A Survey of Advanced Methods for Analysis and Modeling of Propulsion System Reliability

Electronic controls, propulsion system monitoring and health management and application of information technology to maintenance data capture and storage are enabling users to accumulate large amounts of reliability and related maintenance data. Effective analysis and exploitation of these data bases requires advanced tools to extract meaningful and actionable information. The challenges include “competing” failure modes and periodic hard time maintenance that “censors” information from impending failures, and a high number of failure modes that confound analysis. These factors impede accurate assessment of the impact of corrective action and different maintenance procedures on availability and maintenance costs. They confound understanding of the reliability of complex propulsion & power systems that would enable more representative modeling of system availability and maintenance costs for both existing and future applications. Tools providing more complete and accurate characterization of reliability information for complex systems are being developed for aerospace, nuclear and communications industries. These are surveyed and capability gaps identified with respect to commercial and military propulsion & power systems.

A Simplified Method to Evaluate the Impact of Component Design on Engine Performance

This paper presents a simplified and fast method to evaluate the impact of a single engine component design on the overall performance. It consists of three steps. In the first step, an engine system model is developed using available data on existing engines. Alongside the cycle reference point, a sweep of operating points within the flight envelop is simulated. The engine model is tuned to match a wide range of conditions. In the second step, the module that contains the engine component of interest is analyzed. Different correlations between the component design and the module efficiency are investigated. In the third step, the deviations in efficiency related to different component configurations are implemented in the engine baseline model. Eventually, the effects on the performances are evaluated. The procedure is demonstrated for the case of a two-spool turbofan. The effects of tip leakage in the low pressure turbine on the overall engine performance are analyzed. In today’s collaborative engine development programs, the OEMs facilitate the design process by using advanced simulation software, in-house available technical correlations and experience. Suppliers of parts have a limited influence on the design of the components they are responsible for. This can be rectified by the proposed methodology and give subcontractors a deeper insight into the design process. It is based on commercially available PC engine simulation tools and provides a general understanding of the relations between component design and engine performance. These relations may also take into account of aspects like production technology and materials in component optimization.

Turbine Blade Tip Clearance Measurement Instrumentation

New and future gas turbine engines are being required to provide greater thrust with improved efficiency, while simultaneously reducing life cycle operating costs. Improved component capabilities enable active control methods to provide better control of engine operation with reduced margin. One area of interest is a means to assess the relative position of rotating machinery in real-time, in particular hot section turbo machinery. To this end, Hamilton Sundstrand is working to develop a real-time means to monitor blade position relative to the engine static structure. This approach may yield other engine operating characteristics useful in assessing component health, specifically measuring blade tip clearance, time-of-arrival, and other parameters. UTC is leveraging its many years of experience with engine control systems to develop a microwave-based sensing device, applicable to both military and commercial engines. The presentation will discuss a hot section engine demonstration of a blade position monitoring system and the control system implications posed by a microwave-based solution. Considerations necessary to implement such a system and the challenges associated with integrating a microwave-based sensor system into an engine control system are discussed.

The Study of Thermal Properties and Micro Structures of YSZ

Thermal barrier coatings (TBCs) are used in gas turbine engines to achieve a better efficiency by allowing increased turbine inlet temperature and decreasing the amount of cooling air used. Plasma spraying is one of the most reliable methods to produce TBCs, which are generally comprised of a top coating of ceramic and a bond-coat of metal. Usually, the top coating is Yttria-Stabilized-Zirconia (YSZ), providing the thermal barrier effect. The bond-coat is typically a layer of M-Cr-Al-Y (where “M” stands for “metal”), employed to improve the attachment between the ceramic top-coat and the substrate. Due to the extreme temperature gradient presented in the plasma jet and the wide particle size distribution, during the coating process, injected ceramic powders may experience a significantly different heating process. Different heating history, coupled with the substrate preheating temperature, may affect the thermal properties of the YSZ layers. In this paper, four sets of mol 8% YSZ disks are fabricated under controlled temperatures of 1100°C, 1200°C, 1400°C and 1600°C. Subsequently the thermal properties and the microstructures of these YSZ disks are studied. The results indicate a strong microstructure change at a temperature slightly below 1400°C. For a high sintering temperature, a dense YSZ layer can be formed, which is good for gas tight operation; At low sintering temperature, say 1200°C, a porous YSZ layer is formed, which has the advantage of low thermal conductivity. For gas turbine TBC applications, a robust low thermal conductivity YSZ layer is desirable, while for Solid Oxide Fuel Cells, a gas-tight YSZ film must be formed. This study offers a general guideline on how to prepare YSZ layers, mainly by controlling the heating process, to form microstructures with desired properties.

Numerical Simulation of Partial Flame Failure in Gas Turbine Engine

A mathematical model for the simulation of engine start-up thermodynamics has been developed and validated against engine test data. This numerical model has been validated using engine test results for both single and multiple combustor flameouts, and reasonable agreement between test and simulation data has been observed. Numerical simulations have then been generated for flameout cases that have not been available from engine tests, such as flame failure in different combinations of combustors, and at different engine operating conditions. The mathematical model features object modeling of engine components with three gas compositions, being air, fuel, and combustion products. The combustion system has been represented by six combustors, and the gas stream from each combustor has been divided according to the number of the gas path thermocouples downstream from the combustion system. The effects of heat transfer within the combustors and turbine have been modeled. Two sets of thermocouples have been considered, the first being thermocouples installed in multiple combustor burners, and the second being an array of thermocouple probes which are circumferentially positioned in the engine hot gas path. All thermocouples have been modeled as first order dynamic systems. The numerical simulations have been successfully used to support development of a new partial flame failure detection method, which is based on the combined measurements from both sets of thermocouples. A range of numerical simulations have been conducted in order to assess the ability of this new detection algorithm to detect different partial flame failure scenarios, and to examine the sensitivity of the detection algorithm with respect to thermocouples faults.

An IPRP (Integrated Pyrolysis Regenerated Plant) Microscale Demonstrative Unit in Central Italy

The Integrated Pyrolysis Regenerated Plant (IPRP) concept is based on a Gas Turbine (GT) fuelled by pyrogas produced in a rotary kiln slow pyrolysis reactor; pyrolysis process by-product, char, is used to provide the thermal energy required for pyrolysis. An IPRP demonstration unit based on an 80 kWE microturbine was built at the Terni facility of the University of Perugia. The plant is made of a slow pyrolysis rotary kiln pyrolyzer, a wet scrubbing section for tar and water vapor removal, a micro gas turbine and a treatment section for the exhaust gases. This paper describes the plant layout and expected performance with different options for waste heat recovery.

Part Data Mining for Information Re-Use in a PLM Context

Difficulty in locating existing information in order to reuse it constitutes a major challenge to productivity. The use of PLM systems (Product Lifecycle Management) aims in particular to reduce the time and cost of developing a product by facilitating the re-use of existing parts or related information (process plans, tools, FEM, estimates, etc.). When information is alphanumerical, using search engines, such as those made popular on the internet, is efficient. However, a significant portion of information used in engineering rests within CAD (Computer Aided Design) models, making such search tools irrelevant. To aid in the re-use of information, two problems must be resolved: it is first necessary to be able to locate similar parts in the electronic database of the company, and then be able to systematically identify their differences. This article presents some of the results from our work on part, product and process data mining (P3DM). It focuses on tools developed to search similar 3D geometric models and to identify their differences. The PartFinder application locates similar parts by comparing signatures extracted from their solid representations. The 3DComparator aims to identify the differences in terms of Form and Fit between the identified parts. In both cases, the recommended approach is independent of the CAD system, and can also deal with parts represented by IGES or STEP files. Moreover, the approach does not require that the parts occupy the same position and have the same orientation in space. These two points, CAD and position independence, are the main benefits of our approach compared to other existing applications. Lastly, if the comparison takes place between two evolutions of the same geometrical representation of a part, a third tool allows the comparison of the specification trees. The SpecComparator is also presented briefly. An example based on industrial data illustrates the benefit that could be generated.

A Method to Compare Turbine Engine Airstart Times

Turbine engine airstarts are conducted throughout the aircraft airspeed/altitude envelope in ground-based simulation test facilities and in flight tests to ensure safe and reliable engine operation. Differences in airstart times are attributable to variations in engine turnaround speed (the engine core speed at which the airstart is initiated in spooldown airstarts); combustor lightoff time; installation effects such as customer bleed and power extraction; starter motor torque; fuel flow scheduling; and engine-to-engine variation and degradation. An analytical approach is presented to account for these differences and adjust engine airstart time for a low-bypass, twin-spool, military, turbofan engine. Two examples are presented illustrating the difference in airstart times and the analytical approach used to adjust the start times.

Electric Starter and Generator Systems (ESGS) for Gas Turbines: Making Platform Integration Easier

The US Navy has pursued gas turbine electric start systems since 2003. Such a system has been extensively tested at the Naval Surface Warfare Center, Carderock Division (NSWCCD) Land Based Engineering Site (LBES) in Philadelphia, PA. It was demonstrated on a General Electric (GE) LM2500 main propulsion engine as well as a Rolls Royce (RR) MT30 engine. Presently, the system is being refined and repackaged to undergo U.S. Navy qualification for production use. Given the performance success of electric start the next logical step is to extend its application to other engine lines such as the Ship Service Gas Turbine Generators (SSGTG). In order to facilitate platform integration, the electric start concept has been evolved into the Electric Start and Generation System (ESGS). As expected, this system has the ability to start a gas turbine by purely electrical means. Once the engine has reached idle speed or above, the ESGS becomes a generator capable of producing power. This power may be harnessed to address dark start capability on Surface Combatants. The ESGS configuration simplifies integration of bulk energy storage such as a flywheel device or battery pack. This will ensure availability to the engine under a loss of platform power scenario thus providing self-sustainability to all the gas turbine’s electrical functions. Another alternative is to continuously provide ESGS generated power back to the electrical grid in continuous support of the engine auxiliary systems. In this case, flywheels and batteries may be replaced by advanced transfer switches that redirect power where it is needed on demand. This paper describes a program undertaken by NSWCCD to carry out land based testing of an advanced design ESGS. An overview of system requirements is given from a perspective of platform integration. The system architecture is fully described. It is an evolution of ESGS technology that has been extensively tested on RR MT30 and GE LM2500 gas turbines at NSWCCD LBES. Compared with existing air and alternative hydraulic gas turbine starter systems, this system is more compact and provides the benefits of simplified platform integration. It incorporates energy storage to provide black start capability for the gas turbine. Battery and inertial energy storage technologies are discussed in detail for use with the ESGS.

Predicting the Performance of System for the Co-Production of Fischer-Tropsch Synthetic Liquid and Power From Coal

A co-production system based on FT synthesis reactor and gas turbine was simulated and analyzed. Syngas from entrained bed coal gasification was used as feedstock of low temperature slurry phase Fischer-Tropsch reactor. Raw synthetic liquid produced was fractioned and upgraded to diesel, gasoline and LPG. Tail gas composed of unconverted syngas and F-T light component was fed to gas turbine. Supplemental fuel (NG, or refinery mine gas) might be necessary, which was dependent on gas turbine capacity, expander through flow capacity, etc. FT yield information was important to the simulation of this co-production system. A correlation model based on Mobil’s two step pilot plant was applied. This model proposed triple chain-length-dependent chain growth factors and set up correlations among reaction temperature with wax yield, methane yield, and C2-C22 paraffin and olefin yields. Oxygenates in hydrocarbon phase, water phase and vapor phase were also correlated with methane yield. It was suitable for syngas, iron catalyst and slurry bed. It can show the effect of temperature on products’ selectivity and distribution. Deviations of C5+ components yields and distributions with reference data were less than 3%. To light gas components were less than 2%. User models available to predict product yields, distributions, cooperate with other units and do sensitive studies were embedded into Aspen plus simulation. Performance prediction of syngas fired gas turbine was the other key of this system. The increase in mass flow through the turbine affects the match between compressor and turbine operating conditions. The calculation was carried out by GS software developed by Politecnico Di Milano and Princeton University. The simulated performance assumed that the expander operates under choked conditions and turbine inlet temperature equals to NG fired gas turbine. A “F” technology gas turbine was selected to generate power. Various cases were investigated so as to match FT synthesis island, power island and gasification island in co-production systems. Effects of CO2 removal/LPG recovery, co-firing, CH4 content variation were studied. Simulation results indicated that more than 50% of input energy was converted to electricity and FT products. Total yield of gasoline, diesel and LPG was 136g-155g/NM3(CO+H2). At coal feed 21.9kg/s, net electricity exported to grid was higher than 100MW. Total production of diesel and gasoline (and LPG) was 118,000 tons(134,000tons)/Year. Under economic analysis conditions assumed in this paper, co-production system was economic feasible. The after tax profits can research 17 million EURO. Payback times were ranged from 6-7 years.