Persistent Engine Condition Estimation System (PECoES)
Small Business Information
9950 Wakeman Drive, Manassas, VA, -
AbstractABSTRACT: Certain remotely piloted aircraft (RPA) use propulsion systems derived from commercial systems not designed for harsh environments. Operating experience and reliability data obtained in benign environments does not necessarily apply to engines operated in harsh environments, thereby invalidating existing propulsion health management (PHM) methods. There is a need for improved PHM systems for RPA using advanced artificial intelligence approaches being developed by Aurora Flight Sciences. RPA operations produce data that can be used to autonomously develop the needed experience base for engines used in harsh environments. Capturing, in an engine model, the extensive available data organizes it in a useful manner. Furthermore, by using a model of the RPA and engine, probabilistic predictions can be made about the engine"s future state, enabling preemptive maintenance. Aurora proposes the Persistent Engine Condition Estimation System (PECoES) to solve the estimation and data management problem presented above by applying advanced estimation algorithms and other artificial intelligence methods to the sparse data available over communications links and from operations personnel. Using these methods creates a consistent, structured, and robust framework for extracting and storing engine information. PECoES will integrate with the ground station, thereby easing certification work, and making critical data available to ground personnel. BENEFIT: Successful execution of this program will demonstrate the feasibility of the Digital Twin concept for propulsion systems and Aurora"s Persistent Engine Condition Estimation System (PECoES) concept. A Propulsion Digital Twin will enable advanced PHM capabilities for engines operating in harsh environments where operating experience is limited. PECoES, through its models, estimation algorithms, and data management technology, serves as a structured way to accumulate and store the operational and health history of an engine. With information about each engine of a certain type in the Air Force, even more advanced engine health analysis is possible through cross-comparison of the information stored in PECoES. In keeping with its strategic focus, Aurora Flight Sciences will concentrate on commercializing PECoES for use in military and commercial propulsion systems. Initially, the technology will be marketed to manufacturers of relevant current-generation RPA propulsion systems. Aurora has a working relationship with and a letter of interest from Rolls-Royce, so the AE3007H engine on the RQ-4 Global Hawk will be a prime target of our new PHM system. Other RPAs with engines that have limited PHM and onboard sensing capability could benefit from the technology. At the conclusion of Phase II, PECoES will be ready for integration into current-generation RPA systems. Aurora sees integration work on the Global Hawk system being performed during the first 2 years following the Phase II effort, followed by deployment to all Global Hawk systems in later years. Aurora will also seek to transition its technology to other current-generation and future RPA systems through established propulsion system developers and manufacturers. Aurora"s own Orion RPA, with a scheduled first flight in 2012, is another target system that Aurora will investigate. This RPA uses diesel-cycle internal combustion engines, so some modification of the PECoES system will be required, but the same approach still applies. Targeting a new system has the added benefit of closer integration into the core health management system. Finally, Aurora will investigate the applicability of PECoES to other systems that use gas turbine engines, such as commercial transport aircraft, large ships, and power generation plants. These applications would all benefit from the increased operational efficiency and health management capability provided by PECoES. Because similar engines are used, transitioning the technology to these applications would not require an entire reengineering effort; only the system integration would have to be adapted.
* information listed above is at the time of submission.