Online control system for molten steel temperature based on combination of presetting and prediction
The temperature control of molten steel in steelmaking plant has a significant impact on production rhythm,casting blank quality,and energy costs.However,the current mainstream research on molten steel tem-perature control mainly focuses on single process,lacking consideration for process linkage and matching.To achieve precise control of molten steel temperature throughout the steelmaking process,based on investigating the factors influencing molten steel temperature and analyzing ladle thermal conditions,a case-based reasoning(CBR)model was developed for molten steel temperature setting,and a prediction model utilizing Kmeans clustering and backpropagation neural networks(BPNN),along with ladle thermal condition corrections,was established for fore-casting molten steel temperatures at critical nodes from the converter to continuous casting.By combining presetting and prediction,an online temperature control system for molten steel in steelmaking plants was constructed and inte-grated with the dynamic scheduling system of the steel plant.This system provides accurate recommendations and monitoring of molten steel temperatures during production.Application results in Tangshan Iron and Steel New Ar-ea Steelmaking Plant show that the system optimizes target molten steel temperatures,improves the accuracy of temperature predictions,optimizes production scheduling,and reduces fluctuations in molten steel temperatures.Specifically,the temperature drop from the end of converter smelting to the start of RH refining decreases from 44.8 ℃ to 37.4 ℃,and the temperature drop from the end of RH refining to the start of continuous casting decrea-ses from 23.2 ℃ to 21.7 ℃,achieving precise control of molten steel temperature,while providing temperature lev-el decision support for optimizing production process scheduling.
steelmaking plantmolten steel temperaturepredictionladle thermal conditiononline control
NETL
NSTL
万方数据
