The unstart prediction technology is applicable to three main areas within NASA. First, the combination of high-fidelity and high-throughput engineering tools employed and validated in this project enables NASA to design and virtually test the impact of various dual-mode scramjet inlet and combustor chamber designs for hypersonic travel and access to space. Finding the most efficient method to transition between subsonic to supersonic combustion speeds, while avoiding unstart is a key requirement to propel scramjet usage into reality. Second, the image-based machine learning tools can be utilized for detecting physical phenomena across a diverse set of applications, possibly not even limited to fluid-dynamics problems, such as the detection of incipient structural failures. Lastly, the Quantification of Margins and Uncertainties (QMU) methodology can be relevant for many complex systems beyond scramjets. The methodology's focus on identifying uncertainty and rigorously assessing their impact on performance to create more accurate operability margins based on prediction can lead to a more comprehensive strategy that directly combines performance criteria and safety into the design process.
The unstart prediction technology can be used to detect the precursors to engine failure in military and commercial aircrafts and spacecrafts developed by companies such as Lockheed, Boeing, SpaceX, BlueOrigin, and Virgin Galactic. It can also be used to detect incipient structural failures in other sectors such as transportation and energy. The image-based machine learning tools can be utilized for detecting physical phenomena across a diverse set of applications, possibly not even limited to fluid-dynamics problems, such as the detection of incipient structural failures in the transportation and energy sector (i.e. failure in a power generation turbine). Lastly, the Quantification of Margins and Uncertainties (QMU) methodology can be applied to any engineering application that is required to adhere to certain safety margins relating to uncertainty. The methodology's focus on identifying uncertainty and rigorously assessing their impact on performance to create more accurate operability margins based on prediction can lead to a more comprehensive design process that optimizes for operational profitability and safety.
More »