1st day, Tuesday, February 19, 2008
Session 1, Paradigm Shift in Life Prognosis
Peter B. Nagy, Chairperson
Keynote Lecture: Operational Needs for a Paradigm Shift in Life Prognosis, Robert S. Fredell Colonel, HQ AF/ST
The US Air Force operates a 6000-aircraft fleet with an average age of 25 years, while carrying out a recapitalization program that replaces only about 60 aircraft per year. To maintain flight safety, USAF maintenance crews carry a high structural inspection burden to support the successful damage-tolerance-based Aircraft Structural Integrity Program. Recent experience with the F-15 fleet has shown that even a high burden of inspections will not assure flight safety if the inspections are done in the wrong place. Simultaneous breakthroughs on three fronts: structural materials, heavy maintenance, and airframe prognosis offer dramatic new opportunities to the design and development of "carefree" transport wing structures for new aircraft and for the structural life extension of aging aircraft. The “carefree structures” concept, first proposed by Hinrichsen, suggests that truly optimized structures are not designed simply for minimum weight; rather, they are optimized for minimum life cycle costs. This requires material durability, extremely long damage-free lives and freedom from corrosion, along with resistance to impact and lightning and designed-for simplified inspection and repair. Combined with a design for effective state awareness allows an innovative high-velocity maintenance approach that simultaneously offers dramatic life cycle cost-saving benefits over legacy designs while ensuring a high level of safety and availability.
Opportunities and Challenges in Damage Prognosis for Materials and Structures in Complex Systems, James M. Larsen, Reji John, and Eric Lindgren, AFRL/RX
Key challenges, motivations, and historical perspective for prognosis of Air Force systems are outlined from the perspectives of materials and structures in turbine engines and aircraft. The overarching long-term vision for prognosis offers tremendous opportunity to attack the costs of system maintenance and operation, while also enabling the development of new systems having unprecedented capability and reliability.
Prognosis – A Vision for 2030, Thomas A Cruse, DARPA/DSO Consultant, AFRL (retired)
Over the next 25 or so years, the nation will field a variety of new air, space and cyber systems to support the defense needs of the nation. Air vehicles, advanced propulsion, and revolutionary avionics and ISR (intelligence, surveillance, and reconnaissance) electronics packages of the future require, for both cost and readiness purposes, the ability to be fully self-diagnostic. That is, from the mission commander to the maintainer to the logistician (supply chain), the operators responsible for each air vehicle, turbine engine, and electronics package must be informed in real-time of the health status and future capability of the system. Prognosis 2030 is an attempt to motivate the funding of a basic research program to achieve revolutionary changes in the three fundamental elements for prognosis – state modeling that translates system environment history into response at the material damage level, the ability to fully characterize the as-processed and manufactured material state (a so-called material/system “fingerprint”) for each system element, and the idea of a “virtual sensing” strategy that tightly integrates these with new approaches to traditional NDE to define the current and future state of the system as a digital state sensor. The basic research program to support this vision is one that needs to be built around the key challenges where fundamental breakthroughs are required and possible.
Session 2, OEM and End-User Perspectives
Thomas A Cruse, Chairperson
Effective State Awareness Information is Enabling for System Prognosis, Mark Derriso, AFRL/RBSA
The Department of Defense has recently placed emphasis on the development of Integrated Systems Health Management (ISHM) techniques. The ultimate goal of ISHM is to provide a system level capability assessment along with a prognosis regarding the potential risk and success of future missions. To meet this goal, ISHM requires an accurate diagnosis of the current condition of the various subsystems. Consider, for example, the health of the air vehicle structure. The structural design includes numerous uncertainties during the design process. Design techniques are utilized to accommodate for these uncertainties, resulting in a design with an initial prognosis of no failure during the design life. Over time, however, requirements change and the subsystem and system prognosis change. This presentation outlines the role of effective state awareness in enabling system prognosis and discusses the basic technical challenges relating to diagnosing the current state of the vehicle subsystems.
Challenges Associated with Implementing an Integrated Structural Health/Life Management System for Aerospace Structures, Joseph P. Gallagher, Aeronautical Systems Center (retired)
Damage characterizes and defines the material state that directly relates to structural health and to remaining life for airframes. The aircraft structural integrity program (ASIP), as defined by MIL-STD-1530C, provides the framework and processes associated with the initial and continuing airworthiness certification of USAF airframe structures. The ASIP framework and processes facilitate the characterization of the damage state in airframes (aircraft structures) as a function of time in-service. The concept of multiple methods for characterizing the damage state is presented. These multiple methods include: 1) the on-board usage monitoring method (provided by virtual sensors); 2) the off-board damage monitoring method (based on maintenance/inspection data collection; and, 3) storage of damage information), and the on-board damage monitoring method (provided by damage-event sensors). By focusing on the fatigue mechanism and similar aging mechanisms, several types of damage distributions are identified. The presentation identifies several serious challenges associated with characterizing the current state of damage and its accurate projection into a future state. The presentation provides some guidance on the way ahead.
Airframe Manufacturer’s Perspective of Materials State Awareness, Richard Bossi and Kent Ruffing, The Boeing Company
Aircraft mission readiness capability over the lifetime of an airframe is critical. Nondestructive Evaluation (NDE) and Structural Health Monitoring (SHM) technologies play an increasingly interdependent role in providing the material state awareness information used to decide mission readiness. A number of development activities are needed to effectively integrate NDE and SHM. One particularly important area is the service data records information and lessons learned from those records for aircraft airframes and systems.
The Quest for Personalized Engine Maintenance, Edmund Hindle, General Electric Aviation
The United States Air Force is in the midst of a serious operational cost trend that if not corrected will result in a lose-lose situation for both the Air Force and the OEM. A revolutionary paradigm shift must take place to reverse the aircraft sustainment demand for funding. A prognosis based life approach can go a long way towards reversing the operating cost trend but has scientific and technical issues that must be addressed. Current fleet management capability is constrained by uncertainty in the current state of the individual aircraft engines. The ability to sense or measure the damage state of an individual part is limited at best. Further, specific part operational severity is not captured with the current lifing process, hence many components are not operating to their life entitlement because the life is based on fleet weighted average missions. Unlike the fixed interval inspections currently being performed, precise assessment is required for prognosis-based lifing. The key considerations in this assessment are 1) the fidelity of the analysis tools 2) the definition of the boundary conditions (or environmental conditions used by the analysis tools) 3) improved understanding of diagnostics and engine faults with enhanced troubleshooting tools.
Session 3, Prognostics
Dave L. McDowell, Chairperson
Predicting Fatigue Processes in Aluminum Alloys Using Stochastic Multi-Scale Simulations, Paul Wawrzynek and Anthony Ingraffea, Cornell University
The goal of this research is to develop physics-based models for stochastically simulating all stages of microstructurally small fatigue crack formation in aluminum alloys. The salient features of our approach are: A) The use of statistically representative, realistic microstructures as a starting point for our simulations; B) The use of polycrystal plasticity models to compute accurately stress and strain fields in polycrystals using the finite element method; and C) The use of an explicit geometric representational approach in a multi-scale methodology. At each length scale, fatigue crack precursors, such as grain boundary or particle decohesion, are represented geometrically in the finite element model, and allowed to evolve through changes in the underlying geometric and mesh models. We will report on progress in development, verification, and validation of our probabilistic simulation models, and show example simulations specific to aluminum alloy 7075-T651.
Physically-Based Modeling in State-Awareness Monitoring Strategies, Dave McDowell, Georgia Tech
The role of physically-based models for crack formation and early growth near hot spots is considered in terms of its contribution to state-of-awareness of aerospace propulsion and airframe structural applications. It is envisioned that computational microstructure-scale modeling of material deformation and damage will be combined with simulations and experiments characterizing signals from sensors interrogating critical locations in metallic systems, and then linked to algorithms for predicting remaining life with consideration of uncertainty.
The Witness Sample Approach to Prognostics, Alten Grandt, Purdue University
The goal of this talk is to overview a proposed “witness sample” method that could provide a cost effective approach for monitoring structures for evidence of potential fatigue and/or corrosion damage. The approach employed here is to attach a small pre-cracked coupon (crack gage) to the structural member of interest. Structural loads are transferred to the coupon and result in measurable crack growth in the coupon. Fracture mechanics techniques enable one to relate crack growth in the coupon with the nucleation and/or extension of an assumed structural crack. Thus, the crack gage provides a simple and rapid nondestructive technique to assess the potential for fatigue damage in the structural member of interest. Since the witness sample is also sensitive to the same thermal-chemical environment seen by the structure, it also monitors the detrimental effects of environment on structural integrity. The end result would be a simple nondestructive evaluation technique to enable critical maintenance and repair decisions by flight line personnel.
Impact of Multiple On-Board Inspections on Cumulative Probability of Detection, Harry Millwater, University of Texas at San Antonio
This research examines the simulation of recurring automated inspections resulting from simulated on-board “crack” sensors, and their potential effect on reducing the probability-of-fracture of structural components. We assert that recurring inspections for an automated system should be modeled as dependent with respect to the first inspection due to the largely repeatable aspects of the sensor and data collection system. This assertion has a large effect on the computed probability of detecting a crack and alleviates the substantial overprediction of sensor efficacy generated using the assumption of independent inspections for automated systems. Furthermore, it is demonstrated that the fundamental feature that determines the efficacy of a recurring automated on-board sensor is the probability of detecting a crack of critical size, i.e., the size that will cause fracture, and this feature is by and large separate from the shape of the POD curve. This information can be used to determine the required accuracy of an on-board automated inspection to achieve a specified reliability of a structural component.
A New Approach for Investigating Crystal Stresses that Drive the Initiation of Fatigue-Induced Defects in Structural Alloys, Matt Miller and Paul Dawson, Cornell University
Fatigue failure continues to be among the most challenging limitations for the design of aerospace engines and structures. While post-mortem microstructural analyses can provide insight into specific fatigue-related accidents - they yield little information that will improve our ability to actually predict fatigue failures. Design for fatigue continues to follow conservative, yet costly guidelines. Due to complicated material microstructures and anisotropic single crystal properties, the prediction of microcrack initiation within polycrystalline metallic alloys and composite systems has proven illusive. This talk describes a hybrid experimental-simulation project – currently being funded by AFOSR - that focuses on predicting the cyclic loading-induced evolution of micromechanical state that leads to microcrack initiation. The key aspects of the methodology are: i) in-situ synchrotron x-ray diffraction / cyclic loading experiments to measure the evolving distribution of three-dimensional stress states at the scale of a single crystal within the aggregate, ii) elasto-viscoplastic polycrystal plasticity simulations that discretize each crystal using finite elements and iii) a cohesive digital framework for combining and comparing simulated and experimental results. The dynamic nature of the in-situ experiments, the computationally intensive multiscale simulations and the combined experimental and numerical aspects of the overall approach should make the description of this project relevant to the discussion around the proposed BAA for real time assessment of aerospace structural health.
2nd day, Wednesday, February 20, 2008
Session 5, Physical Foundations
Kumar V. Jata, Chairperson
Role of Environment in Mechanism-based Modeling of Airframe Fatigue, Rick P. Gangloff, University of Virginia
Environment impacts the driving forces and damage mechanisms for each stage of fatigue in structural metals including modern aluminum alloys. Such effects must be incorporated in inspection/sensing, properties data, and physics-based damage mechanism modeling inputs to next-generation performance prognosis. With notable exceptions, environmental effects on structural integrity are not addressed quantitatively in life management methods for DoD and civilian structures. Fundamental uncertainties are associated with:
• Environment electrochemistry spectra that vary with base and flight operations, and affect damage accumulation in transient modes.
• Localized-occluded geometric and chemical factors that are associated with a corroded surface and confound state sensing of features governing transition to fatigue cracking.
• Physics-based modeling of the distribution of lives for transition from corrosion to fatigue crack formation and early growth that parallels cutting edge work on microstructure-scale defect-initiated damage accumulation.
• Physics-based modeling of gradated nano-to-micro scale damage at a fatigue crack tip that is limited because fundamental damage mechanisms and failure criteria are not quantified.
Research on mechanisms of these environment-fatigue interactions is highlighted to initiate discussion of environmental effects in prognosis and to extend fundamental plasticity modeling of the stages of fatigue.
Multiphysics Foundations for Material State Change Prognosis in Material Systems, Ken L. Reifsnider, F. Chen, and X. Xue, University of South Carolina
Material systems are nano-structured heterogeneous materials in which there is a “composite effect,” a behavior or performance that is more than the linear sum of that of the constituents. In such systems, discrete events, such as damage or degradation – or other changes of material state, are typically spatially distributed throughout most of the life of the material in an applied environment. The advent of functional material systems, such as fuel cells, electrolyzers, and many membrane-based systems (for processing, medical devices, etc.) has taught us that these distributed events are characterized by changes in many physical aspects of the material state, such as mechanical integrity, conductivity, density, electrical and magnetic properties, and diffusion / transport characteristics. These state changes are often coupled, by temperature, mass transport, heat transport, etc. Analysis of such coupled phenomena has become an important element of engineering. “Multiphysics analysis” is now supported by well established methodology, and even some commercial codes. However, applications of such analysis methods are somewhat disperse, and understandings of the problems addressed by those methods are somewhat remedial at this point.
This talk will examine the fundamental concept of using multiphysics methods and analysis to link mechanical changes in state (e.g., distributed ‘damage’) to other multiphysics-based coupled changes in state using the experimental methodology referred to as electrochemical impedance spectroscopy (EIS). The foundations for the concept will be outlined, and rudimentary (preliminary) results will be shown. Limitations and opportunities will be discussed and some future plans for continuing work will be addressed.
Electromagnetic Materials State Awareness Monitoring, Peter B. Nagy, University of Cincinnati
Electromagnetic methods offer unique opportunities for materials state awareness monitoring. A variety of sensors can be built based on electric, magnetic, electromagnetic, and thermoelectric principles. These very simple and robust sensors can detect and quantitatively characterize subtle environmentally-assisted and/or service-related changes in the state of metals, such as microstructural evolution, phase transformation, plastic deformation, hardening, residual stress relaxation, increasing dislocation density, etc. In most cases, the detection sensitivity is sufficiently high for the purposes materials state awareness monitoring and the feasibility of the sensing method is mainly determined by its selectivity, or the lack of it, to a particular type of damage mechanism.
Nonlinear Ultrasonic Materials State Awareness Monitoring, Larry J. Jacobs and Jianmin Qu, Georgia Institute of Technology
This talk presents an overview of recent progress on the use of nonlinear ultrasonic techniques to quantitatively characterize material state. In particular, this talk will focus on methodologies that are based on measuring higher order harmonics to track microstructure changes in the material due to fatigue damage. As a single frequency wave propagates in a material with damage, higher harmonic frequencies are generated. This higher harmonic generation is quantified with an absolute nonlinearity parameter β. Recent experimental measurements have indisputably established the close correlation between β and the degree of damage. Since b can be measured using nonlinear ultrasound, such a correlation enables the nondestructive characterization of the fatigue life of a metallic component. Various nonlinear ultrasonic techniques based on bulk, Rayleigh and Lamb waves will be discussed and reviewed. Measurement results will be presented for aluminium alloys and Ni-base super alloys subjected to quasi-static monotonic tension and low-cycle fatigue. Recent progress on developing quantitative physics-based models to relate b to material microstructure changes has followed two somewhat different approaches. One is to directly relate b to the dislocation substructure, and the other is to relate b to cumulative plastic strain developed in the material. The former provides a clear physical interpretation of b, while the latter allows for the characterization of material state and the estimation of the remaining fatigue life of a component based on nondestructive nonlinear ultrasonic measurements. Both approaches will be discussed and their predictions are compared to experimental measurements.
Session 6, Technical Capabilities
Mark J. Schulz, Chairperson
Exploring the implications of a Bayesian approach to Materials State Awareness, Bruce Thompson, Iowa State University
Materials State Awareness has, at its end goal, the probabilistic prognosis of the state and future serviceability of materials components, structures and systems. Many tools are seen as providing inputs to this prediction, including models of damage evolution, direct measurements of the operational environments of individual structural components and systems (as opposed to fleet averages) which provide input to those predictions, and sensing the evolution of the state using either local or global sensors. This paper will address the challenge of integrating this disparate information, with emphasis on the sensing aspects, into a probabilistic prediction of the future service life and performance of the component/system. It is argued that a Bayesian approach, in which a priori knowledge of possible states is narrowed by the information available from sensors, is an attractive approach to realize the goal. Examples will be given that illustrate some of the key issues.
An Interdisciplinary to Understanding “Dwell Fatigue” in Ti Compressor Rotor Alloys, Jim Williams, Amit Bhattacharjee, Somnath Ghosh, Mike Mills, and Stan Rokhlin, Ohio State University
There is a fatigue failure mode in high temperature Ti alloys known as “Dwell Fatigue”. This mode is characterized by a major reduction (more than an order of magnitude) in fatigue life when the material is subjected to a loading pattern where the load is held at maximum value for a dwell period, typically several minutes. Further, this failure mode is typified by subsurface crack initiation which complicates detection of early stages of failure during field inspections. Because of its importance, the phenomenology of dwell fatigue has been thoroughly examined and is well defined. However, the fundamental reasons for the occurrence of dwell fatigue are not well understood.
This talk will first describe the phenomenology of dwell fatigue. Following this, results of a multi-year, interdisciplinary study of dwell fatigue in Ti-6Al-2Sn-4Zr-2Mo (+Si) will be described. The study has used solid mechanics modeling, detailed characterization methods and high resolution acoustic microscopy to provide a fundamental, semi-quantitative understanding of dwell fatigue. An important aspect of this failure phenomenon includes internal load re-distribution during the hold time at maximum load. This load re-distribution is caused by variations in stress relaxation rate as a function of local crystallographic orientation. This orientation dependence relative to the loading direction influences the degree of load re-distribution. The role of microtexture or regions of common alpha phase orientation on the reduction in fatigue life during dwell loading has been shown. Qualitative relationships between microtexture and dwell fatigue susceptibility will be described.
The current state of understanding and future directions to develop a quantitative understanding will also be described.
An Unified Approach to Structural Health Management and Damage Prognosis of Metallic Aerospace Structure, Aditi Chattopadhyay, Arizona State University
A comprehensive framework on structural health management (SHM) and damage prognosis of metallic aerospace components is being developed to provide reliable life cycle estimates of metallic aerospace components (AFOSR MURI). The procedure comprises a hierarchical framework of sensor data, information management, models and algorithms that span and integrate scales from microstructure to structural level. The methodological developments are being steered by closed-loop validations, incorporating both simulation and experimental test data. Specific research concentrations include: i) Physically-based multiscale modeling; (2) Models for in situ interrogation and detection; (3) Prognosis via state-awareness and life models. The methodological developments are steered by a closed-loop validation plan that incorporates both simulation and experimental test data. Representative results will be presented to provide an overview of this MURI project in the areas of multiscale modeling, damage classification and structural damage prognosis.
Damage State Awareness in Composite Laminates Via Ultrasonic Guided Waves, Cliff Lissenden and Joe Rose, Pennsylvania State University
As the amount of composite materials used in aircraft increases, so must our ability to monitor the damage state in composites. Ultrasonic guided wave based methods are attractive because they are capable of monitoring large regions from one location. However, the multilayered anisotropic nature of composite laminates makes guided wave propagation quite complex. The ability to understand guided wave excitation and propagation is fundamental to a physically based monitoring system that makes best use of activation, sensing, and signal processing capabilities. Penetration power and sensitivity improvement are the major challenges in guided wave based technologies for detection and quantification of structural damage in composites. We present a physically based understanding of ultrasonic guided wave propagation and excitation behaviors in composites, which is an essential building block for system design.