Systems Simulation for Fusion Using Novel Augmented Component Mode Synthesis Reduction Techniques

Abstract

Systems simulation provides a valuable capability for development of any engineering asset, for targeted-fidelity rapid assessment of design concepts, upfront x-in-the-loop virtual operations and extended use into physical operations as part of a live digital twin for predictive maintenance and lifetime monitoring. This is particularly true for development of commercial fusion power plants, where limited ability for physical testing and a challenging environment for diagnostics dictates a heavier reliance on such techniques.

Realising this value will in turn rely on development of novel reduced order modelling (ROM) techniques to enable efficient simulation at an appropriate fidelity. Although advancing computational capability allows for larger and more complex simulations, the environmental and financial costs must be considered, with development of efficient ROM techniques enabling more effective use of resources including enabling the high throughput computations necessary for rigorous uncertainty quantification.  

This paper presents developments of a novel full-field reduced order modelling technique using an augmented Component Mode Synthesis (CMS) reduction and modal coupling method, describing the reduction process and implementation in Modelica. The approach promises efficient simulation of coupled fluid-thermo-mechanical models of complex components within a systems environment, capturing aspects of non-linear behaviour. The approach is verified against other methods using a series of case studies, including demonstration for a coupled fluid-thermal simulation of a fusion power plant plasma facing component. Plans for further development and application for simulation of fusion systems and in wider industry are discussed in the context of moving towards realisation of a probabilistic real-time digital twin.