论文部分内容阅读
LF精炼工序在炼钢过程起着调节温度的关键作用,准确预报LF精炼终点钢水温度对实际生产有重要意义.传统的LF精炼预报模型包括机理模型与黑箱模型.机理预报模型能够体现各工艺因素对终点钢水温度的影响,但由于LF精炼传热机理研究尚不完善,依靠机理模型预报终点钢水温度,难以达到预期效果;黑箱预报模型能够准确预报终点钢水温度,但不能反映精炼过程各工艺因素对钢水温度的影响,尤其当生产工艺条件发生改变时,黑箱模型在应用上会受到限制.本文以方大特钢LF精炼炉为研究对象,建立一种机理预报模型与黑箱预报模型(BP神经网络预报模型)相结合的LF精炼终点钢水温度灰箱预报模型.该模型既能反映各工艺因素对终点钢水温度的影响,又能准确预测终点钢水温度,其终点钢水温度预测误差在±5℃以内的命中率可以达到95%以上.
LF refining process plays a key role in temperature control during the steelmaking process. Accurately forecasting the temperature of the molten steel at the end of LF refining is of great significance to the actual production.Traditional LF refining forecasting model includes mechanism model and black box model.The mechanism forecasting model can reflect various process factors However, due to the incomplete research on the heat transfer mechanism of LF refining, it is difficult to predict the finish temperature of molten steel by means of mechanism model. The black box prediction model can accurately predict the temperature of the final molten steel, but can not reflect the process factors of the refining process Especially on the change of production process conditions, the black box model will be limited in its application.In this paper, the LF refining furnace of Fangda Special Steel is taken as the research object to establish a mechanism forecasting model and a black box forecasting model (BP nerve Which can not only reflect the influence of process factors on the temperature of the finish molten steel, but also accurately predict the temperature of the finish molten steel.The prediction error of the molten steel temperature at the end is ± 5 ℃ The hit rate can reach more than 95%.