problem oriented mathematical model has been developed, which can predict the lining profile of blast furnace hearth accurately. A testing example of elephant-foot erosion pattern on the basis of an intact ceramic cup synthetic hearth, is employed to examine the effectiveness, stability and accuracy of the model. The results indicate that the model would have a broad prospect of practical applications.