Research On Prediction Of Key Parameters Of Blast Furnace Smelting And Classification Method Of Furnace Condition

The iron and steel smelting is a highly energy intensive and highly polluting industry.In the industrial transformation,the concept of green manufacturing should implement into the sintering,coking,iron-making,steel making and other major links,and use the advanced control technology to improve production efficiency and reduce pollution emissions,and there are the considerable meaning and actually application worthy.In the article,at the first,the basic flow,complexity and abnormal condition of the blast furnace iron making are comprehensively analyzed.At this step,integrate with the theory analysis way and the number analysis way,and in view of the particle swarm optimization algorithm is easy to fall into the local extreme point,the particle swarm optimization is improved to enhance its egocentricity.And the least squares plump vector machine is selected as the principle part of the method of blast furnace hot metal temperature prediction model,which is optimized by the improved particle swarm optimization.It realized a accurate prediction of the hot metal temperature in the blast furnace and improved the prediction accuracy.At the same time,the blast furnace temperature indirectly reflects the temperature of the blast furnace hot metal.Secondly,according to the complexity of blast furnace production and operation data,the deep learning algorithm is introduced into the blast furnace smelting process by using the excellent ability of deep learning to do well with data.And then,the improved particle swarm algorithm is used to revise the parameters of the deep belief network.So the prediction model of blast furnace condition is established by using the deep belief network as the main body of the model.Finally,the validity and accuracy of the blast furnace hot metal temperature prediction model and the blast furnace condition classification prediction model are verified by simulation analysis.It has been an important guiding role for the blast furnace smelting process control.At the same time,according to the operating parameters,and the state parameters as well as the deep belief network furnace condition classification forecasting model of the blast furnace smelting process,a comprehensive monitoring system of the blast furnace condition is designed.With the actual production parameters of the monitoring,the furnace conditions can predict in a timely manner by Visual Studio2010 software platform,it has important reference value in the blast furnace iron-making process control and management of energy-saving emission reduction.