Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2010. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 106-109). / Small satellites are gaining popularity in the space industry and reduction in spacecraft size requires scaling down its propulsion system. Low-power electric propulsion poses a unique challenge due to various scaling penalties. Of high-performance plasma thrusters, the radio-frequency ion engine is most likely to succeed in scaling as it does not require an externally applied magnetic field and is structurally simple to construct. As part of a design package an original two-dimensional simulation code for radio-frequency ion engine discharge is developed. The code models the inductive plasma with fluid assumption and resolves the electromagnetic wave in the time domain. Major physical effects considered include magnetic field diffusion and coupling, plasma current induction and ambipolar plasma diffusion. The discharge simulation is benchmarked with data from an experimental thruster. It shows excellent performance in predicting the load power and the internal power loss of the plasma. Predictability of anode current depends on the operating power but is generally adequate. Optimum skin depth on the order of half of chamber radius is suggested by the simulation. The code also demonstrates excellent scaling ability as it successfully predicts the performance of a smaller thruster with errors less than 10%. Using the code a brief optimization study was conducted and the results suggest the maximum thrust efficiency does not necessarily occur at the same frequency that maximizes the power coupling efficiency of the matching circuit. / by Michael Meng-Tsuan Tsay. / Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2010. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 73-74). / The challenge of achieving safe flight comes into sharp focus in the face of adverse conditions caused by faults, damage, or upsets. When these situations occur, the corresponding uncertainties directly affect the safe operation of the aircraft. A technology that has the potential for enabling a safe flight under these adverse conditions is adaptive control. One of the main features of an adaptive control architecture is its ability to react to changing characteristics of the underlying aircraft dynamics. This thesis proposes the building blocks of an adaptable and reconfigurable control technology that ensures safe flight under adverse flight conditions. This technology enables the synthesis of such controllers as well as the systematic evaluation of their robustness characteristics. The field of adaptive control is a mature theoretical discipline that has evolved over the past thirty years, embodying methodologies for controlling uncertain dynamic systems with parametric uncertainties [1, 2, 3, 4, 5, 6]. Through the efforts of various researchers over this period, systematic methods for the control of linear and nonlinear dynamic systems with parametric and dynamic uncertainties have been developed [7, 8, 9, 10, 11, 12]. Stability and robustness properties of these systems in the presence of disturbances, time-varying parameters, unmodeled dynamics, time-delays, and various nonlinearities, have been outlined in the references [4]-[13] as well as in several journal and conference papers over the same period. / (cont.) In this thesis, we consider the control of a transport aircraft model that resembles the Generic Transport Model [14]. While the vehicles' geometry and aerodynamic model are those of a C5 aircraft, every other aspect has been made to coincide with the GTM, e.g. anti wind-up logic, time-delay due to telemetry, baseline control structure, low-pass and wash-out filters. We delineate the underlying nonlinear model of this aircraft, and introduce various damages, and failures into this model. An adaptive control architecture is proposed which combines a nominal controller that provides a satisfactory performance in the absence of adverse conditions, and an adaptive controller that is capable of accommodating various adverse conditions including actuator saturation. The specific adverse conditions considered can be grouped into the following three categories, (a) upsets, (b) damages, and (c) actuator failures. Specific cases in (a) include flight upsets in initial conditions of various states including angle of attack, cases in (b) include situations where structural failures cause changes in the location of the Center-of-Gravity (CG)[15], while cases in (c) include situations where symmetric and asymmetric failures in control surfaces and engines occur. These failures include losses in control effectiveness, and locked-in-place control surface deflections. The resilience of the adaptive controller to uncertainty is evaluated for safety using the control verification methodology proposed in [16]. / (cont.) This methodology enables the determination of ranges of uncertainty for which a prescribed set of closed-loop requirements are satisfied. This thesis studies several one-dimensional uncertainty analyzes for two flight maneuvers that focus on the longitudinal and lateral dynamics. As compared to the baseline controller, the adaptive controller enlarges the region of safe operation by a sizable margin in all but one of the cases considered. / by Megumi Matsutani. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2003. / Includes bibliographical references (p. 185-188). / The problem of performance optimization of a Common Aero Vehicle (CAV) is considered. In particular, the CAV is modeled as an unpowered high lift-to-drag ratio Earth penetrating re-entry vehicle. The CAV mission design problem is to determine a steering command that takes the CAV from a known initial state to a target on the surface of the Earth while optimizing a given performance index and satisfying all of the constraints imposed during flight. The CAV mission design problem is formulated as an optimal control problem. The optimal control problem is transformed to a nonlinear programming problem using a Legendre Pseudospectral Method. The nonlinear programming problem is then solved using a sparse nonlinear optimization algorithm. Once a solution to the CAV mission design problem is obtained, three main studies are conducted. First, the accuracy of the Legendre Pseudospectral Method is evaluated and the desirable characteristics of the solution to the CAV mission design problem are defined. Second, a study is conducted to demonstrate the effect of the parameters on the performance of the CAV. This parametric study demonstrates the use of the Legendre Pseudospectral method as a design tool and provides insight to the behavior of the CAV. Third, a preliminary investigation is performed on the real-time application of the Legendre Pseudospectral Method to the CAV mission design problem. This real-time analysis includes an assessment of the robustness of the solution to realistic environmental disturbances. / by Kimberley A. Clarke. / S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2005. / Includes bibliographical references (p. 194-198). / Real options analysis is being increasingly used as a tool to evaluate investments under uncertainty; however, traditional real options methodologies have some shortcomings that limit their utility, such as the use of the geometric Brownian motion to model the value of the underlying asset and the assumption of a fixed cost to exercise the option. In this thesis, an alternative real options methodology is developed that overcomes some of the difficulties of traditional approaches. In particular, the methodology proposed here presents an analytical framework that allows the value of completion and the strategy- enabling completion cost (commonly referred to as stock price and strike price in the real options literature, respectively) to be represented by any probability distribution. If these probability distributions can be described analytically, an exact solution to the real options valuation problem can be found. Otherwise, the probability distributions can be generated with numerical simulation (e.g. Monte Carlo simulation), and the answer can then be found numerically. This generalized methodology combines the simplicity of analytical approaches with the flexibility to represent completion costs and the value of completion with any probability distribution. The generalized real options methodology is illustrated with an example from aviation: the decision to launch a new aircraft development program. This type of investment is suitable for real options analysis because of the many uncertainties involved, the long- term nature of the project, and the ability of management to act and influence the project as uncertainties are resolved during its evolution. / (cont.) The analysis shows that investors can use the numerical results of the real options evaluation to determine the investment limits on the different stages of the aircraft program, that managers can use insights from the real options approach to restructure the program to improve the financial feasibility of the project, and that both investors and managers can use the output of derivative analyses to define minimum requirements (in terms of aircraft orders) to ensure program success. / by Bruno Miller. / Ph.D.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1993. / Includes bibliographical references (leaves 103-105). / by Jack Barron Aldrich. / M.S.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2010. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 77-80). / This thesis investigates the use of generalized belief propagation (GBP) and belief propagation (BP) algorithms for distributed inference. The concept of a network region graph is introduced, along with several approximation structures that can be distributed across a network. In this formulation, clustered region graphs are introduced to create a network "backbone" across which the computation for inference is distributed. This thesis shows that clustered region graphs have good structural properties for GBP algorithms. We propose the use of network region graphs and GBP for location-aware networks. In particular, a method for representing GBP messages non-parametrically is developed. As an special case, we apply BP algorithms to mobile networks without infrastructure, and we propose heuristics to optimize degree of network cooperation. Numerical results show a five times performance increase in terms of outage probability, when compared to conventional algorithms. / by Ulric John Ferner. / S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2014. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 77-80). / Shielding critical links enhances network robustness and provides a new way of designing robust networks. We first consider shielding critical links to guarantee network connectivity after any failure under geographical and general failure models. We develop a mixed integer linear program (MILP) to obtain the minimum cost shielding to guarantee the connectivity of a single source-destination (SD) pair under a general failure model, and exploit geographical properties to decompose the shielding problem under a geographical failure model. We extend our MILP formulation to guarantee the connectivity of the entire network, and use Benders decomposition to significantly reduce the running time by exploiting its partial separable structure. We extend the algorithms to guarantee partial network connectivity, and observe that significantly less shielding is required, especially when the failure region is small. To mitigate the effect of random link failures on network connectivity, we consider increasing the effective min-cut of the network by shielding, where shielded links cannot be contained in effective cuts. For a single SD pair, we develop an efficient algorithm to increase the effective min-cut by one, and develop a MILP with a small number of constraints to increase the effective min-cut by an arbitrary value. Then we extend the MILP to obtain the optimal shielding to increase the effective min-cut for the entire network, which can be used to solve realistic size problems. Finally, we consider shielding critical nodes in random graphs. We demonstrate the importance of high degree nodes in random graphs constructed under the configuration model. The occupancy of higher degree nodes leads to a larger size of the giant component. Moreover, shielding a small fraction of nodes in power law random graphs guarantees the existence of a giant component if the exponent is less than three. / by Jianan Zhang. / S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 173-174). / This thesis presents the development of a dimensionless blood microcirculation model for the study of blood microcirculation flows. It is a two dimensional, axially symmetric, incompressible, Newtonian-flow, Krogh cylinder model subjected to axially periodic boundary conditions. This model formulation allows for the use of the streamfunction-vorticity formulation of the Navier-Stokes equation, which offers simplification to boundary conditions and also allows for the use of a non-uniform, collocated mesh. A streamfunction vorticity formulation of the Immersed Boundary Method is also developed, specifically for the boundary conditions along the immersed boundary (red blood cell membrane). Periodic boundary conditions are used, with the assumption of fully-developed flow, in order to focus on the effects of the transient diffusion of oxygen into the surrounding tissue, orthogonal to the capillary flow direction. / by Kevin Sabo. / S.M.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1996. / Includes bibliographical references (p. 115-116). / by Louis J. Lintereur. / M.S.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1993. / Includes bibliographical references (leaves 212-213). / by Jabin Todd Bell. / M.S.