Numerical investigation of a fleet of towed AUVs

Abstract

This paper investigates the influence of the propeller on the drag of twin self-propelled AUVs, firstly, to examine the fleet performance for various propulsive conditions of leading and following AUV and, secondly, to study the parametric influence of transverse separations and longitudinal offsets on the fleet's drag. A series of CFD RANS-SST simulations have been performed at the Reynolds Number 3.2 × 10 6 with a commercial code ANSYS CFX 12.1. Mesh convergence is tested and validated with experimental and empirical results. The RANS-HO and RANS-UT propeller models are selected to estimate the time averaged thrust and torque of the propeller. The results show that the self-propelled vehicles experience an additional drag which is dominated by the thrust distribution of the propeller rather than torque. The drag of the following AUV is increased due to the upstream propeller, defined as a propeller race deduction. For the parametric studies, the results show that increasing the spacing results in a lower drag. The two sources of self-propelled drag increment are the viscous interaction and a direct result of proximity to the propeller race upstream. The result highlights the importance of considering both thrust deduction and any propeller race deductions when calculating the propulsive power consumption.

Numerical investigation of a pair of self-propelled AUVs operating in tandem

“…As the transverse separation increases up to 0.37L, the increase in accelerated flow due to the action of propeller reduces, resulting in no propeller race deduction. These results and the previous towed studies (Rattanasiri et al, 2014) suggested that the distance of 0.5L is required as the minimum transverse separation to exclude body-tobody interaction and the propeller race deduction.…”

“…The computational cost of using the medium mesh (8.9 million elements) is 10 times lower then that of the fine mesh (22.7 million elements) with a less than 2% variation in drag. Details of the fluid domain discretisation and mesh strategy's validation is completely presented in Rattanasiri et al (2014). Results utilising these meshes exhibited good correlation with the pressure distribution, the side-force coefficient, form factor and the towed drag measured in: the wind tunnel experiments (Molland and Utama, 1997), previous numerical analysis (Molland and Utama, 2002) and empirical predictions (Hoerner, 1965).…”

“…Figure 12 shows that while both hulls experience interaction, the follower B2 experiences a drag reduction while the leader B1 experiences a drag augment, due to the body-to-body interaction Rattanasiri et al (2014). The comparison of results at S /L = 0.17 and D/L = 0.57 configuration as shown in Figure 14, implied that to maintain the speed of the follower in the fleet, the propeller has to operate to pull the AUV backward.…”

“…In a previous study of towed pairs of prolate spheroids, Rattanasiri et al (2014) defined seven operating regions depending on the changes to individual drag and the total fleet, see Figure 11, where %B1 and %B2 is defined as the percentage difference of the individual drag of B1 and B2 referenced to a single towed hull drag, respectively. %CB is defined as the combined drag which considered as a percentage difference of the fleet drag referenced to the sum of two single towed hull drags.…”

Higher order hydrodynamic interaction between two slender bodies in potential flow

“…Note added in proof The analyses carried out in the previous sections neglected viscosity effects. As mentioned above a recent study [37,38], which appeared while this paper was under initial review, examined, among others, the impact of spacing between slender bodies on the viscous drag. Following [37], we define C B;j ¼ C DðB;jÞ À C DðsÞ C DðsÞ ; for j ¼ 1; 2…”