Deep Learning-Based Model Reduction for Distributed Parameter Systems

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

53 Scopus Citations
View graph of relations



Original languageEnglish
Pages (from-to)1664-1674
Journal / PublicationIEEE Transactions on Systems, Man, and Cybernetics: Systems
Issue number12
Online published21 Sep 2016
Publication statusPublished - Dec 2016


This paper presents a deep learning-based model reduction method for distributed parameter systems (DPSs). The proposed method includes three phases. In phase I, numerical or experimental data of the spatiotemporal distribution is reduced into low-dimensional representations using the deep auto-encoder (DAE). In phase II, the low-dimensional representations are used to establish the reduced-order model. In phase III, the reduced model is then used to reconstruct the high-dimensional DPS. Experimental studies are conducted to validate the proposed method. The proposed method is compared with the classical proper orthogonal decomposition method and demonstrates better modeling accuracy and efficiency in the experiments.

Research Area(s)

  • Deep learning, distributed parameter system (DPS), model reduction, restricted Boltzmann machine (RBM), spatiotemporal dynamics