Designing a Stealthier Submarine with Design-driven Simulation Tools

Submarine engineering is a challenge – there are few areas of engineering that require such precise control of as many different domains of physics. A successful, safe and reliable underwater vessel design must be able to withstand the pressures of deep ocean water and the stresses of waves – and the shock impacts of combat. It must travel with minimal drag as silently as possible while also minimizing its radar cross section. Its hydraulic, electrical, thermal, ventilation and mechanical systems are safety-critical and all need to operate reliably for months on end.

This means submarine development requires large teams with many different specialist groups working together to produce a fully capable vessel that can carry out undersea defense objectives.
This blog post will demonstrate how MODSIM, unified modeling and simulation, enables the development of modern undersea defense technology, using stealth design as an example of the benefits of a MODSIM approach.

The detectability of a submarine depends on several factors. The noise signature of a submarine affects how readily the vessel is detected by passive sonar. The sound is not just a result of vibrations and noise from machinery inside the vessel – the hydrodynamical design of the hull itself affects noise as water flows turbulently, particularly around control surfaces. A poorly optimized design produces excess noise and increases the vessel’s detectability.

Another important factor is the radar cross section (RCS). When radar waves strike the vessel, they are reflected and scattered. The waves that are back-scattered towards the receiver are what allows an enemy to detect the submarine. Therefore, one of the basic principles of stealth design is to minimize the back-scatter. For undersea vessels near the surface, this is further complicated by hydrodynamics as the wake produces a complex free surface deformation pattern.

To maximize stealth, we need to be able to deal with both these effects at the same time. This means naval architects, radar engineers and computation fluid dynamics (CFD) experts all need to work together in a multi-domain, multi-objective workflow.

Modeling the submarine

The workflow to reduce detectability begins early in development. The sooner stealth factors are considered, the less risk there is of problems being found late in testing, and the more potential there is to find an optimal solution. As soon as the initial CAD model of the hull is available, work can begin.

The project begins by making full use of the “Requirements, Functions, Logical and Physical” (RFLP) data model to build the CAD model shown in Figure 1. The model geometry was built with CATIA roles, with parameters allowing optimization later, and therefore it can be made immediately available for simulation on the 3DEXPERIENCE platform. The CAD model is more detailed than is needed for the simulation, so automatic de-featuring increases simulation performance.

The submarine system’s dynamic behavior is modeled and simulated using the Dymola Behavior Modeling (DBM) application, which is an integrated 1D-modeling and simulation environment on the 3DEXPERIENCE platform. DBM directly generates multibody system models from CAD data and provides 3D visualization and animation capabilities of the system behavior. This is used to generate a multibody model of the submarine including control surfaces. With this, users can study the hydrodynamic behavior of the submarine under varying operational conditions or drive strategies. The control systems of ballast tanks can be accurately modeled, in order to realistically represent surfacing and diving behavior. Different drive units – for instance, diesel engines or fuel cells – can also be considered in the simulation. Above described process is fully summarized in Figure 2.

The results of the control system simulation are compared with the test requirements (see Figure 3). If they are not met, the model parameters and topology can be modified through an iterative process until the target behavior is achieved.

Once we have the geometry and control surfaces optimized, we can calculate the hydrodynamic wake of the submarine. We use the SIMULIA Fluid Dynamics Engineer, which takes advantage of a state-of-the-art unstructured second order fully implicit finite-volume Reynolds-Averaged Navier-Stokes (RANS) solver to model the flow, including turbulence. The interface between water and air, where the surface wake forms, is solved using the multiphase Volume of Fluid (VOF) formulation. The free surface VOF iso-contour plots are the data we need for the next stage of the simulation. Figure 4 shows an example from the semi-submerged submarine.

Any flow simulation requires careful meshing, as this affects both convergence behavior – which determines the speed of the simulation – and the accuracy of the results. Under MODSIM, meshing is a largely automated process.

For this simulation, we used a shooting-and-bounding ray (SBR) model with the Asymptotic Solver in CST Studio Suite. The SBR method has been demonstrated to be feasible and accurate for periodic and random rough surfaces, with agreement between its results and those of the method of moments. The CST Studio Suite SBR method includes shadowing and multi-path effects for greater accuracy, making it an effective electromagnetic tool for RCS evaluation and for the generation of radar signals in maritime environments.

The free-surface VOF iso-contour plots are directly translated into a tessellated mesh from the FMK app into CST Studio Suite through the 3DEXPERIENCE platform. This provides the single source of truth, collecting all the different modeling and simulation data, as well as the RFLP, in a single place accessible to all team members. Everyone has access to the latest versions of all files, reducing the risk of confusion and delays caused by using different versions of data or incompatible file formats. Traceability is preserved across different design iterations. These processes result in the overall radar signature workflow description as highlighted in Figure 5.

To calculate the RCS, a monostatic scattering simulation is carried out along the azimuth angle from 0 to 360 degrees, with a 1 degree step width at a fixed elevation angle of 89 degrees (almost sea level). In other words, the radar scattering is calculated at each azimuth angle around the sweep, and plotted in 1D around the target.

In order to extract the impact of the shape of the water surface, the simulation is repeated for a flat plane. The RCS in each simulation is compared to this one as a control. Figures 6 and 7 shows how the presence of the submarine and the waves affect the RCS – a full 3D far-field has been plotted here for illustrative purposes.

The results show that the submarine has a significant contribution to the overall RCS.  The sub on a planar surface and a realistic surface has a very similar RCS, with the realistic surface dominating the planar surface at some angles. These angles are of interest, as they tell us the directions at which the excitation rays reflect off the surface of a wake, thereby compromising the stealth of the submarine.

Conventional submarines use periscopes and snorkeling masts as vertical appendages to approach and breach the free surface. This means that these are a major source of hydrodynamic disturbance and turbulence. This results in an unsteady flow and a high, steep bow wave at snorkeling speeds below 5 m/s. This wave can be identified by radar or visual observation. To address this issue, baffles are added to the periscope – in this case, a helical strake running up the periscope disrupts the development of the bow wave (Figure 7). The strake model is parameterized so the thickness and pitch of the helix can be varied for a component optimization study.

The addition of helical strakes, especially at 150 mm, significantly reduced the overall RCS (Figure 9). Because a MODSIM methodology has been used, the optimized periscope design from the simulation can be easily returned to the model and integrated into the final submarine design.

MODSIM, unified modeling and simulation, enables engineers to develop cutting-edge designs faster and more efficiently. For an undersea defense vessel, MODSIM enabled a full RCS analysis for stealth purposes, covering multiple physical disciplines. The unification of CAD and CAE meant that the submarine design data could be easily converted into a simulation-ready model, and the integration of different simulation tools on the 3DEXPERIENCE platform allows systems, electromagnetic and fluid simulations to be combined into a single workflow to accurately capture the complex interaction of radar scattering and hydrodynamic wakes. Using MODSIM, the engineers were able to find an optimal periscope geometry quickly, reduce RCS and improve stealth, and implement this in the final design.

The defense industry is moving towards model-based acquisition. This requires a digital transformation in the acquisition process, with new and collaborative ways of working between defense agencies and their suppliers. Find out more on designing and engineering next generation naval defense platforms with the 3DEXPERIENCE Platform.