Fault Identification and Health Monitoring of Gas Turbine Engines Using Hybrid Machine Learning-based Strategies

Ahealth monitoring scheme is developed in this work by using hybrid machine learning strategies to iden-tify the fault severity and assess the health status of the aircraft gas turbine engine that is subject to component degrada-tions that are caused by fouling and erosion.The proposed hybrid framework involves integrating both supervised recur-rent neural networks and unsupervised self-organizing maps methodologies,where the former is developed to extract ef-fective features that can be associated with the engine health condition and the latter is constructed for fault severity modeling and tracking of each considered degradation mode.Advantages of our proposed methodology are that it ac-complishes fault identification and health monitoring objectives by only discovering inherent health information that are available in the system I/O data at each operating point.The effectiveness of our approach is validated and justified with engine data under various degradation modes in compressors and turbines.

Numerical simulation method of surge experiments on gas turbine engines

In order to obtain the surge margin of an aero-engine during its operation,an engine surge experiment is required.A multi-dimensional simulation method for an aero-engine is established in this paper.The simulation of a surge experiment using high-pressure air-injection is then carried out on a turbo-shaft engine to obtain the surge boundary using this method.More specifically,firstly,a body-force model is employed to calculate the compressor performance owing to its capability of capturing the main three-dimensional features of compressor surge and avoiding excessive simulation time required by the traditional fully-three-dimensional Reynolds Averaged Navier-Stokes(RANS)method.Then,a one-dimensional model combining a lumped-parameter plenum model is used for the combustor to account for the propagation of pressure waves and the heat-release process,and a zero-dimensional throttle model is used to mimic the choking effect at the turbine nozzle.Finally,the air-injection system is modeled by imposing an injection boundary condition,which can be used conveniently in changing injection parameters.Based on the established method,the influences of different test parameters,such as the air-injection location,the pressure,the orifice size,the number of injection orifices,and the injection time duration on the surge characteristics and boundary are further studied,which offer effective guidance to optimize an actual experimental design.

Endwall aerodynamic losses from turbine components within gas turbine engines

A survey of research on aerodynamic loss investigations for turbine components of gas tuibine engines is presented.Experimental and numerically predicted results are presented from investigations undertaken over the past 65 plus years.Of particular interest are losses from the development of secondary flows from airfoil/endwall interactions.The most important of the airfoilAmdwall secondary flows are passage vortices,counter voitices,and corner vortices.The structure and development of these secondaiy flows are described as they affect aerodynamic perfonnance within and downstream of turbine passage flows in compressible,high speed flows with either subsonic or transonic Mach number distributions,as well as within low-speed,incompressible flows.Also discussed are methods of endwall contouring,and its consequences in regard to airfoil/endwall secondary flows.

Preparation and characterization of LPPS NiCoCrAlYTa coatings for gas turbine engine

NiCoCrAlYTa coatings have been deposited onto an aircraft gas turbine engine blade using a LPPS unit equipped with a computerized robot. Optimal processing conditions, including spray parameters, the trajectory of the robot, and the synchronized movements between the torch and the blade, have been developed for superior coating properties. Transferred arc treatment, providing a preheating and a cleaning of the substrate surface, enhances the adherence of the coatings to the substrate. The resulting LPPS coatings show dense and uniform characteristics with ideal hardness, and good corrosion resistance to cycle oxidation.

Application of ultrasonic fatigue technology in very-high-cycle fatigue testing of aviation gas turbine engine blade materials:A review

The need for very-high-cycle fatigue(VHCF)testing up to 1010cycles of aviation gas turbine engine blade materials under combined mechanical loads and complex environments has encouraged the development of VHCF testing instrumentation and technology.This article begins with a comprehensive review of the existing available techniques that enable VHCF testing.Recent advances in ultrasonic fatigue testing(UFT)techniques are highlighted,containing their new capabilities and methods for single load,multiaxial load,variable amplitude fatigue,and combined cycle fatigue.New techniques for conducting UFT in high-temperature,humid environments,and corrosive environments are summarized.These developments in mechanical loading and environmental building techniques provide the possibility of laboratory construction for real service conditions of blade materials.New techniques that can be used for in situ monitoring of VHCF damage are summarized.Key issues in the UFT field are presented,and countermeasures are collated.Finally,the existing problems and future trends in the field are briefly described.

A novel model-based multivariable framework for aircraft gas turbine engine limit protection control

Control technologies are innovated to satisfy increasingly complicated control demands of gas turbine engines.In terms of limit protection control,a novel model-based multivariable limit protection control method,which is achieved by adaptive command reconstruction and multiplecontrol loop selection and switch logic,is proposed in this paper to address the problem of balancing smaller thrust loss and safe operations by comparing with widely-used Min-Max logic.Five different combination modes of control loops,which represent the online control loop of last time instant and that of current time instant,is analyzed.Different command reconstructions are designed for these modes,which is based on static gain conversion of amplitude beyond limits by using an onboard model.The double-prediction based control loop selection and switch logic is developed to choose a control loop appropriately by comparing converted amplitude beyond limits regardless of one or more parameters tending to exceed limits.The proposed method is implemented in a twin-spool turbofan engine to achieve limit protection with direct thrust control,and the loss of thrust is improved by about 30% in comparison with the loss of thrust caused by Min-Max logic when limit protection control is activated,which demonstrates the effectiveness of the proposed method.

Experimental study on impacts of fuel type on thermo-acoustic instability in a gas turbine model combustor

Effects of liquid fuel composition variations on characteristics of self-excited thermo-acoustic instabilities in a lean premixed,pre-vaporized gas turbine model combustor were experimentally studied.Test fuels included practical RP-3 jet fuel and its blending with iso-octane and n-dodecane,which were branched and linear alkanes respectively.Under the test conditions,dynamic pressure measurements indicated that the dominant instability frequency was highest for RP-3 flame,while RP-3/ndodecane flame exhibited the strongest instability strength.A further analysis showed that the instability frequency correlated well with the profiles of adiabatic flame temperature,and the strength of the instability highly depended on the ignition delay times of the fuels.Measurements of the flame structure and flow field with OH*chemiluminescence (CL) imaging and twodimensional particle image velocimetry (PIV) techniques indicated that changes in the fuel composition did not alter the unstable modes and general sequences of flame-flow structure oscillations.Further power spectra and proper orthogonal decomposition(POD) analysis suggested that axial oscillations along with precessing vortex core (PVC) induced helical motion predominated periodic flame structure and flow field oscillations.

A comprehensive review of thermally sprayed abradable sealing coatings:Focusing on abradability

Modern aero and stationary gas turbine engines have been designed with much higher compressor-pressure ratios and thrust-weight ratios than earlier models,and these ratios are strongly influenced by the hot-running clearances between the rotating and stationary components.The main benefit of reduction in the clearances is efficiency gains,resulting in lowered fuel consumption and polluting gas emissions,with ecological and economic advantages.However,at these undersized clearances,some rubbing interactions are unavoidable,which can be accommodated by applying Abradable Sealing Coatings(ASCs)on the stationary inner surface.This paper reviews the commercially available abradable materials for thermal spraying at various application positions and temperatures.Emphasis is placed on the abradability and wear mechanisms involved.In addition,considering the tendency of SiC/SiC ceramic matrix composites replacing superalloys as hot section components,the future prospect of ceramic abradables based on Environmental Barrier Coatings(EBCs)in turbine stages is summarized and a new concept of"self-degradable ceramics"based on the corrosive steam environment is proposed for the purpose of high-temperature fillerfree abradables.

3D convolutional selective autoencoder for instability detection in combustion systems

While analytical solutions of critical(phase)transitions in dynamical systems are abundant for simple nonlinear systems,such analysis remains intractable for real-life dynamical systems.A key example is thermoacoustic insta-bility in combustion,where prediction or early detection of the onset of instability is a hard technical challenge,which needs to be addressed to build safer and more energy-efficient gas turbine engines powering aerospace and energy industries.The instabilities arising in combustion chambers of engines are mathematically too complex to model.To address this issue in a data-driven manner instead,we propose a novel deep learning architecture called 3D convolutional selective autoencoder(3D-CSAE)to detect the evolution of self-excited oscillations using spatiotemporal data,i.e.,hi-speed videos taken from a swirl-stabilized combustor(laboratory surrogate of gas turbine engine combustor).3D-CSAE consists of filters to learn,in a hierarchical fashion,the complex visual and dynamic features related to combustion instability from the training videos(i.e.,two spatial dimensions for the image frames and the third dimension for time).We train the 3D-CSAE on frames of videos obtained from a limited set of operating conditions.We select the 3D-CSAE hyper-parameters that are effective for characterizing hierarchical and multiscale instability structure evolution by utilizing the dynamic information available in the video.The proposed model clearly shows performance improvement in detecting the precursors and the onset of instability.The machine learning-driven results are verified with physics-based off-line measures.Advanced active control mechanisms can directly leverage the proposed online detection capability of 3D-CSAE to mitigate the adverse effects of combustion instabilities on the engine operating under various stringent requirements and conditions.

Recent research progress on airbreathing aero-engine control algorithm

Airbreathing aero-engines are regarded as excellent propulsion devices from ground takeoff to hypersonic flight,and require control systems to ensure their efficient and safe operation.Therefore,the present paper aims to provide a summary report of recent research progress on airbreathing aero-engine control to help researchers working on this topic.First,five control problems of airbreathing aero-engines are classified:uncertainty problem,multiobjective and multivariable control,fault-tolerant control,distributed control system,and airframe/propulsion integrated control system.Subsequently,the research progress of aircraft gas turbine engine modelling,linear control,nonlinear control,and intelligent control is reviewed,and the advantages and disadvantages of various advanced control algorithms in aircraft gas turbine engines is discussed.Third,several typical hypersonic flight tests are investigated,and the modelling and control issues of dual-mode scramjet are examined.Fourth,modelling,mode transition control and thrust pinch control for turbine-based combined cycle engines are introduced.Followed,significant hypersonic airframe/propulsion integrated system control is analysed.Finally,the study provides specific control research topics that require attention on airbreathing aero-engines.

A double integral method for quantitative evaluation of influence on thin-walled casing response caused by bearing uncertainties

Bearings in a gas turbine engine are the key connecting components transmitting force and motion between rotors and thin-walled flexible casing.The bearing stiffness and damping of squeeze film damper(SFD)nearby bearings are easily affected by many factors,such as assembly process,load condition and temperature variation,resulting in uncertainties.The uncertainties may influence the response of the measuring point on the casing.Hence,it is difficult to carry out the fault diagnosis,whole machine balancing and other related works.In this paper,a double integral quantitative evaluation method is proposed to simultaneously analyze the influence of two uncertain dynamic coefficients on the response amplitude and phase of casing measuring points.Meanwhile,the coupling influence of stiffness and damping accompanied by dramatic changes with rotational speeds are essentially discussed.As an example,a typical engine bearing-casing system with complex dynamic characteristics is analyzed.The impact of uncertain dynamic coefficients on the unbalance response is quantitatively evaluated.

Reduced-dimensional MPC controller for direct thrust control

With the development of the aircraft gas turbine engine, a control system should be able to achieve effective thrust control to gain better operability. The main contribution of this paper is to develop a novel direct thrust control approach based on an improved model predictive control method through a strategy that reduces the dimension of control sequence. It can not only achieve normal direct thrust control tasks but also maximize the thrust level within the safe operation boundaries. Only the action of switching the objective functions is required to achieve the switch of these two thrust control modes while there is no modification to the control structure. Besides,a shorter control sequence is defined for multivariable control by updating only one control variable at every simulation time instant. Therefore, the time requirement for the solving process of the optimal control sequence is reduced. The proposed controller is implemented to a twin-spool engine.Simulations are conducted in the wide flight envelope, and results show that the average timeconsumption can be reduced up to 65% in comparison with the standard model predictive control,and the thrust can be increased significantly when maximum thrust mode is implemented by using engine limit margins.

Sequential ensemble optimization based on general surrogate model prediction variance and its application on engine acceleration schedule design

The Efficient Global Optimization(EGO)algorithm has been widely used in the numerical design optimization of engineering systems.However,the need for an uncertainty estimator limits the selection of a surrogate model.In this paper,a Sequential Ensemble Optimization(SEO)algorithm based on the ensemble model is proposed.In the proposed algorithm,there is no limitation on the selection of an individual surrogate model.Specifically,the SEO is built based on the EGO by extending the EGO algorithm so that it can be used in combination with the ensemble model.Also,a new uncertainty estimator for any surrogate model named the General Uncertainty Estimator(GUE)is proposed.The performance of the proposed SEO algorithm is verified by the simulations using ten well-known mathematical functions with varying dimensions.The results show that the proposed SEO algorithm performs better than the traditional EGO algorithm in terms of both the final optimization results and the convergence rate.Further,the proposed algorithm is applied to the global optimization control for turbo-fan engine acceleration schedule design.

Mathematical simulation of low cycle fatigue of high-loaded engine parts

The paper discusses main aspects of low cycle fatigue influence on the lifetime ofengine parts.The importance and history of the problem,the main experiments and the effectsimpacting the low cycle fatigue of structural materials are described.A hypothesis about theexistence of a thermomechanical surface of structural material,generalized to the case of acyclical loading was used to approximate the loops of cyclic nonisothermal elastoplasticdeformation curves.The cyclic deformation curve model is based on the following threeparameters:the elastic modulus during unloading,the Bauschinger effect and the conversionparameter of the nonlinear part of its first halfcycle.This model also accounts for theaccumulated plastic strain,as well as the testing temperature.The criterion of durability isformulated,based on the dependence between the number of halfcycles to failure and theaccumulated plastic deformation.Deformation theory of plasticity,generalized to the case ofcyclic deformation,in combination with the durability model and technology of"dying"elements is applied to the finite element analysis of low cycle fatigue of gas turbine engineparts.The results of calculations are demonstrated.