Jeff W. Bird

Benchmarking Gas Path Diagnostic Methods: A Public Approach

Recent technology reviews have identified the need for objective assessments of engine health management (EHM) technology. The need is two-fold: technology developers require relevant data and problems to design and validate new algorithms and techniques while engine system integrators and operators need practical tools to direct development and then evaluate the effectiveness of proposed solutions. This paper presents a publicly available gas path diagnostic benchmark problem that has been developed by the Propulsion and Power Systems Panel of The Technical Cooperation Program (TTCP) to help address these needs. The problem is coded in Matlab™ and coupled with a non-linear turbofan engine simulation to produce “snap-shot” measurements, with relevant noise levels, as if collected from a fleet of engines over their lifetime of use. Each engine within the fleet will experience unique operating and deterioration profiles, and may encounter randomly occurring relevant gas path faults including sensor, actuator and component faults. The challenge to the EHM community is to develop gas path diagnostic algorithms to reliably perform fault detection and isolation. An example solution to the benchmark problem is provided along with associated evaluation metrics. A plan is presented to disseminate this benchmark problem to the engine health management technical community and invite technology solutions.Copyright

Developing Data Mining-Based Prognostic Models for CF-18 Aircraft

The CF-18 (CF denotes Canadian Forces) aircraft is a complex system for which a variety of data are systematically being recorded: flight data from sensors, built-in test equipment data, and maintenance data. Without proper analytical and statistical tools, these data resources are of limited use to the operating organization. Focusing on data mining-based modeling, this paper investigates the use of readily available CF-18 data to support the development of prognostics and health management systems. A generic data mining methodology has been developed to build prognostic models from operational and maintenance data. This paper introduces the methodology and elaborates on challenges specific to the use of CF-18 data from the Canadian Forces. A number of key data mining tasks are examined including data gathering, information fusion, data preprocessing, model building, and model evaluation. The solutions developed to address these tasks are described. A software tool developed to automate the model development process is also presented. Finally, this paper discusses preliminary results on the creation of models to predict F404 no. 4 bearing and main fuel control failures on the CF-18.

Sensitivity of air-coupled ultrasound and eddy current sensors to bearing fault detection

For decades, vibration and oil analysis have usually been used to detect early bearing faults and track their progression over time. Progress has been seen in condition monitoring through vibration analysis of rolling element bearings using improved sensors and advanced signal processing techniques. In this paper, the authors investigate the use of air-coupled ultrasound and eddy current sensors as diagnostic tools for the detection of bearing faults. A series of experiments was carried out in a laboratory environment: localized defects with different sizes were created intentionally on the test bearing components simulating evolving cracks or other related faults. The resulting data for a constant bearing speed and load have shown that both sensors are capable of detecting different types of defects located on the bearing components. The data from the air-coupled ultrasound and eddy current sensors were also compared with those obtained from an accelerometer. The test method and the processing technique are described and the spectra of the different signals are analyzed and discussed.

A Comparative Study of Air-Coupled Ultrasound Sensor and Accelerometer in Detecting Bearing Defects

The aim of this preliminary study is to investigate the capabilities of a non-destructive evaluation (NDE) sensor in detecting bearing faults. An air-coupled ultrasound sensor is used for this purpose. An accelerometer, which is the standard sensor used in industry, is also used in order to compare the output spectrums of the two sensor signals. A defect was created intentionally on the bearing components to simulate a fatigue crack or other relevant defect. The power spectra of vibration signals measured by accelerometer and ultrasound sensor are compared and their advantages and disadvantages are determined.© 2005 ASME

Diagnosis of Turbine Engine Transient Performance With Model-Based Parameter Estimation Techniques

This review surveys knowledge needed to develop an improved method of modelling the dynamics of gas turbine performance for fault diagnosis applications. Aerothermodynamic and control models of gas turbine processes are examined as complementary to models derived directly from test data. Extensive, often proprietary data are required for physical models of components, while system identification (SI) methods need data from specially-designed tests. Current methods are limited in: tuning models to test data, non-linear effects, component descriptions in SI models, robustness to noise, and inclusion of control systems and actuators. Conclusions are drawn that SI models could be formulated, with parameters which describe explicitly the functions of key engine components, to offer improved diagnostic capabilities.© 1994 ASME

Evaluation of the Impact of Alternative Fuel Use on the Emissions and Performance of a Service-Exposed T56 Engine

Alternative fuel sources are becoming an operational reality; these fuels have the potential to reduce emissions, improve combustion characteristics and to increase fuel supply security. A test with a T56 turboprop engine was performed to demonstrate that a CHEFA/JP8 (Camelina Hydroprocessed Ester and Fatty Acids and standard JP8) fuel blend would meet operational requirements. The primary test objective was to assess whether a fuel change had an immediate impact on the engine condition, performance, emissions or vibration characteristics. This paper presents test results comparing engine performance with JP8 and a 50/50 blend of JP8 and CHEFA. Comparison runs were conducted before and after a 20 hour ground durability test with the CHEFA fuel blend. A nearly time-expired, nacelle-dressed T56 on an outdoor test stand was tested. The engine was equipped with minimally-intrusive non-standard pressure, temperature and emissions monitoring equipment, and a field vibration assessment suite in addition to the standard flight instrumentation. This paper discusses the test plan, data acquisition methods, results and data repeatability. The performance and emissions results are compared to the changes predicted theoretically from the fuel properties. Observations from the borescope inspections before, during and after the 20 hour durability test are also presented. The lessons learned in this test could be applied to future fuel or process-change tests, and the results provide a performance baseline for engine health assessment.© 2012 ASME

Review of Metrics and Assignment of Confidence Intervals for Health Management of Gas Turbine Engines

The development and evaluation of new diagnostic systems requires statistically-based methods to measure performance. Various metrics are in use by developers and users of diagnostic systems. Current metrics practices are reviewed, including receiver operating characteristics, confusion matrices, Kappa coefficients and various entropy techniques. A set of metrics is then proposed for assessment of diverse gas path diagnostic systems. The use of bootstrap statistics to compare metric results is developed, and demonstrated for a set of hypothetical data sets with a range of relevant characteristics. The bootstrap technique allows the expected range of the metric to be assessed without assuming a probability distribution. A method is proposed to develop confidence intervals for the calculated metrics. The application of a confidence interval could prevent a good diagnostic technique being discarded because of a lower value metric in one test instance. The strengths and weaknesses of the various metrics with derived confidence intervals are discussed. Recommendations are made for further work.Copyright

Engine Oil Condition Monitoring Using High Temperature Integrated Ultrasonic Transducers

High temperature integrated ultrasonic transducers (IUTs) made of thick piezoelectric composite films have been coated directly onto the lubricant oil supply and sump lines of a modified CF700 turbojet engine. These thick piezoelectric films are fabricated using a sol-gel spray technology. The center frequencies of these IUTs are in the range of 10 to 12 MHz. The top electrodes, electrical wires, conducting adhesive bond, connectors and cables have been tested successfully for temperatures of up to 500°C. By operating these IUT in transmission mode, the amplitude and velocity of transmitted ultrasonic waves across the flow channel of the lubricant oil in supply and sump lines were measured during engine operation. The results have shown that the strength of the ultrasonic waves is sensitive to the presence of air bubbles in the oil and that the ultrasound velocity is linearly dependent on oil temperature. Based on the sensitivity of ultrasound velocity to oil temperature, a method for real-time monitoring of engine oil degradation is proposed.Copyright

Propulsion System Prognostics R&D Through the Technical Cooperation Program

Cooperation and coordination at the international level are key features of cost effective management of propulsion system R&D. The technical cooperation program fosters such efforts for military-directed work among agencies in Australia, Canada, New Zealand, the United Kingdom, and the United States. The general arrangements and opportunities of the program are discussed, for example the means to exchange data, equipment and staff. In particular, the paper concentrates on the activity associated with the Aerospace Systems Group, Technical Panel 3, responsible for propulsion and power systems. Within the disclosure limits, the TP3 program is described. Significant examples of cooperation and successful development are highlighted including the helicopter integrated diagnostics system, other collaborative projects, workshops and data exchanges. Strategic planning directions and opportunities are identified from the points of view of technology development and military need.

A Framework of Prognosis and Health Management: A Multidisciplinary Approach

A multi-disciplinary team covering mechanical, materials and information technologies are investigating and developing appropriate technologies for aircraft prognostics and health management (PHM) applications. To integrate the efforts, a framework for PHM has been developed with three major functional blocks: Life Prediction, State Awareness, and Information Management. This framework provides a hierarchy for development and implementation. Achievements and future plans within this framework are presented with specific applications to military legacy systems and future acquisitions. Opportunities are described within some new initiatives for development and implementation.© 2007 ASME