Peter W. Hovey

Joint Estimation of NDE Inspection Capability and Flaw-Size Distribution for In-Service Aircraft Inspections

Nondestructive evaluation (NDE) is widely used in the aerospace industry, using scheduled maintenance inspections to detect cracks or other anomalies in structural and rotating components. Life prediction and inspection interval decisions in aerospace applications require knowledge of the size distribution of unknown existing cracks and the probability of detecting (POD) a crack, as a function of crack characteristics (e.g., crack length). The POD for a particular inspection method is usually estimated through laboratory experiments on a given specimen set. These experiments, however, cannot duplicate the conditions of in-service inspections. Quantifying the size distribution of unknown existing cracks is more difficult. If NDE signal strength is recorded at all inspections and if crack-length information is obtained after “crack find” inspections, it is possible to estimate the joint distribution of crack length, noise response, and signal response. This joint distribution can then be used to estimate both the in-service POD and the crack-length distribution at a given period of service time. In this article, we present a statistical model to describe the data and illustrate a Bayesian method to do the estimation and quantify uncertainty.

Joint estimation of the flaw size distribution and pod function

Life prediction and inspection interval decisions in aerospace applications require knowledge of the size distribution of unknown existing cracks and the probability of detecting a crack (POD), as a function of crack characteristics (e.g., crack size). The POD for a particular inspection method is usually estimated on the basis of experiments to a given specimen set. These experiments, however, cannot duplicate the conditions of in service inspections. Quantifying the size distribution of unknown existing cracks is more difficult. If reasonably precise crack size information is available for specimens with “crack find” decisions through inspection procedures, it is possible to estimate the joint distribution of crack size and signal response. This joint distribution can then be used to estimate both the in service POD and the crack size distribution. In this paper, we present a statistical model and methodology to do this estimation. We also illustrate the method on one set of simulated data.

USING A BAYESIAN MODEL TO JOINTLY ESTIMATE THE FLAW SIZE DISTRIBUTION AND THE POD FUNCTION

In this paper we extend previous work by the authors to jointly estimate the flaw size distribution and the POD function from simulated field inspection data. Similar to our previous work, we assume that when a crack is above a detection threshold, both the signal amplitude and the flaw size are recorded. For a signal that is above the noise floor, but below the detection threshold, only the amplitude is recorded. At all other locations we know only that the signal is below the noise floor, i.e. left censored. Now our model allows different airplanes to have different crack growth rates, and the distribution of crack growth rates is to be estimated from the data. To estimate the parameters of the model, we use a Bayesian formulation that provides a convenient structure for estimating the plane‐to‐plane differences. The Bayesian formulation also allows the use of prior information based on knowledge of physics or previous experience with similar inspection situations. For example, there may be useful infor...

The Test Plan for the Next Air Force NDI Capability and Reliability Assessment Program

Nondestructive Inspection (NDI) plays a key role in assessing the condition of materials, components and structures comprising aeronautical systems. Formal design specifications developed in the Air Force Aircraft Structural Integrity Program (ASIP) and Engine Structural Integrity Program (ENSIP) call for periodic inspections at defined levels of NDI flaw detection probabilities to avert failure or functional impairment. An essential link in establishing inspection intervals is the ability of the NDI system to find flaws. This paper describes the statistical aspects of the next Air Force evaluation program for NDI in airframe components.

USING THE INFORMATION IN FIELD SERVICE INSPECTIONS TO ASSESS THE DAMAGE STATE OF THE AIRCRAFT AND THE DETECTION CAPABILITY OF THE INSPECTION SYSTEM

Fatigue crack growth is a key contributor to a decline in reliability of aging aircraft. Knowledge of the distribution of crack sizes is essential to properly evaluate the ability of aging aircraft to continue to perform safely in future missions. The primary tool for determining the crack size distribution is nondestructive inspection. However, not all cracks will be detected, so interpreting the information in the sizes of detected cracks is difficult. This paper will discuss aBayesian approach to using the results of in‐service inspections to quantify the POD function for the inspection system and the distribution of the sizes of the cracks present in aging aircraft.

Estimating the Distribution of the Sizes of Flaws Remaining After an Inspection

Abstract : The U.S. Air Force plans for maintenance and retirement of aircraft based in part on fatigue crack growth models. Periodic inspections are used to help assess airworthiness and plan for future inspections. Nondestructive inspections are not perfect so some cracks are missed and the likelihood that an individual crack is detected is a function of the size of the crack when inspected. Additionally, the crack size distribution is related to the number of flight hours the aircraft has experienced, so not all inspection results come from the same distribution. In a recent study several models were compared that utilize the capability of the inspection system and the variation between aircraft and times of inspections to estimate the distribution of sizes of cracks that were missed during the inspection. This white paper summarizes those results and identifies some methods for extending them.