Self-adaptive sinter nature prediction method based on model parameters

Abstract

The invention discloses a self-adaptive sinter prediction method based on model parameters. According to the method, blend physicochemical indexes and historical data of corresponding sinter nature are collected; the relation between the blend physicochemical indexes and the sinter nature is fitted out through an RBF neural network algorithm and is saved as a function; the blend physicochemical indexes are input to predict the sinter nature; and a sinter matching database is established to achieve the self-adaptation function of a model. Through the method, the sinter nature is predicted quickly and accurately based on data modeling, automatic update of the model parameters is realized, and a basis is provided for ore matching strategy production. Meanwhile, the existing prediction methoddepending on manual experience judgment is changed, and the working efficiency of an enterprise is improved.

Description

technical field [0001] The invention belongs to the technical field of sinter property prediction, and in particular relates to a sinter property prediction method based on model parameter self-adaptation. Background technique [0002] The sintering process is a complex process industrial production process, which includes both physical and chemical processes. Among them, there are many influencing factors and uncertainties in the chemical process, and many factors cannot be directly measured. For example, the alkalinity in the product index is mainly composed of SiO in the raw ore. 2 CaO content and CaO content are jointly determined. Although there are mathematical formulas that can quantitatively describe the relationship between the three, the alkalinity is also affected by other sintering auxiliary materials, and the relationship between alkalinity and these auxiliary materials is very nonlinear. Characterized by mathematical formulas, it is necessary to analyze the sp...

AI Technical Summary

Problems solved by technology

Among them, there are many influencing factors and uncertainties in the chemical process, and many factors cannot be directly measured. For example, the alkalinity in the product index is mainly composed of SiO in the raw ore. 2 CaO content and CaO content are jointly determined. Although there are mathematical formulas that can quantitatively describe the relationship between the three, the alkalinity is also affected by other sintering auxiliary materials, and the relationship between alkalinity and these auxiliary materials is very nonlinear. Characterized by mathematical formulas, it is necessary to analyze the specific values ​​of alkalinity and other auxiliary materials in the primary sintering process, and establish a sinter property prediction model based on model parameter adaptation to qualitatively describe the relationship between the three

Images