Machine Learning Software Tool for Rocket Engine Anomaly Detection and Health Monitoring

Project Objective  

This should be your one-sentence project description.  

Project Description 

Include long-term goals, processes, resources being invested, and technologies under development. Be sure that the language is clear, concise, and understandable to the public audience. Be sure to spell out any acronyms.  

previous description published (edit as needed): Artificial Intelligence and Machine Learning (AIML); while a buzzword topic of conversation and a rapidly growing field; has not been explored for the applications of liquid rocket engine design and analysis. Adding AIML to existing test operations and data review procedures offers a supplementary tool for detecting anomalies in engine performance and understanding higher-order relationships between engine control parameters. This addition would lead to weeks of saved workhours per engine firing; improved engine and control design; and avoiding engine and launch failures. AI fault detection benefits future engine programs by quick evaluation of engine performance. This capability will improve designs through rapid fault identification and offer autonomous anomaly detection for earth independent operations (EIO). ML is equivalent to a full engine team reviewing data live in-flight and flagging potential faults for astronauts on deep-space missions. Using the extensive test history of the L3 Harris RS-25 (formerly Aerojet Rocketdyne Space Shuttle Main Engine[SSME]); we will develop machine learning methodologies for rocket engine state prediction and anomaly detection. This proposal asks for 0.2 FTE over 6 months to provide proof-of-concept for autonomous anomaly detection for unlabeled RS-25 engine datasets. Once at that level of confidence; the model can then be tested on other fledgling engine programs. A stretch goal of $15;200 ODC is also requested to bring in an intern with fresh software development and machine learning education and experience to help develop and test this tool. Finally; we request an additional stretch goal of $10;000 to stand-up a machine learning training server to be made available to the center to develop critically needed center experience in this technology.

Project Results and Conclusions 

Provide one to three paragraphs listing specific accomplishments, technology development milestones, data and results, or constructed hardware for the calendar year. Include lessons learned if applicable. 

This is early/mid-stage (TRL 2) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.