Establishment method for sintering end point prediction system based on big data and machine learning

Abstract

The invention discloses an establishment method for a sintering end point prediction system based on big data and machine learning, and belongs to the field of sintering technological process control.The model establishment method comprises the steps that historical data are collected; data pre-processing is carried out; the model input variable and output variable are determined; an output sample is subject to visual presentation according to a drawing method, the data distribution characteristic and the equipment control accuracy are comprehensively considered, and sintering end point position classification is determined according to technological experience; and a GBDT is used for building a prediction model for sintering end point trend change states. The model established through the method can obtain the trend change range, it can be obtained that current ignition sintering mixture sintering end point positions are six classification trend change conditions including undersintering+++, undersintering++, undersintering+, undersintering, normality and oversintering. Well guidance is provided for a sintering process operator to judge sintering end point in advance and taking measures including fine adjustment, early adjustment and the like.

Description

technical field [0001] The invention relates to the establishment method and application of a sintering endpoint prediction system based on big data and machine learning, and belongs to the field of sintering process control. Background technique [0002] In sintering production, the sintering end point is an important production operation index, and the stability of the sintering end point is the key to controlling the output, quality and cost reduction of sintering ore. The sintering end point is advanced, and the production capacity of the sintering machine cannot be fully utilized, resulting in a reduction in the output of sintering ore; the sintering end point is lagging behind, and the mixture is unloaded at the discharge end of the machine tail before it is completely burned, resulting in a decline in the pass rate and a decrease in the output of the sintering machine. And quality goes down, cost goes up. In addition, if the unburnt sinter enters the blast furnace, i...

AI Technical Summary

Problems solved by technology

However, in actual sintering production, there are many influencing factors, and it is difficult to calculate the vertical sintering speed, so it is difficult to obtain accurate sintering end point information by applying this method on site

[0008] To sum up, the existing methods for predicting the end point of sintering either seriously lag behind or cannot be predicted accurately in practice, and there are few models for predicting the end state of sintering

Images