Engineering Commons^™

Model-Based Design Of An Optimal Lqg Regulator For A Piezoelectric Actuated Smart Structure Using A High-Precision Laser Interferometry Measurement System, Grant P. Gallagher

Master's Theses

Smart structure control systems commonly use piezoceramic sensors or accelerometers as vibration measurement devices. These measurement devices often produce noisy and/or low-precision signals, which makes it difficult to measure small-amplitude vibrations. Laser interferometry devices pose as an alternative high-precision position measurement method, capable of nanometer-scale resolution. The aim of this research is to utilize a model-based design approach to develop and implement a real-time Linear Quadratic Gaussian (LQG) regulator for a piezoelectric actuated smart structure using a high-precision laser interferometry measurement system to suppress the excitation of vibratory modes.

The analytical model of the smart structure is derived using the …

Real-Time Predictive Control Of Connected Vehicle Powertrains For Improved Energy Efficiency, Joseph Oncken

Dissertations, Master's Theses and Master's Reports

The continued push for the reduction of energy consumption across the automotive vehicle fleet has led to widespread adoption of hybrid and plug-in hybrid electric vehicles (PHEV) by auto manufacturers. In addition, connected and automated vehicle (CAV) technologies have seen rapid development in recent years and bring with them the potential to significantly impact vehicle energy consumption. This dissertation studies predictive control methods for PHEV powertrains that are enabled by CAV technologies with the goal of reducing vehicle energy consumption.

First, a real-time predictive powertrain controller for PHEV energy management is developed. This controller utilizes predictions of future vehicle velocity …

Utilizing Trajectory Optimization In The Training Of Neural Network Controllers, Nicholas Kimball

Master's Theses

Applying reinforcement learning to control systems enables the use of machine learning to develop elegant and efficient control laws. Coupled with the representational power of neural networks, reinforcement learning algorithms can learn complex policies that can be difficult to emulate using traditional control system design approaches. In this thesis, three different model-free reinforcement learning algorithms, including Monte Carlo Control, REINFORCE with baseline, and Guided Policy Search are compared in simulated, continuous action-space environments. The results show that the Guided Policy Search algorithm is able to learn a desired control policy much faster than the other algorithms. In the inverted pendulum …

Verification Of Stochastic Reach-Avoid Using Rkhs Embeddings, Adam J. Thorpe

Electrical and Computer Engineering ETDs

A solution to the terminal-hitting and first-hitting stochastic reach-avoid problem for a Markov control process is presented. This solution takes advantage of a nonparametric representation of the stochastic kernel as a conditional distribution embedding within a reproducing kernel Hilbert space (RKHS). Because the disturbance is modeled as a data-driven stochastic process, this representation avoids intractable integrals in the dynamic recursion of the reach-avoid problem since the expectations can be calculated as an inner product within the RKHS. An example using a high-dimensional chain of integrators is presented, as well as for Clohessy-Wiltshire-Hill (CWH) dynamics.

Control Oriented Nonlinear Model Reduction For Distributed Parameter Systems, Samir Sahyoun

Doctoral Dissertations

The development of model reduction techniques for physical systems modeled by partial differential equations (PDEs) has been a very active research area. Large number of states is needed to accurately capture the dynamics of such systems which makes them unsuitable for control design. The order of the system must be reduced prior to control design. In this dissertation, new methods that generalize the popular proper orthogonal decomposition (POD) to nonlinear PDEs are investigated. In particular, cluster based POD algorithms are developed and applied to the one and two dimensional Burgers equations that govern a nonlinear convective ow. Each cluster contains …

Optimizing Control Of Shell Eco-Marathon Prototype Vehicle To Minimize Fuel Consumption, Chad Louis Bickel

Master's Theses

Every year the automotive industry strives to increase fuel efficiency in vehicles. When most vehicles are designed, fuel efficiency cannot always come first. The Shell Eco-marathon changes that by challenging students everywhere to develop the most fuel-efficient vehicle possible. There are many different factors that affect fuel efficiency, and different teams focus on different vehicle parameters. Currently, there is no straightforward design tool that can be used to help in Shell Eco-marathon vehicle design. For this reason, it is difficult to optimize every vehicle parameter for maximum fuel efficiency.

In this study, a simulation is developed by using basic vehicle …

An Optimal Energy Management Strategy For Hybrid Electric Vehicles, Amir Rezaei

Dissertations, Master's Theses and Master's Reports

Hybrid Electric Vehicles (HEVs) are used to overcome the short-range and long charging time problems of purely electric vehicles. HEVs have at least two power sources. Therefore, the Energy Management (EM) strategy for dividing the driver requested power between the available power sources plays an important role in achieving good HEV performance.

This work, proposes a novel real-time EM strategy for HEVs which is named ECMS-CESO. ECMS-CESO is based on the Equivalent Consumption Minimization Strategy (ECMS) and is designed to Catch Energy Saving Opportunities (CESO) while operating the vehicle. ECMS-CESO is an instantaneous optimal controller, i. e., it does not …

A Radial Basis Function Method For Solving Optimal Control Problems., Hossein Mirinejad

Electronic Theses and Dissertations

This work presents two direct methods based on the radial basis function (RBF) interpolation and arbitrary discretization for solving continuous-time optimal control problems: RBF Collocation Method and RBF-Galerkin Method. Both methods take advantage of choosing any global RBF as the interpolant function and any arbitrary points (meshless or on a mesh) as the discretization points. The first approach is called the RBF collocation method, in which states and controls are parameterized using a global RBF, and constraints are satisfied at arbitrary discrete nodes (collocation points) to convert the continuous-time optimal control problem to a nonlinear programming (NLP) problem. The …