Method for predicting temperature of hot continuous rolling finish rolling inlet based on data drive

Abstract

The invention belongs to the field of steel and iron rolling, and relates to a method for predicting the temperature of a hot continuous rolling finish rolling inlet based on data drive through an improved gray wolf optimization algorithm optimized error back propagation neutral network (AGWO-BP). A large quantity of data are acquired on site, a finish rolling inlet temperature prediction data setbased on full length of a plate blank is established, a finish rolling inlet temperature prediction model is established through a neutral network algorithm, a gray wolf population is initialized through a chaos Tent sequence, an AGWO-BP prediction model is obtained, and the finish rolling inlet temperature of surface full length of next steel is predicted through relevant data of the rolled plate blank. According to the method, the temperature of the hot continuous rolling finish rolling inlet of surface full length of the plate blank can be predicted at high precision, and the method is applied to prediction of the temperature of the hot continuous rolling finish rolling inlet.

Description

technical field [0001] The invention belongs to the field of iron and steel rolling, and more specifically relates to a method for predicting the entrance temperature of hot continuous rolling and finishing rolling based on data driving by using an improved gray wolf optimization algorithm to optimize error backpropagation neural network. Background technique [0002] The complex industrial production process of steel rolling is a complex industrial large system composed of multi-subsystem processes. One of the characteristics of modern iron and steel enterprise production is multi-channel processing procedures. In the actual production process, the relevant input variables and interference variables have as many as dozens of dimensions, and the input variables are added in stages, and each step has an impact on the final product. Quality makes a difference. Among the many process parameters of hot rolling, the prediction accuracy of temperature is the key to ensure the acc...

AI Technical Summary

Problems solved by technology

In the prior art, it is not easy to measure the temperature at the entrance of finishing rolling accurately, because if the measuring point is placed before and after the flying shear, the slab will be transported for more than 100 meters at this time, and the secondary oxide scale will be formed on the surface, which is not conducive to temperature measurement

If the measuring point is placed behind the finishing descaling box, the surface temperature is too low due to high-pressure water cooling, and the temperature difference between the inside and outside of the slab is too large

[0004] At present, there are many domestic papers and patents on the prediction of the exit temperature and coiling temperature of hot continuous rolling, but there are relatively few patents and documents related to the prediction of the entrance temperature of the finish rolling. Including: "A method for predicting the entrance temperature of hot continuous rolling finish rolling" (Patent No. CN 105537284 B) proposes a method of predicting the temperature of the finish rolling entrance slab using traditional mathematical models. This patent is the only one in the prior art for finish rolling The entrance temperature is predicted, but the technical problem in this patent is to use traditional mathematics to build a finish rolling entrance temperature prediction model. The construction process is cumbersome, with a large amount of calculation and poor generalization ability. , which is simpler and has a strong generalization ability compared with

In the paper "The Temperature Prediction Model of Hot Continuous Rolling Finishing Based on the CMAC Neural Network", the production data is used to train the CMAC network to analyze and simulate the actual temperature forecast of the finishing rolling. The technical problem in this paper is that the temperature change of the entire length of the slab surface is not considered. And the number of samples is too small

Moreover, the temperature prediction at the entrance of finish rolling in the prior art cannot effectively eliminate various interferences received during the measurement of the pyrometer at the entrance of finish rolling, and thus cannot obtain a more realistic predicted temperature, and the prior art does not use AGWO -Patents and literature on temperature prediction by BP method

Images