Long-term prediction method of converter gas generation in metallurgical enterprises based on steelmaking rhythm estimation

Abstract

Provided is a metallurgy industry converter gas generation amount long-term prediction method based on steelmaking rhythm estimation. Steelmaking rhythm estimation mainly comprises two parts, namely feature extraction and feature fusion. First, converter gas generation amount data are divided into data segments which are at equal time intervals with a limited number, a model-matching-based method is used for extracting time domain features and frequency domain features which can describe steelmaking rhythm from different dividing data segments, and description is carried out in the mode of a group of multi-source two-dimensional vectors; second, an improved fuzzy C mean value clustering method is used for carrying out data fusion on the multi-source two-dimensional feature vectors, and two-dimensional feature vectors which represent the steelmaking rhythm universal law are estimated; and finally, based on the estimated steelmaking rhythm feature vectors, a long-term prediction value of converter gas generation amount is reconstructed in an inverse mode. According to the method, generation amount changing in a future long time period of a converter gas system can be accurately predicted, and effective support is provided for site gas balance adjusting personnel to carry out effective gas dispatching.

Description

technical field [0001] The invention belongs to the field of information technology, relates to feature extraction based on model matching, feature description based on vector and feature fusion theory based on clustering, and is a long-term prediction method for converter gas generation in metallurgical enterprises based on steelmaking rhythm estimation. The present invention divides the existing converter gas generation data at the metallurgical enterprise site into finite segments of equal time intervals, and first uses a method based on model matching to extract time domain features and frequency domains that can describe the production rhythm from different divided data segments features, and describe them in the form of a group of multi-source two-dimensional vectors; secondly, the improved fuzzy C-means clustering method is used to fuse the multi-source two-dimensional feature vectors to estimate the two-dimensional Eigenvector; Finally, based on the estimated steelmaki...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is the long-term prediction of converter gas generation in existing metallurgical enterprises

Images