A blast furnace fault diagnosis system and method

Abstract

The invention provides a blast furnace fault diagnosis system and method. The system includes: a historical data collection module, an actual data collection module, a feature weight matrix construction module, a model building module, and a blast furnace fault diagnosis module. The method includes: collecting actual attribute data of blast furnace production conditions, historical attribute data and corresponding blast furnace operation fault state types; determining the characteristic weight of each attribute according to the importance of each attribute for fault diagnosis, and constructing a characteristic weight matrix; establishing a method for blast furnace Feature-weighted twin hypersphere support vector machine model for fault diagnosis; the actual attribute data of the blast furnace production status is brought into the established feature-weighted twin hypersphere support vector machine model to obtain the type of blast furnace operating fault state to which the actual attribute data of the blast furnace production status belongs. Complete blast furnace fault diagnosis. The present invention quantifies the importance of each feature of blast furnace faults and integrates the importance of each feature into the construction process of the learning machine to improve the accuracy of fault diagnosis.

Description

Technical field [0001] The invention belongs to the technical field of blast furnace fault diagnosis, and specifically relates to a blast furnace fault diagnosis system and method. Background technique [0002] The fault diagnosis of blast furnace is of great significance to reduce accidents and economic losses. Blast furnace smelting is the reduction of pig iron ore into iron. It is a continuous and complex production process; in order to prevent abnormal situations, a large number of parameters in the production process need to be monitored, such as hot air temperature, hot air volume, hot air pressure, furnace Top pressure, total pressure difference, upper pressure, lower pressure, oxygen enrichment, permeability index, cross temperature measurement, material speed, physical heat, [Si] content, etc.; therefore, the characteristics of the fault state are high-dimensional and multi-featured signals Therefore, the traditional fault diagnosis method applied to blast furnace fault...

AI Technical Summary

Problems solved by technology

Blast furnace smelting is to reduce pig iron ore to iron, which is a continuous and complicated production process; in order to prevent abnormal situations, a large number of parameters in the production process need to be monitored, for example, hot air temperature, hot air volume, hot air pressure, furnace Top pressure, total pressure difference, upper pressure, lower pressure, oxygen enrichment, air permeability index, cross temperature measurement, material speed, physical heat, [Si] content, etc.; therefore, the characteristics of the fault state are high-dimensional and multi-featured signals comprehensive embodiment; therefore, the traditional fault diagnosis method applied to the blast furnace fault diagnosis effect is not very ideal

Images