Sintering mixed water adding control method based on dual-model collaborative prediction

Abstract

The invention relates to the technical field of sintering mixed water adding control, and provides a sintering mixed water adding control method based on dual-model collaborative prediction. The method comprises: firstly, collecting historical data of the sintering, mixing and water adding process to form a historical data set; secondly, preprocessing the historical data set; constructing and training a water adding amount regression prediction model based on the convolutional neural network by taking the blanking amount of each raw material as input and the water adding amount as output; constructing and training a moisture content classification prediction model based on a convolutional neural network by taking the blanking amount and the water adding amount of each raw material as inputand the moisture content category corresponding to the moisture content of the mixture as output; and finally, controlling the water adding amount based on dual-model collaborative prediction: predicting the water adding amount and the moisture content category in real time, and adjusting the next water adding amount according to the predicted value of the moisture content category. According tothe invention, the accuracy of water addition prediction and control can be improved, the moisture content of the mixture is stabilized in the optimal range, and the prediction control efficiency is high.

Description

technical field [0001] The invention relates to the technical field of sintering mixing water addition control, in particular to a sintering mixing water addition control method based on dual-model cooperative prediction. Background technique [0002] The sintering process is an important process in steel production, which has a great influence on the quality of steel production. The ultimate goal of the sintering process is to provide sintered ore that meets the process requirements for blast furnace production. At present, most sintering plants use the sintering mixing process to mix iron-containing ore powder, various fluxes, fuels and other powdery and granular materials with water, and then ignite and burn them on the sintering machine, and a series of physical and chemical reactions form a block with good air permeability. shaped products. [0003] The amount of water added in the mixing process is a key control parameter in the sintering mixing process. Generally, t...

AI Technical Summary

Problems solved by technology

However, the existing feedforward forecasting method for controlling the amount of water added, on the one hand, the accuracy of the feedforward forecast is not high, and on the other hand, the feedforward forecast has not been corrected, resulting in the use of the obtained forecasted value of water addition to control the amount of water added. Moisture content cannot meet the process requirements, and the efficiency of predictive control is low

Images