Remaining Life Prediction for Individual Components from Sparse Data

There is an immediate need for individual component remaining life prediction capability that can provide near term relief through life management decision support. When failures occur in the fleet and fleet readiness and life cycle cost implications are substantial, early damage detection in individual components becomes critical. The promise of new early damage detection sensors such as the MWM-Arrays is encouraging, but precrack/early damage detection sensors may not always have access to dynamic components. To increase such access to early damage sensor data, JENTEK is developing non-invasive tooling to enable scanning/imaging of precrack/early stage damage and on-board sensor networks for monitoring static components. However, even for on-board solutions, most implementations will use portable data acquisition systems to download data from damage sensors. Thus, a method is needed to predict remaining life from sparse, early damage data. This proposed Phase I will focus on remaining life prediction and investigate feasibility of rapid failure risk assessment for rotorcraft dynamic components. In Phase I, we will demonstrate the feasibility of using sparse data on individual components, from one or two inspections, combined with empirical coupon subcomponent test data to estimate remaining life. In Phase II, we will demonstrate this in a component test. BENEFITS: Reliable, individual component remaining life prediction, using only sparse data on early damage states, would offer a new tool for total ownership cost reduction. It would also improve readiness and increase safety margins. The existence of a small number or even a single failed component presents an uncomfortable decision environment for both military and civilian aircraft operators. Expanding demands on rotary aircraft in war time and for natural disaster relief is further elevating the need to maintain high readiness levels, small logistics footprints and controlled life cycle costs.