This paper presents a novel finite-element (FE)-based digital twinning (DT) framework combined with reduced order modelling (ROM), which aims to augment the physical testing of heat exchanger component prototypes for fusion energy applications. There is an ongoing need for DT in the fusion energy experimental facilities, such as UKAEA’s HIVE [1], as it allows for effective autonomous or semi-autonomous control of the experiment in the extreme environment. The assimilation of sparse temperature measurements collected from a few sensors within the domain into the computational model allows for the full thermal solution reconstruction. It is conducted using previously proposed FE-based approach [2]. Furthermore, various ROM strategies are implemented and trailed to determine which one achieves best results when combined with this FE-based approach. Some of the trialled strategies include operator learning correcting the solution reconstructed on a coarse mesh to obtain a fine-mesh solution [3] and Craig-Bampton method based on thermal eigenmodes [4]. The modified framework is showcased by analysing heat conduction in a HIVE sample and controlling the maximum temperature within the domain using PID controller. References [1] R. Lewis, Simulation driven machine learning methods to optimise design of physical experiments and enhance data analysis for testing of fusion energy heat exchanger components, Phd thesis, Swansea University, 2023. [2] W. Bielajewa, M. Tindall, P. Nithiarasu, Digital twinning using novel nonlinear finite-element approach for sparse-data solution reconstruction, Sixth International Conference on Computational Methods for Thermal and Energy Problems (ThermaEComp), 6: 129–132, 2024. [3] F. José Cantarero-Rivera, R. Yang, H. Li, H. Qi, J. Chen, An artificial neural network-based machine learning approach to correct coarse-mesh-induced error in computational fluid dynamics modeling of cell culture bioreactor, Food and Bioproducts Processing, 143: 128-142, 2024. [4] A. Miyamoto, H. Sauerland, H. Xu, R. W. De Doncker, Automatic Model Order Reduction Technique for Real-time Temperature Monitoring of Oil-Cooled Electric Machines, IEEE Transactions on Industry Applications, 69(1): 477-485, 2024.