Prediction of Multi-Converter Gas Holders Level in Steel Enterprises based on Mechanism and Data

Le résumé fourni par la source

Linz Donawitz converter gas (LDG) is an important secondary energy source in iron and steel enterprises. The accurate prediction of converter gas holder level is of great significance for the optimal scheduling of the LDG systems. Considering that there are multi-operating conditions caused by the affection of different gas holders from each other in a multiple gas holders system, this paper proposes a mechanism and data-based converter gas holder level prediction method, in which via physical mechanism analysis the operating conditions as well as their switching logics are built according to different pressures in each gas holder and valve states. Then, the least squares support vector machine (LSSVM) is employed to establish gas holders level prediction models for each operating condition. The actual operation data in an iron and steel plant of China are employed for simulation experiments. The results show that the proposed method not only can effectively identify the operating conditions of dual gas holders system, but also can achieve higher prediction accuracies compared to those of other prediction methods, providing effective guidance for the scheduling of the LDG systems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.