Temperature prediction of roller kiln based on mechanism and SSA-ELM data-driven integrated model

Abstract

Temperature prediction in roller kiln is an important problem for lithium-ion battery cathode materials. However, since the complexity of process and industrial noise, it is difficult to obtain accurate prediction models. This paper presents an integrated modeling method that combines mechanism with data-driven model. Firstly, on the basis of heat transfer mechanism and heat balance theory, the mechanism model of the sagger temperature in the roller kiln is established. Then, aiming at the estimation error caused by industrial process complexity and industrial noise, a data-driven error compensation model based on real-time process operation data is constructed. In view of the large lag of temperature change in roller kiln and the high correlation and strong nonlinearity between process variables, the error compensation model of sparrow search algorithm optimization extreme learning machine SSA-ELM is established. Then the mechanism model and error compensation model are combined to obtain a comprehensive temperature prediction model. Finally, the numerical experiments using real data show that the comprehensive prediction model has high prediction accuracy.