Intelligent Modeling Method For High-charge Surface Temperature Field

The burden surface temperature field of Blast Furnace (BF) is the most direct manifestation of the distribution of gas flow in the throat of BF. However, due to the complex physical, chemical reaction and dynamic process, it is difficult to establish correct model of burden surface temperature field, which is of very importance in forecasting the conditions of gas flow, optimizing production operation and ensuring stable operation of BF. This study has taken on great significance in academic research and good prospect in applications.The relationships between burden surface temperature field and multi-source information, such as infrared image, crossing temperature, stock rod, gas temperature from riser tube, etc. are analyzed by mechanism analysis. Based on the infrared image, feature extraction techniques of isotherm, central position and radial temperature distribution in burden surface temperature field are studied. Aiming to solve the imprecision problem of conventional calibration methods that only make use of one kind of detecting information, a novel method for modeling burden surface temperature field in BF is presented. It is based on information fusion and makes full use of information detected from the throat of BF. A dynamic temperature calibration method based on two-point method is taken as the benchmark calibration method, in which the nonlinear error involved is adjusted by an improved BP neural network based on Genetic Algorithm (GA). This method has a lot of advantages, such as more precision of temperature calibration, simpler neural network structure, less computation and higher convergence rate. The simulation results show that, this model of burden surface temperature field can depict distribution of temperature field more exactly than conventional calibration methods, and also confirm the accuracy and feasibility of this method.Using the proposed modeling method, a real-time monitoring system of burden surface temperature field in BF is developed and successfully operated in Lianyuan Steel 2200m~3 BF. The visualization interface of monitoring system is more effective to understand the distribution of burden surface temperature field and instruct the operation of burden distribution.Finally, this thesis makes some conclusions, and presents some issues for future research.