A Blast Furnace Molten Iron Quality Monitoring Method Based on Adaptive Threshold pls

Abstract

The present invention provides a blast-furnace molten iron quality monitoring method based on adaptive threshold PLS. The method comprises: collecting blast furnace operation parameters and molten iron quality variables at the same moment; selecting data in the blast-furnace molten iron normal process as a training set, calculating a mean value and a standard deviation and performing standardization processing; constructing a PLS model; obtaining the new blast furnace operation parameter sample data of the blast-furnace ironmaking process and performing standardization processing; aiming at a test set, employing a Q statistical magnitude and an Hotelling's T<2> statistical magnitude to detect whether the blast-furnace ironmaking process generates anomaly or not, calculating the Q statistical magnitudes and the T<2> statistical magnitudes of the test set samples, and calculating a fixed control limit; and calculating the index weight mobile mean value of each sample statistical magnitude at present in real time so as to determine the T<2> statistical magnitude adaptive threshold and the Q statistical magnitude adaptive threshold at the current moment and complete the fault detection of the test set. The method provided by the invention obviously reduces the fault false alarm rate and ensures the accuracy and the sensitivity of a fault detection result.

Description

technical field [0001] The invention belongs to the technical field of blast furnace molten iron quality monitoring, in particular to a blast furnace molten iron quality monitoring method based on an adaptive threshold PLS. Background technique [0002] Blast furnace ironmaking is the main method of modern ironmaking. The iron produced by this method accounts for more than 95% of the world's total iron. The steel produced is widely used in machinery manufacturing, transportation, medical equipment and military development. industry. Blast furnace ironmaking is an important link in steel production, and the quality of molten iron produced directly determines the quality of subsequent converter steelmaking. At present, comprehensive molten iron quality indicators are usually measured by silicon content ([Si]), phosphorus content ([P]), sulfur content ([S]) and molten iron temperature (MIT). However, the forward motion of the blast furnace (normal working condition) is the re...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the above problems of high false alarm rate and unsatisfactory detection effect in the blast furnace molten iron quality monitoring based on the PLS (partial least squares) method, the present invention is based on the idea of ​​exponential moving weighted average, and the fixed control The limit is improved to the method of applying adaptive threshold, which is applied to the quality monitoring method of blast furnace molten iron based on partial least squares, and improves the monitoring effect of blast furnace molten iron quality

Images