Robust soft measurement method for quality of blast-furnace hot metal

Abstract

The invention provides a robust soft measurement method for the quality of blast-furnace hot metal. The robust soft measurement method comprises the following steps: acquiring blast furnace bosh gas volume, cold-blast air flow, oxygen enrichment flow, breathability, oxygen enrichment rate and theoretical combustion temperature at the current moment; carrying out normalization processing on acquired data; and carrying out robust soft measurement of the blast-furnace hot metal by using a blast-furnace hot metal quality robust soft measurement model constructed by a multiplex random weight neutral network to obtain a Si content estimation value, a P content estimation value, an S content estimation value and a hot metal temperature estimation value. According to the robust soft measurement method provided by the invention, blast-furnace body parameters obtained by real-time measurement are used as input data of the model, the sequential relationship between the hysteresis characteristic of the blast-furnace smelting process and input and output variables is sufficiently considered, and the blast-furnace hot metal quality robust soft measurement model with a nonlinear autoregression structure is constructed; and moreover, robust soft measurement of hot metal quality parameters such as Si content, P content, S content and hot metal temperature is realized, and hysteresis caused by offline testing and uncertainty caused by manual operation are avoided.

Description

technical field [0001] The invention belongs to the technical field of automatic control of blast furnace smelting, and in particular relates to a method for robust soft sensing of blast furnace molten iron quality. Background technique [0002] Blast furnace ironmaking is an extremely complex nonlinear dynamic process of reducing iron from iron ore and other iron-containing compounds to smelt molten iron of acceptable quality. As the most important production index in the blast furnace ironmaking process, the molten iron quality index directly determines the quality of subsequent steel products and the energy consumption state of the blast furnace smelting process. In order to achieve the goals of high quality, low consumption, high yield and long life, it is necessary to monitor and control the blast furnace ironmaking process in real time. At present, parameters such as silicon [Si] content (chemical heat), molten iron temperature (physical heat), sulfur [S] content, and...

AI Technical Summary

Problems solved by technology

These methods mainly consider the soft sensing of molten iron quality parameters under ideal furnace conditions, and their robustness is poor. When the modeling data contains outliers, these methods cannot suppress the interference of outliers and predict the molten iron quality parameters more accurately.

[0007] The patent application number "201610118914.7" applied for "a multivariate molten iron quality soft sensor method based on robust random weight neural network" can predict the multivariate molten iron quality parameters and reflect the inherent characteristics of the blast furnace ironmaking process, but it only solves the problem of modeling training. The output of the data contains outliers (abnormalities in the Y direction), but it cannot solve the robustness problem that the input of the training data also contains outliers (abnormalities in the X direction).

When the input and output data in the modeling training data contain outliers (that is, both X and Y directions are abnormal), the robustness of the modeling method will be poor

At the same time, it cannot solve the problem that the multicollinearity of the output of the hidden layer has an adverse effect on the modeling

To sum up, at present, there is no multivariate robust soft-sensing method for molten iron quality parameters (Si content, P content, S content and molten iron temperature) in the blast furnace smelting process at home and abroad.

Images