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

The CF-18 aircraft is a complex system for which a variety of data are systematically being recorded: operational flight data from sensors and Built-In Test Equipment (BITE) and maintenance activities recorded by personnel. These data resources are stored and used within the operating organization but new analytical and statistical techniques and tools are being developed that could be applied to these data to benefit the organization. This paper investigates the utility of readily available CF-18 data to develop data mining-based models for prognostics and health management (PHM) systems. We introduce a generic data mining methodology developed to build prognostic models from operational and maintenance data and elaborate on challenges specific to the use of CF-18 data from the Canadian Forces. We focus on a number of key data mining tasks including: data gathering, information fusion, data pre-processing, model building, and evaluation. The solutions developed to address these tasks are described. A software tool developed to automate the model development process is also presented. Finally, the paper discusses preliminary results on the creation of models to predict F404 No. 4 Bearing and MFC (Main Fuel Control) failures on the CF-18.

An Automated Energy Conservation Decision Support System for Navy Gas Turbines

Energy conservation measures currently employed by U.S. Navy surface combatants require labor-intensive, time-consuming data entry from which fuel curves are generated to drive each ship’s propulsion plant machinery alignment. From these rudimentary curves optimal transit speeds, configurations, and refueling requirements are determined for specific operational demands and mission profiles. This paper describes an automated process for optimizing shipboard fuel consumption rates by integrating advanced diagnostic and maintenance optimization techniques with the onboard data information system. The automated energy conservation decision support system described herein addresses fossil fuel propulsion (gas turbines, steam turbines, and diesel engines), power generation and auxiliary systems. The software tool consists of diagnostic, fuel management, and maintenance modules. The diagnostic module tracks and trends the health state of components that use fuel (and their supporting systems) to provide real-time information on the impact of their current condition on fuel consumption. The fuel management module automates data collection and the generation of fuel curves through open-systems architecture communication with ICAS. It also enables planning by recommending an optimal machinery configuration to minimize fuel consumption based on either speed or time to destination constraints. Additionally, a fuel management module provides real-time information on fuel consumption and optimizes the load of each component based on its health condition, operating requirements and the number and condition of similar components. Finally, overall decision support comes from the maintenance management module that tracks the maintenance actions being performed on fuel consuming systems and recommends future maintenance to be performed (from a fuel conservation standpoint) based on current health information.

Compressor Performance 2D Modelling at Reverse Flow Conditions

After a shaft failure the compression system of a gas turbine is likely to surge due to the heavy vibrations induced on the engine after the breakage. Unlike at any other conditions of operation, compressor surge during a shaft over-speed event is regarded as desirable as it limits the air flow across the engine and hence the power available to accelerate the free turbine. It is for this reason that the proper prediction of the engine performance during a shaft over-speed event claims for an accurate modelling of the compressor operation at reverse flow conditions. The present study investigates the ability of the existent two dimensional algorithms to simulate the compressor performance in backflow conditions. Results for a three stage axial compressor at reverse flow were produced and compared against stage by stage experimental data published by Gamache. The research shows that due to the strong radial fluxes present over the blades, two dimensional approaches are inadequate to provide satisfactory results. Three dimensional effects and inaccuracies are accounted for by the introduction of a correction parameter that is a measure of the pressure loss across the blades. Such parameter is tailored for rotors and stators and enables the satisfactory agreement between calculations and experiments in a stage by stage basis. The paper concludes with the comparison of the numerical results with the experimental data supplied by Day on a four stage axial compressor.

Development and Experimental Validation of a Modelling Tool for Humid Air Turbine Saturators

This work presents the re-engineering of the TRANSAT 1.0 code which was developed to perform off-design and transient condition analysis of Saturators and Direct Contact Heat Exchangers. This model, now available in the 2.0 release, was originally implemented in FORTRAN language, has been updated to C language, fully coded into MATLAB/Simulink® environment and validated using the extensive set of data available from the MOSAT project, carried out by the Thermochemical Power Group of the University of Genoa. The rig consists of a fully instrumented modular vertical saturator, which is controlled and monitored with a LABVIEW® computer interface. The simulation software showed fair stability in computation and in response to step variation of the main parameters driving the thermodynamic evolution of the air and water flows. Considering the actual mass flow rates, a geometric similitude was used to avoid calculation instability due to flows under 100 g/s. Overall the model proved to be reliable and accurate enough for energy system simulations.

Performance Study of a 1 MW Gas Turbine Using Variable Geometry Compressor and Turbine Blade Cooling

This work presents the performance study of a 1 MW gas turbine including the effects of blade cooling and compressor variable geometry. The axial flow compressor, with Variable Inlet Guide Vane (VIGV), was designed for this application and its performance maps synthesized using own high technological contents computer programs. The performance study was performed using a specially developed computer program, which is able to numerically simulate gas turbine engines performance with high confidence, in all possible operating conditions. The effects of turbine blades cooling were calculated for different turbine inlet temperatures (TIT) and the influence of the amount of compressor-bled cooling air was studied, aiming at efficiency maximization, for a specified blade life and cooling technology. Details of compressor maps generation, cycle analysis and blade cooling are discussed.

Development of a Medium-Size Steam Reformer for CHP Applications Based on PEM Fuel Cells

This paper describes technological of a fuel processor for hydrogen production able to convert 10 cubic meters of methane per hour. This device has been developed to feed hydrogen CHP suitable for the most common residential applications. The measured conversion efficiencies are extremely high: after the steam reformer the results are 76%H2; 18%CO2; 0,5%CH4; 5%CO; but the carbon monoxide is totally reduced throughout the water gas shift and the partial oxidation which contemporarily increase the hydrogen to over 77%. According to these results, this fuel processor is one of the first middle sized reformer to achieve, at comparable costs per cubic meter, conversion performance that were normally obtained only by industrial reforming plants.

First Unsteady Pressure Measurements With a Fast Response Cooled Total Pressure Probe in High Temperature Gas Turbine Environments

This paper presents the first experimental engine and test rig results obtained from a fast response cooled total pressure probe. The first objective of the probe design was to favor continuous immersion of the probe into the engine to obtain time series of pressure with a high bandwidth and therefore statistically representative average fluctuations at the blade passing frequency. The probe is water cooled by a high pressure cooling system and uses a conventional piezo-resistive pressure sensor which yields therefore both time-averaged and time-resolved pressures. The initial design target was to gain the capability of performing measurements at the temperature conditions typically found at high pressure turbine exit (1100–1400K) with a bandwidth of at least 40kHz and in the long term at combustor exit (2000K or higher). The probe was first traversed at the turbine exit of a Rolls-Royce Viper turbojet engine, at exhaust temperatures around 750 °C and absolute pressure of 2.1bars. The probe was able to resolve the high blade passing frequency (≈23kHz) and several harmonics up to 100kHz. Besides the average total pressure distributions from the radial traverses, phase-locked averages and random unsteadiness are presented. The probe was also used in a virtual three-hole mode yielding unsteady yaw angle, static pressure and Mach number. The same probe was used for measurements in a Rolls-Royce intermediate pressure burner rig. Traverses were performed inside the flame tube of a kerosene burner at temperatures above 1600 °C. The probe successfully measured the total pressure distribution in the flame tube and typical frequencies of combustion instabilities were identified during rumble conditions. The cooling performance of the probe is compared to estimations at the design stage and found to be in good agreement. The frequency response of the probe is compared to cold shock tube results and a significant increase in the natural frequency of the line-cavity system formed by the conduction cooled screen in front of the miniature pressure sensor were observed.

Gas Turbine Starting Systems Used Aboard U.S. Navy Large Surface Combatants

There are various methods used to start marine gas turbine engines on large naval surface combatants. Methods include pneumatic, mechanical, hydraulic, and electric starting systems. This paper gives an overview of basic starting requirements, describes each method used on large surface combatants, and identifies which systems are used on many of the U.S. Navy surface combatants.

Optimization of Turbofan Propulsion Cycle Using a Genetic Algorithm

This paper presents an evolutionary approach as the optimization framework to design for the optimal performance of a high-bypass unmixed turbofan to match with the power requirements of a commercial aircraft. The parametric analysis had the objective to highlight the effects of the principal design parameters on the propulsive performance in terms of specific fuel consumption and specific thrust. The design optimization procedure based on the genetic algorithm PIKAIA coupled to the developed engine performance analyzer (on-design and off-design) aimed at finding the propulsion cycle parameters minimizing the specific fuel consumption, while meeting the required thrusts in cruise and takeoff and the restrictions of temperatures limits, engine size and weight as well as pollutants emissions. This methodology does not use engine components’ maps and operates on simplifying assumptions which are satisfying the conceptual or early design stages. The predefined requirements and design constraints have resulted in an engine with high mass flow rate, bypass ratio and overall pressure ratio and a moderate turbine inlet temperature. In general, the optimized engine is fairly comparable with available engines of equivalent power range.

Impact of Operating Conditions and Design Parameters on Gas Turbine Hot Section Creep Life

This paper investigates the relationship between design parameters and creep life consumption of stationary gas turbines using a physics based life model. A representative thermodynamic performance model is used to simulate engine performance. The output from the performance model is used as an input to the physics based model. The model consists of blade sizing model which sizes the HPT blade using the constant nozzle method, mechanical stress model which performs the stress analysis, thermal model which performs thermal analysis by considering the radial distribution of gas temperature, and creep model which using the Larson-miller parameter to calculate the lowest blade creep life. The effect of different parameters including radial temperature distortion factor (RTDF), material properties, cooling effectiveness and turbine entry temperatures (TET) is investigated. The results show that different design parameter combined with a change in operating conditions can significantly affect the creep life of the HPT blade and the location along the span of the blade where the failure could occur. Using lower RTDF the lowest creep life is located at the lower section of the span, whereas at higher RTDF the lowest creep life is located at the upper side of the span. It also shows that at different cooling effectiveness and TET for both materials the lowest blade creep life is located between the mid and the tip of the span. The physics based model was found to be simple and useful tool to investigate the impact of the above parameters on creep life.