Abstract

The results of a passive industrial experiment on coking at EVRAZ NTMK are analyzed using a selective matrix, rather than the whole data set, taking account of a comprehensive factor affecting the M₂₅ and M₁₀ values of the coke: the coking index of the batch CI^ν0. In this approach, broad variation in the response function (M₂₅ or M₁₀) is ensured with a symmetric matrix, ruling out the predominance of a particular range of CI^ν0 or the M₂₅ and M₁₀ values. Applying this approach to actual data for the industrial coking of batch at coke batteries 5–6 (wet quenching) and 9–10 (dry quenching) in a passive industrial experiment, we obtain mathematical models for predicting the quality of wet- and dry-quenched industrial coke (the M₂₅ and M₁₀ values) as a function of the batch properties (CI^ν0), with the existing conditions of coal preparation and coking at those batteries. Verification of the models for predicting the quality of wet- and dry-quenched coke (the M₂₅ and M₁₀ values) on the basis of extensive data for industrial coking at batteries 5–6 (62 runs) and 9–10 (58 runs) shows that the M₂₅ and M₁₀ values are predicted with acceptable accuracy for practical purposes.

中文翻译：

---

在被动工业实验中预测 EVRAZ NTMK 的湿熄焦和干熄焦的质量。2. 预测 M25 和 M10

摘要

在 EVRAZ NTMK 的焦化被动工业实验的结果使用选择性矩阵而不是整个数据集进行分析，同时考虑了影响焦炭M ₂₅和M ₁₀值的综合因素：批次的焦化指数CI ^ν0。在这种方法中，通过对称矩阵确保响应函数（ M ₂₅或M ₁₀ ）的广泛变化，排除了 CI ^ν0或M ₂₅和M _{10的特定范围的优势}价值观。将这种方法应用于被动工业实验中焦炭电池 5-6（湿熄焦）和 9-10（干熄焦）批次的工业焦化实际数据，我们获得了用于预测湿熄焦和干熄焦质量的数学模型。淬火工业焦炭（M ₂₅和M ₁₀值）作为批次特性 (CI ^ν0 ) 的函数，以及这些电池组的煤炭制备和焦化的现有条件。根据电池 5-6（62 次运行）和 9-10（58 次运行）的工业焦化的大量数据，验证用于预测湿熄焦和干熄焦质量的模型（M ₂₅和M ₁₀值） ) 表明M ₂₅和M ₁₀值的预测精度对于实际目的来说是可以接受的。