钢水温度的精确管控有利于提高铸坯质量和降低生产成本.针对目前炼钢—连铸区段钢水温度在线管控方面存在的不足,在分析钢水温度影响因素的基础上,建立了基于案例推理的炼钢—连铸区段钢水温度在线管控模型.同时通过调整案例推理算法的相似度计算方法、权重计算方法、重用案例个数等参数提高模型的精度.结果表明:转炉出钢温度预定模型平均降低转炉出钢温度6℃,精炼结束温度预定模型提高连铸开浇温度命中率2.33%.精炼开始温度,精炼结束温度和连铸开浇温度预报模型误差小于10℃的命中率分别达到75.33%、98.33%和95.67%,且均高于神经网络模型.
The accurate management and control of molten steel temperature is benefit to improve the slab quality and reduce production cost.Online management and control model of molten steel temperature by adopting case-based reasoning was established,which was based on the combination of the analysis of various factors affecting the temperature and the limitations of the status of temperature management and control for BOF-CC process.The precision of model was optimized by adj usting the parameters of the case-based reasoning algorithm such as simi-larity calculation method,the weight calculation method and reusing case amount.The result shows that the pre-setting model of BOF tapping temperature reduces tapping temperature by 6℃,the hit radio of the start tempera-ture of continuous casting increases by 2.33% on the basis of presetting model of refining end temperature.The hit radio of prediction model of the start temperature of secondary refining,the end temperature of secondary refi-ning and the start temperature of continuous casting respectively are 75.33%,98.33% and 95.67%,all higher than other neural network model.
参考文献
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%