<p dir="ltr">I conducted an in-depth analysis of the frequency and severity of external load accidents involving Chinook helicopters over a period of 30 years. The literature review encompassed General Aviation (GA) and ground-based safety organizations, while the data analysis predominantly relied on secondary data from the Army Combat Readiness Center (ACRC). In conducting this study, I aimed to identify key trends, causes, and effects of these accidents, particularly emphasizing material failures, human errors, and the substantial impact of rotor downwash as horizontal wind velocities in proximity to the ground. The study's goal was to improve safety and operational efficiency in Chinook external load operations by identifying frequency and severity of accidents over a 30-year period. The hope was that this would provide valuable insights for improvements in risk mitigation techniques.</p><p dir="ltr">By using an exploratory secondary data analysis of both publicly available U.S. Army accidents and accident data provided by the U.S. ACRC, I found that Chinook rotor downwash, which manifests as horizontal wind velocity when in close proximity to the ground, is the most significant and underreported factor. Based on the findings of this research, I recommend improved classification and documentation of such accidents. The findings highlighted the urgency of updating training and operational procedures to effectively address the unique challenges posed by rotor downwash and high gross weights in proximity to the ground, typical of Chinook external load Pickup and Landing Zone (PZ/LZ) operations. Implementing these recommendations is expected to enhance safety measures in both training and practical operations, ultimately reducing future accidents and improving safety standards in the aviation industry.</p>

Orientadores: Jose Manoel Balthazar, Paulo R. G. Kurka / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica / Made available in DSpace on 2018-08-09T22:56:56Z (GMT). No. of bitstreams: 1 Pereira_DaniloCarlos_D.pdf: 3711639 bytes, checksum: 2413d33f619760c04be8cc5320c0a84b (MD5) Previous issue date: 2007 / Resumo: Neste trabalho analisou-se a dinâmica não linear de uma aeronave em vôo longitudinal. Efetuou-se a análise do comportamento bifurcacional da aeronave F-8 ¿Cruzader¿. Na análise bifurcacional foi estudado o comportamento topológico desta aeronave tomando-se dois parâmetros de controle: a deflexão do profundor e a alteração da massa da referida aeronave. Ante a pesquisa desenvolvida, foi proposto um projeto de controle linear ótimo com o objetivo de estabilizar as oscilações do ângulo de ataque, considerando-se regiões criticas do comportamento não linear da aeronave. Adicionalmente, incluiu-se no modelo matemático a variação da velocidade longitudinal da aeronave, visto tratar-se de simulações numéricas em um túnel de vento virtual / Abstract: In this work it was analyzed the non linear dynamic of an aircraft taken onto longitudinal flight. It was done analysis of the bifurcacional behavior of the aircraft F-8 ¿Cruzader¿. In the bifurcational analysis was studied the topological behavior of this aircraft taken into account two parameters of control: the deflection of the elevator and the alteration of the mass of the related aircraft. In the face of the developed research, an optimum linear control project was proposed with the objective of stabilizing the oscillations of the angle-of-attack. Additionally, the variation of the longitudinal speed of the aircraft was included in the mathematic model in order to simulate the oscillatory movement of the aircraft considered, in a tunnel of virtual wind / Doutorado / Materiais e Processos de Fabricação / Mestre em Engenharia Mecânica

Le décrochage profond est un cas particulier du décrochage d’un avion, où l'empennage horizontal est entièrement situé dans le sillage décollé de la voilure principale. Le plan perd ainsi son efficacité, ce qui se traduit par une position d'équilibre en tangage stable, à une incidence élevée, dont il est impossible de sortir par une manœuvre simple. L’objectif de cette étude est de caractériser l’aérodynamique associée à ce phénomène et de proposer une procédure d’identification et de récupération. Il est proposé une démarche visant à déterminer la dynamique bidimensionnelle de l’écoulement autour d’une configuration aéronautique de référence. Les coefficients aérodynamiques, obtenus dans une large plage d’incidence, mettent en évidence l’effet de l’interaction entre les profils sur le décrochage, qui impacte principalement le profil aval. L’analyse des champs de vitesse fournit l’étendue et l’évolution axiale des sillages des profils. Un traitement des champs de vitesse par moyennes de phase permet de reconstruire la dynamique temporelle. À partir de ces résultats, un modèle potentiel de forçage de l’écoulement autour du profil aval permet d’expliquer la modification du coefficient de portance imposé par l’interaction. Des simulations numériques de l’écoulement, qui fournissent des champs résolus en temps, permettent de retrouver certaines évolutions expérimentales. L’ensemble des résultats est utilisé, en parallèle à des données issues d’un aéronef réel, dans un modèle de vol longitudinal afin d’analyser le comportement dynamique de l’avion. Des critères permettant d’identifier la dynamique qui conduit à cet équilibre, fournissent une détection précoce de ce dernier. / Deep stall is a specific type of airplane stall, in which the horizontal tail is inside the detached wake of the main wing. The tail loses its efficiency, leading to a stable pitching equilibrium position with a high angle-of-attack, without any easy recovery procedure. The aim of the study is to characterize the aerodynamic associated to that phenomenon in order to propose an identification and recovery procedure. The approach consists in a two-dimensional flow characterization based on an aeronautical reference configuration. Aerodynamic coefficients, obtained for a wide range of angles-of-attack, show the interaction between the airfoils on the stall of the downstream airfoil. The analysis of velocity fields gives the width and the axial development of the airfoils wakes. Phase-averages of velocity fields lead to the synthesis of flow time-development. With these results, a potential model of flow forcing on the downstream airfoil explains the lift coefficient alteration imposed by the interaction. Flow numerical simulations, giving time-resolved fields, provide good accordance with experimental developments .The whole set of results is used, concurrently with real aircraft data, inside a longitudinal flight model in order to analyze the airplane dynamical behavior. Criteria for the identification of the dynamic leading to that equilibrium provide a rapid detection of deep stall and the implementation of a recovery strategy.

This thesis proposes a general approach to solve the offline flight-maneuver identification problem using machine learning methods. The purpose of the study was to provide means for the aircraft professionals at the flight test and verification department of Saab Aeronautics to automate the procedure of analyzing flight test data. The suggested approach succeeded in generating binary classifiers and multiclass classifiers that identified six flight maneuvers of different complexity from real flight test data. The binary classifiers solved the problem of identifying one maneuver from flight test data at a time, while the multiclass classifiers solved the problem of identifying several maneuvers from flight test data simultaneously. To achieve these results, the difficulties that this time series classification problem entailed were simplified by using different strategies. One strategy was to develop a maneuver extraction algorithm that used handcrafted rules. Another strategy was to represent the time series data by statistical measures. There was also an issue of an imbalanced dataset, where one class far outweighed others in number of samples. This was solved by using a modified oversampling method on the dataset that was used for training. Logistic Regression, Support Vector Machines with both linear and nonlinear kernels, and Artifical Neural Networks were explored, where the hyperparameters for each machine learning algorithm were chosen during model estimation by 4-fold cross-validation and solving an optimization problem based on important performance metrics. A feature selection algorithm was also used during model estimation to evaluate how the performance changes depending on how many features were used. The machine learning models were then evaluated on test data consisting of 24 flight tests. The results given by the test data set showed that the simplifications done were reasonable, but the maneuver extraction algorithm could sometimes fail. Some maneuvers were easier to identify than others and the linear machine learning models resulted in a poor fit to the more complex classes. In conclusion, both binary classifiers and multiclass classifiers could be used to solve the flight maneuver identification problem, and solving a hyperparameter optimization problem boosted the performance of the finalized models. Nonlinear classifiers performed the best on average across all explored maneuvers.

<p>  </p> <p>As space agencies look to conduct more scientific missions beyond Earth orbit, low-cost access to space becomes indispensable. Small satellites (smallsats) fulfill this need as they can be developed at a fraction of the cost of traditional large satellites. Consequently, smallsats are being envisioned for planetary science missions at several destinations including Mars. However, a significant challenge for interplanetary smallsats is performing fully-propulsive orbit insertion because modern smallsat propulsion technologies have limited total velocity change capabilities. At destinations with significant atmospheres, this challenge can be circumvented via <em>aerocapture</em>, a technique that uses a single atmospheric pass to convert a hyperbolic approach trajectory into a captured elliptical orbit. Aerocapture has been shown to enable significant propellant mass savings as compared to fully-propulsive orbit insertion, making it an attractive choice for smallsats. Performing aerocapture with smallsats requires a suitable vehicle design that satisfies the associated control requirements and volumetric constraints. To address this requirement, this dissertation proposes the <em>morphable entry system </em>(MES), a conceptual deployable entry vehicle that utilizes shape morphing to follow a desired atmospheric flight profile during aerocapture. The aerocapture performance of the MES at Mars is investigated using a six degree-of-freedom aerocapture simulation environment. The shape morphing strategy employed by the MES is shown to be feasible for targeting desired angle of attack and sideslip angle profiles that lead to successful orbit captures. Furthermore, the robustness of the MES to simulated day-of-flight uncertainties while employing angle of attack control is demonstrated through a Monte Carlo dispersion analysis. The major contributions of this research as well as areas of future work are described.</p>

<p dir="ltr">The absence or limited availability of operational statistics at general aviation airports restricts airport managers and operators from assessing comprehensive operational data. The traditional manual compilation of operational statistics is labor-intensive and lacks the depth and accuracy to depict a holistic picture of a general aviation airport’s operations. This research developed a reliable and efficient approach to address the problem by providing a comprehensive and versatile flight status identification framework. </p><p dir="ltr">Leveraging the BlueSky flight simulation module, the research can generate a synthetic flight database to emulate real-world general aviation aircraft’s flight scenarios. Two neural network architectures, namely, an RNN-GAN network and a refined Seq2Seq network, were explored to examine their capability to reconstruct flight trajectories. The Seq2Seq network, which demonstrated better performance, was further employed to estimate the simulated aircraft’s different metrics, such as internal mechanical metrics and flight phase. Additionally, this research undertook an array of diverse tailored evaluation techniques to assess the efficacy of flight status predictions and conducted comparative analyses between various configurations. </p><p dir="ltr">Furthermore, the research concluded by discussing the future development of the framework, emphasizing its potential for generalization across various flight data applications and scenarios. The enhanced methodology for collecting operational statistics and the analysis tool will enable airport managers and regulators to better receive a comprehensive view of the airport’s operations, facilitating airport planning and development.</p>

This thesis investigates the effectiveness of employing pretraining and a discrete "control signal" bottleneck layer in a neural network trained in aircraft navigation through deep reinforcement learning. The study defines two distinct tasks to assess the efficacy of this approach. The first task is utilized for pretraining specific parts of the network, while the second task evaluates the potential benefits of this technique. The experimental findings indicate that the network successfully learned three main macro actions during pretraining. flying straight ahead, turning left, and turning right, and achieved high rewards on the task. However, utilizing the pretrained network on the transfer task yielded poor performance, possibly due to the limited effective action space or deficiencies in the training process. The study discusses several potential solutions, such as incorporating multiple pretraining tasks and alterations of the training process as avenues for future research. Overall, this study highlights the challanges and opportunities associated with combining pretraining with a discrete bottleneck layer in the context of simulated aircraft navigation using reinforcement learning. / Denna studie undersöker effektiviteten av att använda förträning och en diskret "styrsignal" som fungerar som flaskhals i ett neuralt nätverk tränat i flygnavigering med hjälp av djup förstärkande inlärning. Studien definierar två olika uppgifter för att bedöma effektiviteten hos denna metod. Den första uppgiften används för att förträna specifika delar at nätverket, medan den andra uppgiften utvärderar de potentiella fördelarna med denna teknik. De experimentella resultaten indikerar att nätverket framgångsrikt lärde sig tre huvudsakliga makrohandlingar under förträningen: att flyga rakt fram, att svänga vänster och att svänga höger, och uppnådde höga belöningar för uppgiften. Men att använda det förtränade nätverket för den uppföljande uppgiften gav dålig prestation, möjligen på grund av det begränsade effektiva handlingsutrymmet eller begränsningar i träningsprocessen. Studien diskuterar flera potentiella lösningar, såsom att inkorporera flera förträningsuppgifter och ändringar i träningsprocessen, som möjliga framtida forskningsvägar. Sammantaget belyser denna studie de utmaningar och möjligheter som är förknippade med att kombinera förträning med ett diskret flaskhalslager inom kontexten av simulerad flygnavigering och förstärkningsinlärning.

Applications such as urban surveillance, search and rescue, agricultural applications, military applications, etc., require miniature air vehicles (MAVs) to fly for a long time. But they have restricted flight duration due to their dependence on battery life, which necessitates optimal path planning. The generated optimal path should obey the curvature limits prescribed by the minimum turn radius/ maximum turn rate of the MAV. Further, in a dynamically changing environment, the final configuration that the MAV has to achieve may change en route, which demands the path to be replanned by an airborne processor in real-time. As MAVs are small in size and light in weight, wind has a very significant effect on the flight of MAVs and the computation of the minimum-time path in the presence of wind plays an important role. The thesis develops feasible trajectory generation algorithms which are fast, efficient, optimal and implementable in an onboard computer for rectilinear and circular path convergence problems and waypoint following problems both in the absence and in the presence of wind. The first part of the thesis addresses the problem of computation of optimal trajectories when MAVs fly on a two-dimensional (2D) plane maintaining a constant altitude. The shortest path is computed for MAVs from a given initial position and orientation to a given final path with a specified direction as required for a given mission. Unlike the classical Dubins problem where the shortest path was computed between two given configurations (position and orientation), the final point in this case is not specified. However, the final path, which can either be a rectilinear path or a circular path, and the direction to which the MAV should converge, is specified. The time-optimal path of MAVs is developed in the presence of wind mainly using the geometric approach although a few important properties are also obtained using optimal control theory, specifically, Pontryagin’s minimum principle (which provides only the necessary condition for optimality) for control-constrained systems. The complete optima l solution to this problem in all its generality is a major contribution of this thesis as existing methods in the literature that address this problem are either not optimal or do not give a complete solution. Further, the time-optimal path for specified initial and final configurations is generated in reasonably short time without computing all the path lengths of possible candidate paths, which is the method that exists in the literature for similar problems. Simulation results illustrate path generation for various cases, including the presence of steady and time-varying wind. Another problem in MAV path planning in 2D addressed in this thesis computes an extremal path that transitions between two consecutive waypoint segments (obtained by joining two way points in sequence) in a time-optimal fashion. This designed trajectory, named as γ-trajectory, is also used to track the maximum portion of waypoint segments in minimum time and the shortest distance between this trajectory and the associated waypoint can be set to a desired value. Another optimal path, called the loop trajectory, that goes through the way points as well as through the entire waypoint segments, is also proposed. Subsequently, the thesis proposes algorithms to generate trajectories in the presence of steady wind and compares these with the optimal trajectory generated using nonlinear programming based multiple shooting method to show that the generated paths are optimal in most cases. In three-dimensional (3D) space, if the initial and final configurations – in terms of (X,Y,Z) position, heading angle and flight path angle- of the vehicle are specified then shortest path computation is an interesting problem in literature. The proposed method in this thesis is based on 3D geometry and, unlike the existing iterative methods which yield suboptimal paths and are computationally more intensive, this method generates the shortest path in much less time. Due to its simplicity and low computational requirements, this approach can be implemented on a MAV in real-time. But, If the path demands very high pitch angle (as in the case of steep climbs), the generated path may not be flyable for an aerial vehicle with limited range of flight path angles. In such cases numerical methods, such as multiple shooting, coupled with nonlinear programming, are used to obtain the optimal solution. The time-optimal 3D path is also developed in the presence of wind which has a magnitude comparable to the speed of MAVs. The simulation results show path generation for a few sample cases to show the efficacy of the proposed approach as compared to the available approach in the literature. Next, the path convergence problem is studied in 3D for MAVs. The shortest path is generated to converge to a rectilinear path and a circular path starting from a known initial position and orientation. The method is also extended to compute the time-optimal path in the presence of wind. In simulation, optimal paths are generated for a variety of cases to show the efficacy of the algorithm. The other problem discussed in this thesis considers curvature-constrained trajectory generation technique for following a series of way points in 3D space. Extending the idea used in 2D, a γ-trajectory in 3D is generated to track the maximum portion of waypoint segments with a desired shortest distance between the trajectory and the associated waypoint. Considering the flyability issue of the plane a loop-trajectory is generated which is flyable by a MAV with constrained flight path angle. Simulation results are given for illustrative purposes. The path generation algorithms are all based on a kinematic model, considering the vehicle as a point in space. Implementing these results in a real MAV will require the dynamics of the MAV to be considered. So, a 6-DOF SIMULINK model of a MAV is used to demonstrate the tracking of the computed paths both in 2D plane and in 3D space using autopilots consisting of proportional-integral-derivative (PID )controllers .Achieving terminal condition accurately in real-time, if there is noisy measurement of wind data, is also addressed.

The study is conducted to determine practical modelling and simulation techniques to perform dynamic stability and performance analysis on a 3 Degrees of freedom aircraft model using a Modelica-based commercial tool called Modelon Impact. This study is based on a conceptual aircraft model where in-depth details about the aircraft configuration are unknown and the aim is to determine a suitable model that can capture the longitudinal dynamics and aerodynamic constraints of the aircraft during the conceptual design phase. Requirements include short execution time, easy model development, and minimal data requirements. Therefore, this thesis aims at developing plant and control architectures in  Modelon Impact which can be utilized for the rapid development of aircraft concepts with adequate fidelity in a longitudinal mission-based tracking environment. In a conceptual aircraft design environment, to identify a suitable methodology that mitigates the limitations of a traditional feedback controller, two methodologies are considered for comparison: Sequential DAE resolution (SDR) and Dynamic inversion (DI) control which is discussed from an object-oriented aircraft model. The advantages and shortcomings of each of the models discussed above are compared by conducting several experiments in increasing order of longitudinal mission complexity, and the most appropriate model among the two for a conceptual stage of aircraft design development is ascertained. The two methodologies discussed are compared for their level of complexity, code structure, readability, and ease of usability.

<p>The main accomplishments of this research are </p> <p>(1) developed a high fidelity computational methodology based on large eddy simulation to capture lean blowout (LBO) behaviors of different fuels; </p> <p>(2) developed fundamental insights into the combustion processes leading to the flame blowout and fuel composition effects on the lean blowout limits; </p> <p>(3) developed artificial intelligence-based models for early detection of the onset of the lean blowout in a realistic complex combustor. </p> <p>The methodologies are demonstrated by performing the lean blowout (LBO) calculations and statistical analysis for a conventional (A-2) and an alternative bio-jet fuel (C-1).</p> <p>High-performance computing methodology is developed based on the large eddy simulation (LES) turbulence models, detailed chemistry and flamelet based combustion models. This methodology is employed for predicting the combustion characteristics of the conventional fuels and bio-derived alternative jet fuels in a realistic gas turbine engine. The uniqueness of this methodology is the inclusion of as-it-is combustor hardware details such as complex hybrid-airblast fuel injector, thousands of tiny effusion holes, primary and secondary dilution holes on the liners, and the use of highly automated on the fly meshing with adaptive mesh refinement. The flow split and mesh sensitivity study are performed under non-reacting conditions. The reacting LES simulations are performed with two combustion models (finite rate chemistry and flamelet generated manifold models) and four different chemical kinetic mechanisms. The reacting spray characteristics and flame shape are compared with the experiment at the near lean blowout stable condition for both the combustion models. The LES simulations are performed by a gradual reduction in the fuel flow rate in a stepwise manner until a lean blowout is reached. The computational methodology has predicted the fuel sensitivity to lean blowout accurately with correct trends between the conventional and alternative bio-jet fuels. The flamelet generated manifold (FGM) model showed 60% reduction in the computational time compared to the finite rate chemistry model. </p> <p>The statistical analyses of the results from the high fidelity LES simulations are performed to gain fundamental insights into the LBO process and identify the key markers to predict the incipient LBO condition in swirl-stabilized spray combustion. The bio-jet fuel (C-1) exhibits significantly larger CH₂O concentrations in the fuel-rich regions compared to the conventional petroleum fuel (A-2) at the same equivalence ratio. It is observed from the analysis that the concentration of formaldehyde increases significantly in the primary zone indicating partial oxidation as we approach the LBO limit. The analysis also showed that the temperature of the recirculating hot gases is also an important parameter for maintaining a stable flame. If this temperature falls below a certain threshold value for a given fuel, the evaporation rates and heat release rated decreases significantly and consequently leading to the global extinction phenomena called lean blowout. The present study established the minimum recirculating gas temperature needed to maintain a stable flame for the A-2 and C-1 fuels. </p> The artificial intelligence (AI) models are developed based on high fidelity LES data for early identification of the incipient LBO condition in a realistic gas turbine combustor under engine relevant conditions. The first approach is based on the sensor-based monitoring at the optimal probe locations within a realistic gas turbine engine combustor for quantities of interest using the Support Vector Machine (SVM). Optimal sensor locations are found to be in the flame root region and were effective in detecting the onset of LBO ~20ms ahead of the event. The second approach is based on the spatiotemporal features in the primary zone of the combustor. A convolutional autoencoder is trained for feature extraction from the mass fraction of the OH ( data for all time-steps resulting in significant dimensionality reduction. The extracted features along with the ground truth labels are used to train the support vector machine (SVM) model for binary classification. The LBO indicator is defined as the output of the SVM model, 1 for unstable and 0 for stable. The LBO indicator stabilized to the value of 1 approximately 30 ms before complete blowout.