Modal analysis of non-ducted and ducted propeller wake under axis flow

Abstract

Modal decomposition techniques, flow field and spectral analysis are employed to investigate the wake dynamics and destabilisation mechanisms of a four-bladed marine propeller with or without a nozzle. Numerical simulations are conducted using the Delayed Detached Eddy Simulation (DDES) model for the wake and the Arbitrary Mesh Interface (AMI) method for the blade rotation. The presence of the nozzle significantly reduces the wake's streamwise velocity, delays the wake destabilisation, increases the wake length, and changes the morphologies of wake vortices. In particular, the hub vortex in the ducted propeller wake is broken down into chaotic turbulence by the perturbation of the backflow. From modal analysis, the spatial scale of flow phenomena decreases with the increase of modal frequency. Underlying destabilisation mechanisms in the wake correspond to some characteristic frequencies. The interaction of each sheet vortex with the previously shed tip (leakage) vortices occurs at blade passing frequency (BPF). The pairing of adjacent tip (leakage) vortices occurs at half-BPF. The long-wave instability of the hub vortex and the wake meandering are stochastic processes, each of which occurs at a frequency lower or equal to shaft frequency (SF). These four destabilisation mechanisms can approximately reconstruct the large-scale flow phenomena in the wake. Moreover, each sheet vortex's alternating connection and disconnection with the previously shed tip (leakage) vortices cause the short-wave instability of the tip (leakage) vortices and generate the secondary vortices. The radial expansion motion of large-scale helical vortices in the outer slipstream dominates the wake meandering phenomenon.

Effects of a nozzle on the propeller wake in an oblique flow using modal analysis

“…2018), as further confirmed by Shi et al. (2022), Foti et al. (2018 a ) and Bastankhah & Porté-Agel (2017) using reduced-order reconstruction.…”

“…2021) and its variants, including detached DES (DDES) (Zhang & Jaiman 2019; Shi et al. 2022) and improved DDES (IDDES) (Qin et al. 2021; Wang et al.…”

“…2011) or numerical probes (Shi et al. 2022) are combined with time-frequency transform (or analysis) methods to explore the characteristic frequencies of flow phenomena and to investigate the energy transfer process in the propeller wake. The spectral analysis of individual spatial points can only qualitatively associate the flow signals with visible flow phenomena (Felli et al.…”

“…2021 a ; Shi et al. 2022) but cannot visualize the flow phenomena with specific frequencies. An alternative way to address this problem is to decompose the whole flow field and visualize the spatial modes using machine learning-based, reduced-order models (ROMs, Rowley & Dawson 2017; Taira et al.…”

“…(2018), Shi et al. (2022) and Wang et al. (2022) identified that the passage of helixes and the pairing of adjacent helixes are characterized by blade frequency and half blade frequency, respectively.…”