A model-free adaptive control system and method for blast furnace multi-element molten iron quality

Abstract

The invention provides a model-free adaptive control system and method for multi-element molten iron quality in a blast furnace. Set the expected value of the multi-element molten iron quality index of the blast furnace; obtain the blast furnace body production condition parameter variables at several historical moments required for parameter correction of the model-free adaptive controller and the measured values ​​of the multi-element molten iron quality index at the corresponding time. For the sensitive model-free adaptive controller Parameters and initial values ​​of pseudo-partial derivatives are corrected offline; a model-free adaptive controller is used to online control the molten iron quality index to track the set expected value of the multi-element molten iron quality index of the blast furnace. Combining the model-free adaptive control technology to establish an online recursive controller for the multi-element molten iron quality of the blast furnace; and establish a recursive subspace prediction model between the multi-element molten iron quality of the blast furnace and the two control quantities as an auxiliary model for parameter correction, using multi-parameter sensitivity analysis, The genetic algorithm offline corrects the controller parameters and the initial value of the pseudo-partial derivative to adapt it to the current dynamic characteristics of the blast furnace and achieve effective control of the quality of multi-element molten iron.

Description

technical field [0001] The invention belongs to the technical field of automatic control of blast furnace smelting, and in particular relates to a model-free adaptive control system and method for multi-element molten iron quality of a blast furnace. Background technique [0002] The ultimate purpose of blast furnace ironmaking is mainly to produce high-quality pig iron, which provides qualified primary products for downstream processes such as converter steelmaking. Therefore, the optimal control of blast furnace ironmaking usually also means the control of the quality of molten iron, the final product of the blast furnace. At present, the quality index of molten iron is usually measured by Si content, molten iron temperature, S content and P content. Among them, Si content and molten iron temperature are important parameters to characterize the chemical heat and physical heat of the blast furnace, respectively. The silicon content of molten iron has been the main indicat...

AI Technical Summary

Problems solved by technology

However, the blast furnace smelting system is an uncertain dynamic time-varying system. Changes in ore grade and working conditions will lead to changes in the model parameters. Therefore, the offline model cannot fully reflect the dynamics of the blast furnace at different times. The controller is only suitable for specific parameters of blast furnace conditions, unable to meet the control requirements of the entire production process, the actual application value is low

In addition, the data model based on the existing control method has a large calculation complexity and a long calculation time, which is not suitable for online control and is difficult to apply to real industrial sites

To sum up, there is currently no multivariate dynamic on-line control method for hot metal quality indicators Si content and hot metal temperature in the blast furnace smelting process at home and abroad.

Images