The goal of the project is to develop an intelligent framework to seamlessly integrate state-consistent linear parameter-varying (LPV) reduced order modeling (ROM) for aeroservoelastic (ASE) and robust aerostructural control synthesis. Leveraging on significant achievements by the proposing team in prior research, this Phase I effort represents a significant contribution to ROM technology adoption by ASE control engineers in NASA, and includes several emerging techniques: dual genetic algorithm (GA)-guided, fully automated ROMs, model order reduction with strong state consistent enforcement, and GA-optimized LPV control synthesis for enhanced vehicle stabilization. A modular software framework to streamline the entire workflow and efficiently transition from the model reduction to control synthesis will be established. The feasibility of the proposed technology will be demonstrated for ASE problems of NASA interest (e.g., X-56A MUTT, MADCAT, High-speed ASE, etc.) The Phase II effort will focus on: (1) LPV ROM and control synthesis engine optimization in terms of execution efficiency, robustness, and autonomy; (2) further process automation and exact input/output formatting for direct integration of the ‘intelligent’ environment into NASA workflow; and (3) extensive software validation and demonstration for ASE and flight control design of realistic aircrafts of current NASA interest.
The proposed technology will deliver NASA flight control engineers a valuable tool to (1) automate the entire process of ASE ROM and control synthesis on a single platform; (2) design advanced, robust aerostructural controller; and (3) perform real-time ASE simulation and analysis. It will significantly decrease simulation validation and workflow lag time, and markedly reduce development costs and cycles of aerospace vehicles. NASA projects like MUTT, MADCAT, and High-speed ASE will benefit from the technology.
The non-NASA applications are vast, and will focus on aerospace, aircraft, and watercraft engineering for fluid-structural interaction (FSI) and fatigue analysis, control and optimization, hardware-in-loop simulation, and others. The proposed development will provide a powerful tool which can be used for fault diagnostics, optimized design, simulation and experiment design and planning, and more.