Fuzzy estimation for heat flux distribution at the slab continuous casting mold surface

A fuzzy inference method for estimating the heat flux distribution at the metal-mold interface in slab continuous casting is established in this paper. The finite element method is applied to solve the direct heat conduction problem. A set of decentralized fuzzy inference units are established, and the deviations between the calculated and measured temperatures acquired with the thermocouples buried in the mold are taken as the input parameters of fuzzy inference units, and the corresponding inference components are obtained by fuzzy inference. Then the inference components are weighted and synthesized by a weighted integrated approach based on normal distribution function to gain the compensations of the guessed heat flux distribution. Finally, a decentralized fuzzy inference (NDFI) method based on normal distribution weighted approach is formed. Some numerical experiments are performed to study the effects of the number of temperature sensors, different initial guesses of heat flux distribution and measurement errors on the reversion results. Comparisons with both the existing decentralized fuzzy inference (DFI) method and the conjugate gradient method (CGM) are also conducted, and they all show the validity of the inversion method established in this paper. (C) 2014 Elsevier Masson SAS. All rights reserved.