PhD, EngD and MSc by research theses (SATM)

Permanent URI for this collection

https://dspace.lib.cranfield.ac.uk/handle/1826/8739

Browse

Now showing 1 - 10 of 10

Open Access
Adhesive joint geometry variation in non-rigid aircraft structures
(Cranfield University, 2019-11) Coladas Mato, Pablo; Webb, Phil; Xu, Yigeng
Adhesive bonding is a proven alternative to mechanical fasteners for structural assembly, offering lighter and thus more fuel efficient aircraft and cost-effective manufacturing processes. The effective application of bonded structural assemblies is however limited by the tight fit-up requirement, which is with tolerance ranges of hundreds of microns; this can be a challenge for the industry to meet considering the variability of current part manufacturing methods and the conservative nature of the conventional tolerance stack-up analysis method. Such a (perceived) limitation can discourage effective exploitation of bonding technologies, or lead to development of overengineered solutions for assurance. This work addresses such challenge by presenting an enhanced bondline thickness variation analysis accounting for part deflection of a bonded skinstringer assembly representing a typical non-rigid airframe structure. A semianalytical model accounting for unilateral contact and simplified 1D adhesive flow has been developed to predict bondline thickness variation of the assembly given the adherends’ mechanical properties, adhesive rheological properties, and external assembly forces or boundary conditions. A spectral-analysis method for assembly force requirement estimation has also been tested. The bondline dimensions of several representative test articles have been interrogated, including a reconfigurable test assembly designed specifically to test the input conditions that affect bondline geometry variation. It has been demonstrated that the part deflections need to be accounted for regarding the fit-up requirement of bonded non-rigid structural assembly. The semi-analytical model has been found to more reliable and realistic prediction of bondline thickness when compared to a rigid tolerance stack-up. The analysis method presented can be a major technology enabler for faster, more economical development of the aircraft of the future, as well as of any analogue structures with high aspect ratios where weight savings and fatigue performance may be core objectives.
Open Access
Advanced numerical methods for dissipative and non-dissipative relativistic hydrodynamics
(Cranfield University, 2020-05) Townsend, Jamie F.; Könözsy, László Z. ; Jenkins, Karl W.
High-energy physical phenomena such as astrophysical events and heavy-ion collisions contain a hydrodynamic aspect in which a branch of fluid dynamics called relativistic hydrodynamics (RHD) is required for its mathematical description. The resulting equations must be, more often than not, solved numerically for scientists to ascertain useful information regarding the fluid system in question. This thesis describes and presents a twodimensional computational fluid dynamics (CFD) solver for dissipative and non-dissipative relativistic hydrodynamics, i.e. in the presence and absence of physically resolved viscosity and heat conduction. The solver is based on a finite volume, Godunov-type, HighResolution Shock-Capturing (HRSC) framework, containing a plethora of numerical implementations such as high-order Weighted-Essentially Non-Oscillatory (WENO) spatial reconstruction, approximate Riemann solvers and a third-order Total Variation Diminishing (TVD) Runge–Kutta method. The base numerical solver for the solution of non-dissipative RHD is extensively tested using a series of one-dimensional test cases, namely, a smooth flow problem and shock-tube configurations as well as the two-dimensional vortex sheet and Riemann problem test cases. For the case of non-dissipative relativistic hydrodynamics the relativistic CFD solver is found to perform well in terms of the orders of accuracy achieved and its ability to resolve shock wave patterns. Numerical pathologies have been identified when the relativistic HLLC Riemann solver is used in multi-dimensions for problems exhibiting strong shock waves. This is attributed to the so-called Carbuncle problem which is shown to occur because of pressure differencing within the process of restoring the missing contact discontinuity of its predecessor, the HLL Riemann solver. To avoid this numerical pathology and improve the robustness of numerical solutions that make use of the HLLC Riemann solver, the development of a rotated-hybrid Riemann solver arising from the hybridisation of the HLL and HLLC (or Rusanov and HLLC) approximate Riemann solvers is presented. A standalone application of the HLLC Riemann solver can produce spurious numerical artefacts when it is employed in conjunction with Godunov-type high-order methods in the presence of discontinuities. It has been found that a rotated-hybrid Riemann solver with the proposed HLL/HLLC (Rusanov/HLLC) scheme could overcome the difficulty of the spurious numerical artefacts and presents a robust solution for the Carbuncle problem. The proposed rotated-hybrid Riemann solver provides sufficient numerical dissipation to capture the behaviour of strong shock waves for relativistic hydrodynamics. Therefore, focus is placed on two benchmark test cases (odd-even decoupling and double-Mach reflection problems) and the investigation of two astrophysical phenomena, the relativistic Richtmyer– Meshkov instability and the propagation of a relativistic jet. In all presented test cases, the Carbuncle problem is shown to be eliminated by employing the proposed rotated-hybrid Riemann solver. This strategy is problem-independent, straightforward to implement and provides a consistent robust numerical solution when combined with Godunov-type highorder schemes for relativistic hydrodynamics...[cont.]
Open Access
Aircraft engine transient performance modelling with heat soakage effects
(Cranfield University, 2019-11) Li, Zhuojun; Li, Yiguang
Transient performance design and assessment is a very crucial step of aircraft engine development, especially for acceleration and deceleration process. Normally, the assessment of transient performance stability would be done during the detained design stage while component design parameters are available. As a result, design iterations might be necessary and costly if the transient performance assessment is not satisfactory. To make engine design more cost and time efficiently, it has become more and more important to assess the transient performance stability at conceptual and preliminary design stage with the inclusion of key impact factors such as fuel control schedule, rotor dynamics, volume dynamics and heat soakage. However, due to the lack of detailed engine structural and geometrical information at the initial design stage, such transient performance simulation and assessment may have to ignore heat soakage effects. Therefore, in this paper, a novel generically simplified heat soakage and tip clearance model for three major gas path components of gas turbine engines including compressors, turbines and combustors and has been developed to support more realistic transient performance simulation of gas turbine engines at conceptual and preliminary design stage. Such heat soakage model including heat transfer and tip clearance only requires thermodynamic design parameters as input, which is normally available during such design stages. This generic heat soakage method has been applied to two engine models to test its effectiveness through an in-house developed performance code. The case study of heat-soakage effects could demonstrate that results are promising and the simplified heat soakage model is satisfactory.
Open Access
Development of model free flight control system using deep deterministic policy gradient (DDPG)
(Cranfield University, 2019-09) Budiarti, Dewi H.; Tsourdos, Antonios; Shin, Hyo-Sang
Developing a flight control system for a complete 6 degree-of-freedom for an air vehicle remains a huge task that requires time and effort to gather all the necessary data. This thesis proposes the use of reinforcement learning to develop a policy for a flight control system of an air vehicle. This method is designed to be independent of a model but it does require a set of samples for the reinforcement learning agent to learn from. A novel reinforcement learning method called Deep Deterministic Policy Gradient (DDPG) is applied to counter the problem with large and continuous space in a flight control. However, applying the DDPG for multiple action is often difficult. Too many possibilities can hinder the reinforcement learning agent from converging its learning process. This thesis proposes a learning strategy that helps shape the way the learning agent learns with multiple actions. It also shows that the final policy for flight control can be extracted and applied immediately for a flight control system.
Open Access
High aspect ratio wings on commercial aircraft: a numerical and experimental approach
(Cranfield University, 2019-12-20) Pontillo, Alessandro; Lone, Mudassir M.; Whidborne, James F.
The aim of this work is to assess the aeroelastic response to gust of a flexible high aspect ratio wing (HARW) single-aisle commercial aircraft and to design a viable open-loop Gust Load Alleviation (GLA) system. Aeroservoelastic assessment was carried out by adopting the low-fidelity Cranfield Accelerated Aircraft Load Model (CA2LM ) aeroelastic framework. Wind tunnel testing of two flexible wing models was carried out to assess the limitations of low-fidelity numerical frameworks in modelling highly flexible structures. The numerical work firstly focused on upgrading the CA2LM framework by including the non-linear aerodynamic effects of spoiler deflection into the low-fidelity model. The novel method was able to locally change the wing lift distribution evaluated with strip theory by combining ESDU 14004 experimental data with the numerical estimation. Finally, the aeroelastic response of the High Aspect Ratio Technology ENabler (HARTEN) concept aircraft to gust input was carried out for a single flight condition (h=26000 ft and v=200 m/s) and for two different structural configurations: rigid wing and flexible wing structure. Tuned discrete gust analysis, as specified in CS-25, was adopted in this analysis. Results showed that tuned gust is able to excite flexible wing dynamics along with the rigid-body dynamics, having a detrimental impact on aircraft performance. Finally, an open-loop GLA system was designed to alleviate Wing Root Bending Moment (WRBM) increment due to gust load. The GLA deflected spoilers and ailerons for a fixed amount of time (hold time) once a specific vertical load factor was crossed. An optimization algorithm was used to optimize parameters such as control surfaces deflection, hold time and load factor threshold. Several configurations of the GLA were evaluated. The optimal GLA configuration was able to alleviate WRBM from a minimum of 2.4% to a maximum of 8.1% with respect to the non-alleviated scenario. Two wind tunnel models were built with the common spar and skin configuration, while a novel approach for the skin manufacturing was introduced: the skin was 3D printed with PolyJet technology which allowed to provide a continuous aerodynamic shape removing the typical gaps necessary for flexible models to allow wing bending, limiting the impact of the skin to less than 12.5% of the overall model stiffness. The first model was tested in the Cranfield Weybridge wind tunnel at 27 m/s (Re = 3.5e5) and α = 6 ◦ . The model span was 0.840 m and Aspect Ratio AR = 12. The model was successfully tested to prove the ability of the skin to retain the aerodynamic shape and sustain the load under large deformation, reaching a max wingtip displacement of 32% of the model span. The second model was tested in the Cranfield 8x6 ft wind tunnel in the speed range of 20 m/s to 40 m/s (3.1e5 < Re < 6.2e5) at −2 ◦ < α < 8 ◦ . The model span was 1.5 m and AR = 18.8. The main result showed that in the most severe aerodynamic load scenario (v = 40 m/s and α = 8 ◦ ), the spanwise force coefficient accounted for 10% of the wing overall CL and was 2.5 times higher than CD . The overall damping was also estimated for different velocities at α = 6 ◦ , reaching a maximum of 26.9% at 35 m/s and a minimum of 17.8% at 20 m/s, with aerodynamic damping accounting for a minimum of 61% to a maximum of 74% of the overall damping. Maximum displacement of the wing tip was 13.7% of the model span (0.21 m). In both tests a low-cost acquisition system built with off-the-shelf components was used. The system was based on Raspberry Pi board able to acquire accelerations and rotations from four MPU6050 IMU boards, with the main benefit being the small size of the sensors, which were able to fit within tiny volumes typical of HARW wind tunnel models.
Open Access
Intelligent-based hybrid-electric propulsion system for aero vehicle
(Cranfield University, 2020-03) Wang, Siqi; Tsourdos, Antonios; Economou, John T.
To address the sustainability challenges for air transport, electrified aviation delivers promising benefits to the whole air transportation system. Focusing on reducing environmental impact and raising competitiveness, this thesis presents a research regarding the Distributed Series Hybrid-electric Propulsion System for aero vehicles, which involves study fields of system configuration design, component sizing and energy management strategies. Based on the state-of-art of hybrid-electric aircraft and hybrid-electric propulsion systems, the study firstly improved the conventional series hybrid configuration by adopting distributed propulsion technology and more electric aircraft concept. These improvements can compensate for the drawbacks caused by the conventional series hybrid layout, so that the new designed propulsion system has the potential to reduce system weight and increase fuel economy. After that, a comprehensive sizing method was particularly designed for the proposed system. The engine, as the primary power source, was firstly selected via the battery parametrisation criteria. Then, other components were selected according to a proposed sizing flowchart by using the genetic algorithm. System performance can also be demonstrated during the sizing process. Finally, three different control methods had been applied to manage energy flows. The first supervisory controller is a deterministic rule-based controller, which was designed based on human experiences and can reduce 12% fuel consumption. The second is a battery-friendly fuzzy controller. It was particularly designed to improve the battery operating environment and can simultaneously achieve a 5% improvement on fuel economy compared to the rule-based. The third controller applied model predictive control algorithm, which can further improve the fuel efficiency by 4% and reveal the relationship between the fuel consumption and emissions.
Open Access
Inverse design of transonic/supersonic aerofoils based on deep neural networks
(Cranfield University, 2019-12) Feria Alanis, Aaron; Antoniadis, Antonios
Transonic and supersonic aerofoil inverse design for different flight conditions is carried out using Deep Neural Networks (DNN). DNN are combined with a comprehensive and complete database of aerodynamic data and aerofoil geometry parameters to form the pillars of a surrogate inverse aerodynamic design tool. The framework of this research starts with the aerofoil parameterisation. The Class/Shape Transformation functions (CST) was selected for the parameterisation process due to its high accuracy and flexibility when describing complex shapes. An automated mesh technique is created and implemented to discretise the flow domain. The aerodynamic computations are performed for 395 aerofoils. Spatial discretisation is accomplished with the Jameson-Schmidt-Turkel (JST) scheme and convergence is reached by the backward Euler implicit numerical scheme. Data are collected and managed with the CST parameters for all aerofoils and their respective aerodynamic characteristics from the CFD solver. The Deep Neural Network is then trained, validated using cross-validation and evaluated against CFD data. An extensive investigation of the effect from different DNN configurations takes place in this research. Within this thesis, different case studies are presented for different numbers of design objectives. For the inverse design process the NACA 66-206 aerofoil was selected as the baseline aerofoil, to reduce the aerodynamic drag coefficient while maintaining or improving the lift coefficient, to obtain a superior lift/drag ratio compared with the baseline aerofoil. The framework of this thesis have proved to output aerofoil designs with an improved lift/drag ratio in comparison with the baseline aerofoil.
Open Access
Methodology for avionics integration optimisation
(Cranfield University, 2020-10) Radaei, Mohammad; Jia, Huamin; Lawson, C. P.
Every state-of-art aircraft has a complex distributed systems of avionics Line Replaceable Units/Modules (LRUs/LRMs), networked by several data buses. These LRUs are becoming more complex because of the increasing number of new avionics functions need to be integrated in an avionics LRU. The evolution of avionics data buses and architectures have moved from distributed analogue and federated architecture to digital Integrated Modular Avionics (IMA). IMA architecture allows suppliers to develop their own LRUs/LRMs capable of specific features that can then be offered to Original Equipment Manufacturers (OEMs) as Commercial-Off-The-Shelf (COTS) products. In the meantime, the aerospace industry has been investigating new solutions to develop smaller, lighter and more capable avionics LRUs to be integrated into avionics architecture. Moreover, the complexity of the overall avionics architecture and its impact on cable length, weight, power consumption, reliability and maintainability of avionics systems encouraged manufacturers to incorporate efficient avionics architectures in their aircraft design process. However, manual design cannot concurrently fulfil the complexity and interconnectivity of system requirements and optimality. Thus, developing computer-aided design (CAD), Model Based System Engineering (MBSE) tools and mathematical modelling for optimisation of IMA architecture has become an active research area in avionics systems integration. In this thesis, a general method and tool are developed for optimisation of avionics architecture and improving its operational capability. The tool has three main parts including a database of avionics LRUs, mathematical modelling of the architectures and optimisation algorithms. The developed avionics database includes avionics LRUs with their technical specifications and operational capabilities for each avionics function. A MCDM method, SAW, is used to quantify and rank each avionics LRU’s operational capability. Based on the existing avionics LRUs in the database and aircraft level avionics requirements two avionics architectures are proposed i.e. AFCS architecture (SSA) and avionics architecture (LSA). The proposed avionics architectures are then modelled using mathematical programming. Further, the allocation of avionics LRUs to avionics architecture and mapping the avionics LRUs to their installation locations are defined as an assignment problem in Integer Programming (IP) format. The defined avionics architecture optimisation problem is to optimise avionics architecture in terms of mass, volume, power consumption, MTBF and operational capability. The problems are solved as both single-objective and multi-objective optimisation using the branch-and-bound algorithm, weighted sum method and Particle Swarm Optimisation (PSO) algorithm. Finally, the tool provides a semi-automatic optimisation of avionics architecture. This helps avionics system architects to investigate and evaluate various architectures in the early stage of design from an LRU perspective. It can also be used to upgrade a legacy avionics architecture.
Open Access
Power management for energy harvesting
(Cranfield University, 2020-05) Alsader, Moner; Petrunin, Ivan; Tsourdos, Antonios
The use of wireless sensor networks in aircraft health management grew exponentially over the past few decades. Wireless sensor networks provide technology that reduces the amount of wiring for aircraft, thereby reducing the weight and cost of aircraft. One of the most significant limitations in the use of wireless sensor networks in aircraft health management systems is the availability of power sources. Developing Wireless Sensor Network nodes that can generate and harvest their autonomous power supply continuously is a bottleneck that has been the preoccupation of engineers for many years. The amount of energy a network of Wireless Sensors can harvest fluctuates and is difficult to predict. As a result, existing predictors of energy harvesting are prone to errors. Models-free schemes such as expert systems are thus preferred for energy management strategies. The main aim of this thesis is to propose expert-based systems for energy harvesting in aircraft to enhance wireless sensor nodes life span by improving energy harvesting, energy storage and packet loss probability. In this context, a novel integrated approach based on the Markov chain was proposed for energy harvesting in aircraft. Simulation results and quantitative analysis showed that the integration of Piezoelectric and Thermoelectric harvesters with stochastic scheduling had a better performance in terms of energy storage, energy harvesting and packet loss probability. There was also an increase in energy storage with five Markov states compared to that of two Markov states. The packet loss probability of the integrated approach with five Markov states was better than that of two Markov states. The results also showed that the integrated approach with five Markov states harvested more energy than two Markov states. The novel integration of LTspice and NS-3 simulators was proposed. The LTspice and NS-3 integration was validated by deploying the Fuzzy logic control approach in energy harvesting. Simulation results and quantitative analysis based on Fuzzy control logic expert system indicated that the integration of LTspice and NS-3 was found to be better in energy harvesting compared to non-fuzzy control systems. The downtime ratio and energy utilization efficiency of the wireless sensor nodes were also found to be better than non-fuzzy control. The power management based LEACH routing protocol was also proposed. The simulation results and quantitative analysis showed that the average harvested energy based on the LEACH routing protocol deployed with fuzzy logic and Markov chain was better compared to those with direct communication based on Markov chain and fuzzy logic systems.
Open Access
Reliable and safe control and navigation for autonomous vehicles in dynamic urban environments
(Cranfield University, 2019-09) Philippe, Charles; Tsourdos, Antonios; Adouane, L.; Shin, Hyo-Sang; Thuilot, B.
In this thesis is presented an algorithmic architecture for systematic risk evaluation, mitigation and management intended for autonomous transportation vehicles. The methods presented span low level control, trajectory tracking and multi-vehicle coordination. A task separation between low level steering control and trajectory tracking has been implemented to spread the design effort across two functional blocks. A robust low level controller has been designed, and a comfortable and flexible Model Predictive Controller (MPC) has been implemented for trajectory tracking. This controller has been associated with a supervision mechanism that monitors its performance in real time to evaluate the probability to underperform. When such a risk is identified, the speed of the system is adapted. The multi-vehicle coordination block fulfils the planning task. It is a decentralized, probabilistic optimization algorithm that is naturally risk-adverse. It has been made compatible with mixed-traffic scenarios with human drivers on the road. Results show that risks are monitored and managed across the whole architecture. Furthermore, easy to understand risk metrics are outputted to make the algorithms decisions understandable by the users and engineers working on the system. The work in this thus proposes systematic risk management techniques transposable to all autonomous vehicles systems. It has been tested in simulations and on test vehicles.

Browse

Browsing PhD, EngD and MSc by research theses (SATM) by Course name "Aerospace"

Results Per Page

Sort Options