Online prediction system for terminal composition of low-carbon ferrochromium in silicothermic smelting

Abstract

The invention discloses an online prediction system for terminal composition of low-carbon ferrochromium in silicothermic smelting, which relates to the technical field of ore smelting. The system comprises an information acquisition device and a computer processing system; the information acquisition device communicates with the computer processing system through a network and is used for acquiring the data required for online prediction of the terminal composition of the low-carbon ferrochromium; the computer processing system constructs a work condition-based online prediction model for the terminal composition of the low-carbon ferrochromium in the silicothermic smelting for the acquired data by a multi-scale SVM (support vector machine) information merging method, thereby implementing online prediction for the terminal composition of the low-carbon ferrochromium in the silicothermic smelting, and making a webpage of the prediction result for real-time online query. The invention has the beneficial effects that by utilizing the method provided by the invention, the prediction accuracy is effectively enhanced, and the prediction result is more accurate, more reasonable and more practical.

Description

Technical field [0001] The invention relates to the technical field of submerged arc furnace smelting, in particular to an on-line prediction system of low-carbon ferrochromium end point composition in electric silicon thermal smelting. Background technique [0002] The electro-silicothermal method refers to a method in which silicon in a silicon-chromium alloy is used to reduce the oxides of chromium and iron in chromium ore to produce low-carbon ferrochrome under the condition of making alkaline slag in a submerged arc furnace. The end-point composition of low-carbon ferrochromium in electrosilicothermal smelting includes temperature, carbon content and silicon content. The online prediction of these three variables is not only related to the judgment of the end point of smelting, but also to determine the iron tapping time. It also plays a key guiding role in the power supply system, production quality, energy saving and consumption reduction of the entire smelting process. [...

AI Technical Summary

Problems solved by technology

[0006] The present invention solves the technical problems in the prior art that the training time of data samples with a large amount of data is too long, and that there are errors when the whole smelting process is predicted by a single model, and provides an end point of low-carbon ferrochrome smelted by electric silicon thermal method Composition online prediction system, which makes full use of various sensor information in different smelting conditions, and uses multi-scale support vector machine for fusion analysis, and the prediction results have a high accuracy rate, which can meet actual needs, and the friendly interface design is more Simple and practical

Images