Proper Orthogonal Decomposition and Extended- Proper Orthogonal Decomposition Analysis of Pressure Fluctuations and Vortex Structures Inside a Steam Turbine Control Valve

Abstract

In steam turbine control valves, pressure fluctuations coupled with vortex structures in highly unsteady three-dimensional flows are essential contributors to the aerodynamic forces on the valve components, and are major sources of flow-induced vibrations and acoustic emissions. Advanced turbulence models can capture the detailed flow information of the control valve; however, it is challenging to identify the primary flow structures, due to the massive flow database. In this study, state-of-the-art data-driven analyses, namely, proper orthogonal decomposition (POD) and extended-POD, were used to extract the energetic pressure fluctuations and dominant vortex structures of the control valve. To this end, the typical annular attachment flow inside a steam turbine control valve was investigated by carrying out a detached eddy simulation (DES). Thereafter, the energetic pressure fluctuation modes were determined by conducting POD analysis on the pressure field of the valve. The vortex structures contributing to the energetic pressure fluctuation modes were determined by conducting extended-POD analysis on the pressure–velocity coupling field. Finally, the dominant vortex structures were revealed conducting a direct POD analysis of the velocity field. The results revealed that the flow instabilities inside the control valve were mainly induced by oscillations of the annular wall-attached jet and the derivative flow separations and reattachments. Moreover, the POD analysis of the pressure field revealed that most of the pressure fluctuation intensity comprised the axial, antisymmetric, and asymmetric pressure modes. By conducting extended-POD analysis, the incorporation of the vortex structures with the energetic pressure modes was observed to coincide with the synchronous, alternating, and single-sided oscillation behaviors of the annular attachment flow. However, based on the POD analysis of the unsteady velocity fields, the vortex structures, buried in the dominant modes at St = 0.017, were found to result from the alternating oscillation behaviors of the annular attachment flow.

Unsteady behavior of wall-detached flow inside a steam turbine control valve

“…As a result, massive reverse flows are produced in the core region of the circular pipe. The unsteady flow behaviors of such wall-attached flow, i.e., the in-phase and out-phase oscillations of the annular wall-attached jet and the intermediate shear layer instabilities, were reported in detail by Wang et al 22 However, the wall-detached flow pattern generally appears under more extreme flow conditions with larger opening ratios or lower pressure ratios and thus may generate more severe aerodynamic problems in the valve system. In such a flow pattern, the supersonic jet would detach from the surrounding wall surfaces and directly impinge into the core region of the downstream pipe.…”

“…20,21 However, comprehensively understanding the wall-detached flow is still insufficient, especially the dominant vortex structures and energetic pressure fluctuations, which contribute mostly to the highly unstable three-dimensional (3D) flows inside steam turbine control valves. 22,23 Continuing a previous study by Wang et al, 22 the present study focuses primarily on the unsteady behavior of wall-detached flow inside a bell-shaped steam turbine control valve. To this end, the dependency of wall-detached flow inside an ultra-supercritical steam turbine 24 control valve on the valve's opening ratio and pressure ratio was established first.…”

“…scitation.org/journal/phf influence of dominant POD modes on the unsteady behaviors of the flow dynamics can be further investigated. The procedures for performing POD and flow reconstruction to analyze complex valve flows can be found in the work of Wang et al 21,22 POD analysis on the fluctuating velocity field and pressure field was conducted, and dominant velocity fluctuation modes and pressure fluctuation modes were successfully extracted. Figure 11 plots the normalized eigenvalues of the velocity POD and pressure POD modes.…”

Data Assimilation of Steam Flow Through a Control Valve Using Ensemble Kalman Filter

“…Indeed, advanced computational fluid dynamics (CFD) methods such as large eddy simulation (LES) and detached eddy simulations (DES) can be applied to steam flow simulations in control valves, achieving higher accuracy in the prediction of valves' flow patterns. Wang et al [1][2][3] conducted several studies on the unsteady flow behaviors in steam control valves using DES and proper orthogonal decomposition (POD) analysis. Results showed that the pressure fluctuation intensity comprised the axial, antisymmetric, and asymmetric pressure modes.…”