Staff publications (SATM)

Permanent URI for this collection

https://dspace.lib.cranfield.ac.uk/handle/1826/8738

Browse

Now showing 1 - 20 of 219

Open Access
Abrupt fault detection and isolation for gas turbine components based on a 1D convolutional neural network using time series data
(AIAA, 2020-08-17) Zhao, Junjie; Li, Yiguang
The FDI step identifies the presence of a fault, its level, type, and possible location. Gas turbine gas-path fault detection and isolation can improve the availability and economy of gas turbine components. Data-driven FDI methods are studied in this paper. Some notable gas turbine FDI challenges include: insensitivity to operating conditions, robust separation of faults, noisy sensor readings and missing data, reliable fault detection in time-varying conditions, and the influence of performance gradual deterioration. For conventional ML methods, the problem with handling time series data is its volume and the associated computational complexity; therefore, the available information must be appropriately compressed via the transformation of high-dimensional data into a low-dimensional feature space with minimal loss of class separability. In order to improve the detection and isolation sensitivity, this paper develops a method for FDI based on CNNs. Work in this paper includes: (1) Defining the problem and assembling a dataset. (2) Preparing data for training, validation and test: data generation, feature engineering, data pre-processing, data formatting. (3) Building up the model. (4) Training and validating the model (evaluation protocol). (5) Optimizing: a. deciding the model size. b. regularizing the model by getting more training data, reducing the capacity of the network, adding weight regularization or adding dropout. c. tuning hyperparameters. (6) Evaluation.
Open Access
Advancing fault diagnosis in aircraft landing gear: an innovative two-tier machine learning approach with intelligent sensor data management
(AIAA, 2024-01-04) Kadripathi, K. N.; Ignatyev, Dmitry; Tsourdos, Antonios; Perrusquía, Adolfo
Revolutionizing aircraft safety, this study unveils a pioneering two-tier machine learning model specifically designed for advanced fault diagnosis in aircraft landing gear systems. Addressing the critical gap in traditional diagnostic methods, our approach deftly navigates the challenges of sensor data anomalies, ensuring robust and accurate real-time health assessments. This innovation not only promises to enhance the reliability and safety of aviation but also sets a new benchmark in the application of intelligent machine-learning solutions in high-stakes environments. Our method is adept at identifying and compensating for data anomalies caused by faulty or uncalibrated sensors, ensuring uninterrupted health assessment. The model employs a simulation-based dataset reflecting complex hydraulic failures to train robust machine learning classifiers for fault detection. The primary tier focuses on fault classification, whereas the secondary tier corrects sensor data irregularities, leveraging redundant sensor inputs to bolster diagnostic precision. Such integration markedly improves classification accuracy, with empirical evidence showing an increase from 95.88% to 98.76% post-imputation. Our findings also underscore the importance of specific sensors—particularly temperature and pump speed—in evaluating the health of landing gear, advocating for their prioritized usage in monitoring systems. This approach promises to revolutionize maintenance protocols, reduce operational costs, and significantly enhance the safety measures within the aviation industry, promoting a more resilient and data-informed safety infrastructure.
Open Access
Adversarial proximal policy optimisation for robust reinforcement learning
(AIAA, 2024-01-04) Ince, Bilkan; Shin, Hyo-Sang; Tsourdos, Antonios
Robust reinforcement learning (RL) aims to develop algorithms that can effectively handle uncertainties and disturbances in the environment. Model-free methods play a crucial role in addressing these challenges by directly learning optimal policies without relying on a pre-existing model of the environment. This abstract provides an overview of model-free methods in robust RL, highlighting their key features, advantages, and recent advancements. Firstly, we discuss the fundamental concepts of RL and its challenges in uncertain environments. We then delve into model-free methods, which operate by interacting with the environment and collecting data to learn an optimal policy. These methods typically utilize value-based or policy-based approaches to estimate the optimal action-value function or the policy directly, respectively. To enhance robustness, model-free methods often incorporate techniques such as exploration-exploitation strategies, experience replay, and reward shaping. Exploration-exploitation strategies facilitate the exploration of uncertain regions of the environment, enabling the discovery of more robust policies. Experience replay helps improve sample efficiency by reusing past experiences, allowing the agent to learn from a diverse set of situations. Reward shaping techniques provide additional guidance to the RL agent, enabling it to focus on relevant features of the environment and mitigate potential uncertainties. In this paper, a robust reinforcement learning methodology is adapted utilising a novel Adversarial Proximal Policy Optimisation (A-PPO) method integrating an Adaptive KL penalty PPO. Comparison is made with DQN, DDQN and a conventional PPO algorithm.
Open Access
Aero gas turbine flight performance estimation using engine gas path measurements
(AIAA, 2015-01-05) Li, Yi-Guang
Mature gas turbine performance simulation technology has been developed in the past decades and, therefore, gas turbine performance at different ambient and operating conditions can be well predicted if good thermodynamic performance software and necessary engine performance information are available. However, the performance of gas turbine engines of the same fleet may be slightly different from engine to engine due to manufacturing and assembly tolerance and may change over time due to engine degradation. Therefore, it is necessary to monitor and track important performance parameters of gas turbine engines, particularly those that cannot be directly measured, to ensure safe operation of the engines. For that reason, a novel gas turbine performance estimation method using engine gas path measurements has been developed to predict and track engine performance parameters at different ambient, flight, degraded, and part-load operating conditions. The method is based on the influence coefficient matrix of thermodynamic performance parameters of gas turbine engines and the Newton Contrary to the conventional gas turbine off-design performance predictions where component characteristic maps are essential, it has the advantage that no component characteristic maps are required for the predictions and, therefore, it is relatively simple thermodynamically, fast in calculation, and desirable in engineering applications. It is able to make important invisible performance parameters visible to gas turbine users, which is a useful complement to current engine condition monitoring techniques. The developed method was applied to the performance prediction of a model gas turbine engine similar to EJ200 low-bypass turbofan engine running at different altitudes, Mach numbers, and part load, with and without degradation, by using simulated gas path measurements to test the effectiveness of the method. The results show that the method is able to predict the engine performance with good accuracy without the consideration of measurement noise and with slightly lower accuracy when measurement noise is included. It takes about 30 s for a typical prediction point, which is suitable for offline performance tracking and condition monitoring. Theoretically, the method can be applied to the performance estimation of any types of gas turbine engines.–Raphson mathematical algorithm.
Open Access
Aeroacoustic analysis of a closely installed chevron nozzle jet using the high-order discontinuous Galerkin method
(AIAA, 2023-06-08) Lindblad, Daniel; Sherwin, Spencer J.; Cantwell, Chris D.; Lawrence, Jack; Proenca, Anderson; Moragues Ginard, Margarida
In this paper, we use Large Eddy Simulations (LES) in combination with the Ffowcs Williams - Hawkings method to study the influence of chevrons on the flow field as well as the noise produced by a closely installed M = 0.6 jet. The LES simulations are performed with the spectral/hp element framework Nektar++. Nektar++ uses the high-order discontinuous Galerkin method and an implicit scheme based on the matrix-free Newton-GMRES method to discretize the unfiltered Navier-Stokes equations in space and time, respectively. The far-field noise is computed using Antares. Antares solves the Ffowcs Williams - Hawkings equation for a permeable integration surface in the time-domain using a source-time dominant algorithm. The aerodynamic results show good agreement with experimental data obtained in the Doak Laboratory Flight Jet Rig, located at the University of Southampton. Some discrepancies are observed in terms of the far-field noise levels, especially for higher polar observer angles relative to the downstream jet axis. In terms of noise reduction potential, the simulations predict that the chevrons reduce the OASPL by 1dB compared to an installed round nozzle for all observers located on the unshielded side of the wing. This should be compared to the experiments, which predict a 1.5dB noise reduction for the same chevron nozzle.
Open Access
Aerodynamic analysis of civil aeroengine exhaust systems using computational fluid dynamics
(AIAA, 2018-06-25) Otter, John J.; Goulos, Ioannis; MacManus, David G.; Slaby, Michal
As the specific thrust of civil aeroengines reduces, the aerodynamic performance of the exhaust system will become of paramount importance in the drive to reduce engine fuel burn. This paper presents an aerodynamic analysis of civil aeroengine exhaust systems through the use of Reynolds-averaged Navier–Stokes computational fluid dynamics. Two different numerical approaches are implemented, and the numerical predictions are compared to measured data from an experimental high-bypass-ratio separate-jet exhaust system. Over a fan nozzle pressure ratio range from 1.4 to 2.8, a comparison is drawn between values of the thrust coefficient calculated numerically and those obtained from experimental measurements. In addition, the effects of the freestream Mach number and extraction ratio on the aerodynamic behavior of the exhaust system are quantified and correlated to fundamental aerodynamic parameters.
Open Access
Aerodynamic analysis of Saab 340B aircraft with data fusion implementation
(AIAA, 2024-01-04) Sahin, Kadir; Gomec, Fazil; Millidere, Murat; Whidborne, James
This paper conducts an aerodynamic analysis of the Saab 340B passenger aircraft, employing AVL and as numerical methods and DATCOM as a collection of engineering methods and empirical data that provide a set of aerodynamic coefficients for assessment. The investigation is focused on examining the longitudinal aerodynamic behavior of aircraft, specifically considering the impacts of flaps and elevators at varying angles of attack. The outcomes from the clean configuration are compared with the results obtained from computational fluid dynamics studies found in existing literature. While the results may not precisely match, they capture the general trends in the behavior of the aircraft. This aligns with the expected outcomes from preliminary methods like AVL and DATCOM. Data fusion techniques are strategically employed to integrate insights from these diverse sources, enhancing the overall accuracy and reliability of the aerodynamic assessment. The research aims to provide a comprehensive understanding of the clean configuration's aerodynamic performance, contributing significantly to the advancement of aviation technology.
Open Access
Aerodynamic effects of propulsion integration for high bypass ratio engines
(AIAA, 2017-05-26) Stankowski, Tomasz; MacManus, David G.; Robinson, Matthew; Sheaf, Christopher
This work describes the assessment of the effect of engine installation parameters such as engine position, size, and power setting on the performance of a typical 300-seater aircraft at cruise condition. Two engines with very high bypass ratio and with different fan diameters and specific thrusts are initially simulated in isolation to determine the thrust and drag forces for an isolated configuration. The two engines are then assessed in an engine–airframe configuration to determine the sensitivity of the overall installation penalty to the vertical and axial engine location. The breakdown of the interference force is investigated to determine the aerodynamic origins of beneficial or penalizing forces. To complete the cruise study, a range of engine power settings is considered to determine the installation penalty at different phases of cruise. This work concludes with the preliminary assessment of cruise fuel burn for two engines. For the baseline engine, across the range of installed positions, the resultant thrust requirement varies by 1.7% of standard net thrust. The larger engine is less sensitive with a variation of 1.3%. For an assessment over a 10,000 km cruise flight, the overall effect of the lower specific thrust engine shows that the cycle benefits of −5.8% −5.8% in specific fuel consumption are supplemented by a relatively beneficial aerodynamic installation effect but offset by the additional weight to give a −4.8% −4.8% fuel-burn reduction.
Open Access
Aerodynamic interference for aero-engine installations
(AIAA, 2016-01-02) Stankowski, Tomasz P.; MacManus, David G.; Sheaf, Christopher; Grech, Nicholas
Open Access
Aerodynamics of aero-engine installation
(AIAA, 2016-01-02) Stankowski, Tomasz P.; MacManus, David G.; Sheaf, Christopher; Grech, Nicholas
Small internal combustion engines, particularly those ranging in power from 1 kW to 10 kW, propel many remotely piloted aircraft (RPA) platforms that play an increasingly significant role in the Department of Defense. Efficiency of these engines is low compared to conventional scale engines and thermal losses are a significant contributor to total energy loss. Existing thermal energy loss models are based on data from much larger engines. Whether these loss models scale to the engine size class of interest, however, has yet to be established. The Small Engine Research Bench (SERB) was used to measure crank angle resolved gas temperature inside the combustion chamber of a small internal combustion engine (ICE). A 55 cc, two stroke, spark-ignition ICE was selected for this study. The engine was modified for optical analysis using sapphire rods 1.6 mm in diameter on opposite sides of the combustion chamber. The engine modification was found to have no measurable impact on indicated mean effective pressure or heat rejection through the cylinder. FTIR absorption thermometry was used to collect mid-infrared absorption spectra. The FTIR was allowed to scan continuously while simultaneously recording the scanning mirror position and crank angle associated with each data point, then data was re-sorted by crank angle. Measured spectra were compared with lines generated using CDSD-4000 and HITEMP line list databases. The line of best fit corresponded to the mean gas temperature through the combustion chamber. In this way temperature was determined as a function of crank angle for three operating conditions: 4,300, 6,000, and 7,500 revolutions per minute, all at wide open throttle. High cycle-to-cycle variation in the regions of combustion and gas exchange degraded temperature measurements at the affected crank angles. Future research will attempt to improve signal to noise in these measurements.
Open Access
Aeroelastic scaling for flexible high aspect ratio wings
(AIAA, 2019-12-31) Yusuf, Sezsy; Pontillo, Alessandro; Weber, Simone; Hayes, David; Lone, Mudassir
This paper provides an overview of the work conducted as part of the Cranfield BEAmReduction and Dynamic Scaling (BeaRDS ) programme, which aims to develop a methodologyfor designing, manufacturing and testing of a dynamically scaled High Aspect Ratio (HAR)Wing inside Cranfield 8’x6’ wind tunnel. The aim of this paper is to develop a methodologythat adopts scaling laws to allow experimental testing of a conceptual flexible-wing planformas part of the design process. Based on the Buckinghamπtheorem, a set of scaling lawsare determined that enable the relationship between a full-scale and sub-scale model. Thedynamically sub-scaled model is manufactured as a combination of spar, skin, and addedmass representing the stiffness, aerodynamic profile, and aeroelastic behaviour respectively.The spar was manufactured as a cross-sectional shape using Aluminium material, while theskin was manufactured using PolyJet technology. Compromises due to the manufacturingprocess are outlined and lessons learned during the development of the sub-scaled model arehighlighted.
Open Access
AFJPDA: a multiclass multi-object tracking with appearance feature-aided joint probabilistic data association
(AIAA, 2024-01-02) Kim, Sukkeun; Petrunin, Ivan; Shin, Hyo-Sang
This study addresses a multiclass multi-object tracking problem in consideration of clutters in the environment. To alleviate issues with clutters, we propose the appearance feature-aided joint probabilistic data association filter. We also implemented simple adaptive gating logic for the computational efficiency and track maintenance logic, which can save the lost track for re-association after occlusion or missed detection. The performance of the proposed algorithm was evaluated against a state-of-the-art multi-object tracking algorithm using both multiclass multi-object simulation and real-world aerial images. The evaluation results indicate significant performance improvement of the proposed method against the benchmark state-of-the-art algorithm, especially in terms of reduction in identity switches and fragmentation.
Open Access
AI for real-time tolerance to critical flight data errors in large aircraft
(AIAA, 2023-06-08) Koopman, Cynthia; Zammit-Mangion, David
The environment in the cockpit of large transport aircraft is currently highly complex due to an increasing amount of automation systems. This complexity can cause pilots to become less aware of how all systems work and interact. It becomes a severe issue when sensor or data failures occur, as such failures can contribute to a situation in which it is difficult for a pilot to assess what actually is happening and, possibly, where the fault originated from and how to resolve the problem. Erroneous sensor information is known to cause automation to fail or malfunction and there are several instances where such errors led to fatal accidents. This paper presents a method, based on artificial intelligence, for detecting and identifying incorrect critical flight control data in real-time. The aim of this method is to help the pilot assess the state of the aircraft and reduce the risk of confusion due to automation. A novel combination of Reinforcement Learning and an auxiliary denoising autoencoder is proposed to identify where the failures are occurring and to provide command inputs to the aircraft’s flight control and guidance systems, allowing the aircraft to perform the correct manoeuvre to counter the failure and/or to avoid or recover from flight upsets and loss of control. Tests in nominal as well as stall conditions with a partially blocked Pitot tube were conducted. These tests show that the proposed combination of Machine Learning methods creates a system to accurately detect failures (2.5s average detection time), reconstruct input data (RMSE < 6 ft/s for airspeed), and provide stable directions for the flight controls. Due to the specifically designed architecture and training schedule it is possible for the proposed system to achieve this level of performance using only a single neural network. To conclude, a comparison with the performance of the system trained without the auxiliary denoising autoencoder was made to highlight the significant advantages of the proposed architecture for learning meaningful neural connections and how this relates to creating systems with AI to improve situational awareness for pilots and execute appropriate automatic manoeuvres to successfully counter the effect of sensor failures.
Open Access
AI-based multifidelity surrogate models to develop next generation modular UCAVs
(AIAA, 2023-01-19) Karali, Hasan; Inalhan, Gokhan; Tsourdos, Antonios
The next generation low-cost modular unmanned combat aerial vehicles (UCAVs) provide the opportunity to implement innovative solutions to complex tasks, while also bringing new challenges in design, production, and certification subjects. Solving these problems with tools that provide fast modeling in line with the digital twin concept is possible. In this work, we develop an artificial intelligence (AI) based multifidelity surrogate model to determine performance parameters of innovative modular UCAVs. First, we develop a data generation algorithm that includes a high-fidelity model based on computational fluid dynamics methods and a low-fidelity model based on computational aerodynamic approaches. In the next step, a new transfer learning-based surrogate model is generated using multifidelity data. Thanks to this approach, the developed AI model more accurately predicted the flow conditions that were missing in the high-fidelity data with the data obtained from the low-order model. The performance of the proposed AI-based surrogate model is to be investigated in terms of accuracy, robustness, and computational cost using a generic modular UCAV configuration.
Open Access
AI-driven multidisciplinary conceptual design of unmanned aerial vehicles
(AIAA, 2024-01-04) Karali, Hasan; Inalhan, Gokhan; Tsourdos, Antonios
This paper presents a multidisciplinary conceptual design framework for unmanned aerial vehicles based on artificial intelligence-driven analysis models. This approach leverages AI- driven analysis models that include aerodynamics, structural mass, and radar cross-section predictions to bring quantitative data to the initial design stage, enabling the selection of the most appropriate configuration from various optimized concept designs. Due to the design optimization cycle, the initial dimensions of key components such as the wing, tail, and fuselage are provided more accurately for later design activities. Simultaneously, the generated structure enables more suitable design point selection through the feedback loop within the design iteration. Therefore, in addition to reducing design costs, this approach also offers a substantial time advantage in the overall design process.
Open Access
Airliner conceptual designs for the application of alternative fuels
(AIAA, 2023-06-08) Chan, Joseph; Sun, Yicheng; Smith, Howard
This paper investigates the application of alternative fuels in conventional tube-and-wing aircraft configuration. Potential alternative fuel choices include biofuel, liquid hydrogen, liquefied natural gas, ammonia and methanol. A comprehensive mission range starting from 2000 nautical miles to 8500 nautical miles is explored, designed and analysed using GENUS, an in-house built multidisciplinary design analysis and optimization platform. It is expected to see a progressive change for each fuel type and capacity combination over different mission ranges. The gradient of these trends should be different which potentially demonstrates the additional benefits of using alternative fuels. The aim of this study is 1) to determine how different fuel properties will affect the designs and performance of aircraft, 2) to quantify the reduction of GHG emission by utilising alternative fuel. The results show cryogenic fuel, which required fuselage fuel tanks and external fuel tanks, will noticeably increase energy consumption due to drag and weight penalties. Meanwhile, the use of hydrogen and biofuel can significantly reduce life cycle GHG emissions if only the fuel is produced in sustainable pathways.
Open Access
Analysis of triangular cross-section slender bodies in supersonic regime using RANS simulations
(AIAA, 2023-06-08) Bourny, Quentin; Proenca, Anderson; Di Pasquale, Davide; Prince, Simon A.
This paper presents an investigation on the ability of RANS simulations to capture the aerodynamic forces and the flow topology of triangular cross-section slender bodies. At Cranfield University’s Transonic and Supersonic Wind Tunnel, force measurements and Schlieren images were obtained at zero incidence, Mach number equals to 2.5, and Reynolds number of 2.38 · 10^5 (based on section width). Tests were performed for three bodies of different nose geometries, but constant nose fineness ratio of 1.732. Tests were compared with RANS simulations for three turbulence models: Spalart-Allmaras, k− epsilon Realizable and k − omega SST using the ANSYS Computational Fluid Dynamics software Fluent. In addition, simulations for a configuration presented in the literature which investigated several angles of attack were also conducted. At Mach 2.5, the normal force was predicted accurately by all turbulence models. The axial force, however, was clearly predicted more accurately with the k − epsilon Realizable model. At the other hand, this turbulence model showed inferior ability to capture the flow features, particularly the leeside vortices. Spalart-Allmaras and k − omega SST gave similar results.
Open Access
Analysis of visualization systems in flight simulators
(AIAA, 2023-06-08) Barrio, Luis D.; Korek, Wojciech; Millidere, Murat; Whidborne, James F.
This paper details an analysis of different visualization systems for use in an academic flight simulator, Future Systems Simulator (FSS). First, an overview of off-the-shelf flight simulators is done, detailing the primary features of flight simulators such as Flight Gear, Prepar3D, X-Plane, and Microsoft Flight Simulator (2020). Then, the current setup of the FSS is presented (which uses FlightGear), followed by the process of introducing X-Plane as a scenery-generation tool. To conduct a comparative analysis between FlightGear and X-Plane visual systems, a total of twelve participants with varying levels of experience were invited to participate in the study. The participants performed flight trials in a simple landing scenario at Heathrow Airport. Additionally, the more complex approach at London City Airport was performed with a group of only four highly experienced participants. Participants then gave their feedback and completed a questionnaire. The data from their attempts were recorded for qualitative and quantitative comparison. The results were analyzed to determine which of the two visual systems could be used in the FSS moving forward.
Open Access
Analysis of wake surfing benefits using a fast unsteady vortex lattice method
(AIAA, 2019-01-08) Fleischmann, Dominique; Lone, Mudassir
The computer simulation framework Flexit is used to analyse the fuel economy benefit of aircraft wake surfing. Wake surfing involves multiple aircraft flying in close formation during cruise conditions to reduce overall induced drag and improve overall fuel efficiency. The aircraft fly in echelon such that the kinetic energy lost in vortices generated by the lead aircraft can be partially recovered by the following aircraft flying in regions of the wake where induced velocities have an upwardly directed vertical component. We build on recent theoretical and flight test work by developing a medium fidelity methodology using Flexit for predicting potential performance benefits of wake surfing. We present results from a specific systematic parametric study that corresponds to a series of recent flight tests with two C-17 transport aircraft to demonstrate the methodology and predict the fuel savings that can be obtained by different arrangements of aircraft in a wake surfing formation. The predictions are compared with the flight test data and the trends observed in our simulations agree with the trends of the full scale tests.
Open Access
Application of a semi-empirical method to model subsonic vortex lift over sharp leading-edge delta wings
(AIAA, 2023-01-19) Huynh, Daniel; Di Pasquale, Davide; Prince, Simon A.; Ahuja, Vivek
A semi-empirical method is applied as a complement to the FlightStream solver, to more accurately model subsonic vortex lift over a sharp leading-edge delta wing. The method, based on the prediction of flow patterns and the application of the Polhamus method, is particularly well-adapted to a preliminary aerodynamic design phase. Within a few minutes, it can accurately predict the aerodynamic forces generated by the flow over a delta wing, which are strongly affected by the presence of a leading-edge vortex. This study presents a detailed analysis where computed results are compared against experimental data. Those were obtained from a test case of a 65° subsonic delta wing experiment (case 1), along with a sensitivity analysis against sweep angle and aspect ratio where multiple subsonic delta wings were tested (case 2). A good agreement is observed between computed data and experimental results, within pre-stall, before the vortex bursts. Analysed results demonstrate the validity of the method, for multiple wing configurations associated with different flow conditions.