A sequential conjugate gradient method to estimate heat flux for nonlinear inverse heat conduction problem. (15th December 2020)

• Record Type: Journal Article
• Authors: Xiong, Ping; Deng, Jian; Lu, Tao; Lu, Qi; Liu, Yu; Zhang, Yong
• Is Part Of: Annals of nuclear energy. Volume 149(2020)
• Publication Date: 2020-12-15
• Subjects: Inverse heat conduction problem -- Nonlinear -- Sequential function specification method -- Sequential conjugate gradient method; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2020.107798
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: A sequential conjugate gradient method is presented to reconstruct the heat flux. The square wave and sine wave heat flux are used to evaluate the proposed method. The influence of the number of future time step and measurement errors on the inversion results is studied. The numerical experiments verify the effectiveness and timeliness of the proposed nonlinear IHCP algorism. Abstract: In this paper, a sequential conjugate gradient method is presented to reconstruct the undetermined surface heat flux for nonlinear inverse heat conduction problem (IHCP). This modified method combines the merits of sequential function specification method (SFSM) and conjugate gradient method (CGM). The proposed method can effectively eliminate the leading error which is inevitably generated in SFSM due to the existence of temporary assumption of the unknown parameters and reduce the computing time of nonlinear IHCP. The square wave and sine wave heat flux are used to evaluate the proposed method and the influence of the number of future time steps and measurement errors on the predicted results is studied. The numerical experiments verify the effectiveness and timeliness of the proposed nonlinear IHCP algorism in reconstructing the heat flux. The comparisons with the typical SFSM and CGM are conducted to demonstrate the superiority of the proposed method. The results show that proposed method is more accurate compared with that of the SFSM under the condition of large measurement errors and the computing time is much faster than that of CGM.