ADAPTATION-BASED LOCAL DYNAMIC COKE QUALITY PREDICTION METHOD

WIPO Patent Application WO/2024/077876

A1

Provided in the present invention is an adaptation-based local dynamic coke quality prediction method. The method comprises the following steps: acquiring a complete collection of historical coal blending solutions; converting a coal blending solution to be subjected to prediction and the historical coal blending solutions into a matrix, and performing distance measurement; according to the distance d between the coal blending solution to be subjected to prediction and each historical coal blending solution and a preset threshold value δ, extracting some historical coal blending solutions from the complete collection of historical coal blending solutions, and taking same as a coal blending solution neighborhood subS; according to the coal blending solution neighborhood subS and a preset condition, training a preset model by using the historical coal blending solutions in the complete collection of historical coal blending solutions or in the coal blending solution neighborhood subS; and inputting the coal blending solution to be subjected to prediction into a target prediction model to perform prediction. Further provided in the present invention is an adaptation-based local dynamic coke quality prediction system. The present invention is particularly suitable for coal blending prediction and scientific coal blending of coking enterprises which produce a plurality of types of coal, thereby achieving real cost reduction and efficiency improvement.

LI XINRONG (CN)
DONG WUJUN (CN)
ZHANG TAO (CN)

PCT/CN2023/083565

April 18, 2024

March 24, 2023

Click for automatic bibliography generation  

UNIDT SHANGHAI CO LTD (CN)

G06Q10/06

UNITALEN ATTORNEYS AT LAW CO., LTD. (CN)

View/Download PDF      PDF Help

Previous Patent: SELECTIVE ETCHING SOLUTION FOR 3D NAND STRUCTURAL SHEET

Next Patent: FOLDABLE TENT FRAME HAVING BED BOARDS