Our Research & Development vision is to discover, incubate, and accelerate innovative and validated research into viable technology to solve the most challenging data problems and address real needs.
How We Are Different
Our R&D work always has application and implementation as a goal, but we enjoy working on challenges on the frontier of knowledge, discovering new phenomena. Our experience includes:
- NASA’s first ever Sequential Phase II STTR awardee
- $2.5 million in SBIR and STTR awards since 2011 ($1.8MM since 2014)
- Network of trusted industry and research institution partnerships
- Successful technology transfer through commercialized software and peer-reviewed publication
- Technical staff with skills honed through 25 years of practical experience in applied AI/ML/DS consulting
- Biomarker Identification and Detection – ML-based approaches to the detection of pathogens and degenerative neurological conditions like Alzheimer’s
- Corrosion Fatigue – Modular, physics-based frameworks for predicting gas-turbine engine hot corrosion and fatigue life of critical components
- Geospatial – Novel applications of Deep Learning techniques and ML to complex data fusion and analysis of geospatial datasets
- Machine Reliability – Innovative applications of ML to complex fault diagnosis and detection problems in IIoT systems for critical infrastructure management
- Text Mining – Pioneering applications of NLP, Deep Learning, and ML technology to anomaly detection using unstructured text information
- Air Force
Hot Corrosion of Gas-Turbine Engine Components
Problem – Increasing operating temperatures in gas-turbine engines in the pursuit of greater efficiency has led to higher rates of hot corrosion of critical Ni-based superalloy components. However, the design and environmental factors that lead to hot corrosion was not well-understood. Frequently, engines would be in for minor service only to have hot corrosion unexpectedly discovered. This seriously degrades readiness of air platforms and increases operation and maintenance costs.
Solution – A team composed of researchers from Elder Research, Southwest Research Institute, the University of Virginia, and Rolls-Royce created HOTPITS: a modular, physics-based framework for the prediction and simulation of hot corrosion of Ni-based superalloy components. The HOTPITS framework predicts:
- The “active state” required for hot corrosion based on engine design parameters and environmental factors
- The rate of pit incubation on Ni-based superalloy parts
- The rate of pit growth
- The transition of corrosion pits to fatigue cracks
Although initially developed for Ni-based superalloy systems, HOTPITS can be redefined for any material system, and is being enhanced to model the effects of coating materials.
Results – HOTPITS was successfully validated against laboratory experiments at the University of Virginia. This Phase II success led to this team being awarded NASA’s first ever Sequential Phase II STTR. In this ongoing R&D program, the modular nature of HOTPITS has provided greater insight for GTE design engineers into the processes at work in hot corrosion. This active state prediction was validated against fielded engine performance data from across several platforms at Rolls-Royce.
HOTPITS is implemented in the MicroFaVa material modeling software and DARWIN system for reliability analysis used extensively in the industry.