Keywords: high-speed trains, surface pressure field, compressive sensing, proper orthogonal decomposition, particle swarm optimisation, sparse sensor placement, flow field reconstruction.

This paper presents a rapid prediction framework for the real-time reconstruction of surface pressure fields on complex three-dimensional high-speed trains using extremely sparse sensor data. The methodology integrates generalised proper orthogonal decomposition, compressive sensing, and an improved particle swarm optimisation algorithm. By extracting the dominant spatial modes of the flow field, the number of significant coefficients required for reconstruction is reduced to fewer than ten, fundamentally minimising the required physical sensors. The optimisation algorithm successfully configures a minimal sensor array of seven pressure measurement points on the leading car. The framework accurately reconstructs the surface pressure field, achieving an average computation time of 0.0033 seconds per case and an average relative reconstruction error of three per cent. Evaluations across varied operating conditions confirm high fidelity and aerodynamic rationality. By bypassing the need to resolve the entire volumetric flow field, the proposed approach delivers a transformative reduction in computational effort compared to traditional computational fluid dynamics, providing a robust foundation for real-time aerodynamic load calculation and operational safety monitoring.

download the full-text of this paper (PDF, 16 pages, 926 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description