Description:
TECHNOLOGY AREA(S): Air Platform, Battlespace
OBJECTIVE: Develop efficient and accurate real-time reduced order modeling (ROM) methods for four-dimensional computational fluid dynamics (CFD) predicted ship airwake data. Develop user friendly tools to process CFD datasets to create ROMs and to implement ROMs in shipboard simulation in a plug and play manner.
DESCRIPTION: NAVAIR employs computational fluid dynamics (CFD) to create high fidelity ship airwake models for integration with real-time six-degree-of-freedom aircraft simulation models for shipboard aircraft launch and recovery simulations. The unsteady, time-accurate CFD airwake models are stored on three dimensional grids encompassing the flight path of interest (a subset of the original CFD). Velocity data consisting of three orthogonal components are stored at each grid node. Because the CFD models are time-accurate, multiple sets of grid data are stored for a given time span and data frequency (typically 60 seconds at 10Hz). Datasets for just one wind-over-deck (WOD) angle can require ~6-10GB of storage. Reduced order modeling (ROM) methods are sought to reduce the data storage requirements for each WOD solution set by at least a factor of 1000 while maintaining data spatial and temporal correlation and frequency content in the 0.2 to 2.0 Hz range. ROM methods exist that are based on stochastic atmospheric models; however, while data storage requirements are significantly reduced, temporal and spatial correlation and high frequency content are not adequately preserved. Additionally, implementation of both CFD and ROM models is problematic for multiple reasons including: development of real-time search algorithms, alignment of multiple reference frames (e.g. airwake, aircraft, ship), and portability to various operating systems (e.g. Linux and Windows). This topic also seeks to develop simulation integration verification tools including: airwake alignment visualization tools, standardized verification processes and verification datasets [Refs 1-3]. Novel methods to extract data from original CFD datasets during real time simulations are also sought. All tools must be portable to Windows and Linux platforms.
PHASE I: Design, develop and determine feasibility of creating reduced order modeling methods that reduce data storage by at least three orders of magnitude while maintaining spatial and temporal data correlation and data frequency content in the 0.2Hz-20Hz range. Demonstrate feasibility of proposed method for using a government furnished information (GFI) CFD airwake dataset.
PHASE II: Develop and demonstrate a prototype of the real-time reduced order modeling tool from Phase I. Develop simulation integration methods that allow real-time implementation of ROM models derived from airwake volume grids consisting of up to two million nodes and assuming velocity data are required at approximately 100 points at a 10 Hz frame rate. Expand methods to use original CFD data source in addition to extracted subsets. Develop simulation integration verification tool box to ensure ROM models are properly aligned with multiple aircraft simulation reference systems including both numerical and visual verification. Source code and compilation documentation is required. Demonstrate integration and execution of ROM model with GFI NAVAIR Ship Airwake Analysis for Enhanced Dynamic Interface (SAFEDI) offline flight simulation tool for both rotary-wing and fixed-wing applications [Ref 1].
PHASE III: Finalize and transition the real-time reduced order modeling tool to include user selectable fidelity and scope. Refine as needed ROM fidelity to support future air platform control law development and dynamic interface modeling and simulation efforts. Develop necessary elements to integrate model(s) with manned flight simulation and to test effectiveness for pilot training. Private Sector Commercial Potential: The toolset developed under this STTR is relevant to all flight simulation applications where environment-specific air turbulence models are required including ship board aircraft launch and recovery simulations and air-to-air refueling applications. The underlying technologies can be used with any traditional 6DOF aircraft simulation including commercial flight simulators.
REFERENCES:
1. Polsky, S. A., Wilkinson, C. H., Nichols, J., et.al. (2016). Development and Application of the SAFEDI Tool for Virtual Dynamic Interface Ship Airwake Analysis. AIAA Paper 2016-1771, presented at the AIAA SciTech Conference, San Diego, CA. Retrieved from http://arc.aiaa.org/doi/abs/10.2514/6.2016-1771
2. Xin, H., & He,C. (2007). A Statistical Turbulence Model for Shipboard Rotorcraft Simulations. American Helicopter Society 63rd Annual Forum, Virginia Beach, VA. Retrieved from https://vtol.org/store/product/a-statistical-turbulence-model-for-shipboard-rotorcraft-simulations-3486.cfm
3. Roscoe, M.F., & Wilkinson, C.H. (2002). DIMSS “ JSHIPs Modeling and Simulation Process for Ship Helicopter Testing & Training AIAA-2002-4432. Presented at the AIAA Modeling and Simulation Technologies Conference and Exhibit, Monterey, CA. Retrieved from https://www.researchgate.net/publication/248743959_DIMSS_-_JSHIP's_Modeling_and_Simulation_Process_for_ShipHelicopter_Testing_and_Training-
KEYWORDS: Reduced Order Model (ROM); Airwake; Turbulence; Modeling And Simulation; CFD; 6DOF Simulation
