Improved signature prediction through coupling of ShipIR and CFD

Most existing platform signature models use semi-empirical correlations to predict flow convection on internal and external surfaces, a key element in the prediction of accurate skin signature. Although these convection algorithms are capable of predicting bulk heat transfer coefficients between each surface and the designated flow region, they are not capable of capturing local effects such as flow stagnation, flow separation, and flow history. Most computational fluid dynamics (CFD) codes lack the ability to predict changes in background solar and thermal irradiation with the environment and sun location, nor do they include multi-bounce radiative surface exchanges by default in their solvers. Existing interfaces between CFD and signature prediction typically involve a one-directional mapping of CFD predicted temperatures to the signature model. This paper describes a new functional interface between the NATO-standard ship signature model (ShipIR) and the ANSYS Fluent model, where a bi-directional mapping is used to transfer the thermal radiation predictions from ShipIR to Fluent, and after re-iteration of the CFD solution, transfer the wall and fluid temperatures back to ShipIR for further refinement of local-area heat transfer coefficients, and re-iteration of the ShipIR thermal solution. Both models converge to an RMS difference of 0.3 °C within a few successive iterations (5-6). This new functional interface is described through a detailed thermal/IR simulation of an unclassified research vessel, the Canadian Forces Auxiliary Vessel (CFAV) Quest. Future efforts to validate this new modelling approach using shipboard measurements are also discussed.