Search

This dissertation presents several modern control methods for autonomous reversing of long combination vehicles (LCVs). These approaches not only significantly improve the performance of previous autonomous reversing systems, but also address a major gap in the reversing control literature for LCVs. The methods were validated by implementing them at full-scale on experimental articulated vehicles owned by the 'Cambridge Vehicle Dynamics Consortium' (CVDC). Experimental results were in very good agreement with simulation results. Previous path following control methods for autonomous reversing of LCVs have focussed on minimising the tracking error between the rear-end of the combination and a desired path, irrespective of the motion of the rest of the vehicle. A significant disadvantage of these strategies is that the other parts of the vehicle, particularly the tractor unit, can experience large excursions from the reference path, thus making their implementation infeasible for manoeuvres in spatially-limited areas, such as warehouse yards and normal roads. The 'Minimum Swept Path Control' (MSPC) method was devised to reduce this problem by relaxing the requirement for very accurate path following, while minimising the maximum excursion of the vehicle. This strategy weights both the path-following error at the rear end of the vehicle and the swept path of the front end of the vehicle. MSPC enables the swept path to be reduced by about 50%, compared with path following control, which gives this method more realistic applications. The 'Lane-bounded Reversing Control' (LBRC) method requires a vehicle to satisfy the reversing objectives while constraining the motion to be within a specified 'lane'. The LBRC controller is 'intelligent', which means it can pursue an optimum route without tracking a desired path generated by a path planner and can proactively avoid future potential clashes. Hence, this controller enables the autonomous reversing system to perform a precise, minimum-cost and collision-free manoeuvre to a specified terminal position by planning ahead and making optimal decisions. The controller performance was evaluated in numerous realistic scenarios, both simulated and in field tests at full-scale. The analysis of LBRC provides a solid foundation for the development of more advanced control methods. Adaptive Lane-bounded Reversing Control (ALBRC) and Adaptive Bi-directional Control (ABC) systems were designed to improve the LBRC method. The tuning of the LBRC controller was based on empirical experience and there are many weights to be tuned in the controller configuration. To offset this drawback, the ALBRC algorithm was developed by attaching 'virtual bumpers' to the vehicle system states, and allowing the controller weights to adapt to lane boundaries and obstacles. The ALBRC method simplifies the original tuning process significantly. The ALBRC controller performs well in most cases. However, a solution is not guaranteed if the preview and control horizons are not tuned properly or a vehicle reverses from an arbitrary position. Hence, the ABC algorithm incorporates a so-called 'cusp technology', which allows a vehicle to move forward and backward to realign its position and orientation between attempts at the reversing manoeuvre. In this case, the preview and control horizons do not need to change in different scenarios. The ABC method can significantly reduce computational time compared with using a long preview horizon.

The dissertation consists of two parts. The first part (Chapter 1 to 4) is on some contributions to the development of a non-linear analysis on the quintessential fractal set Sierpinski gasket and its probabilistic interpretation. The second part (Chapter 5) is on the asymptotic tail decays for suprema of stochastic processes satisfying certain conditional increment controls. Chapters 1, 2 and 3 are devoted to the establishment of a theory of backward problems for non-linear stochastic differential equations on the gasket, and to derive a probabilistic representation to some parabolic type partial differential equations on the gasket. In Chapter 2, using the theory of Markov processes, we derive the existence and uniqueness of solutions to backward stochastic differential equations driven by Brownian motion on the Sierpinski gasket, for which the major technical difficulty is the exponential integrability of quadratic processes of martingale additive functionals. A Feynman-Kac type representation is obtained as an application. In Chapter 3, we study the stochastic optimal control problems for which the system uncertainties come from Brownian motion on the gasket, and derive a stochastic maximum principle. It turns out that the necessary condition for optimal control problems on the gasket consists of two equations, in contrast to the classical result on ℝd, where the necessary condition is given by a single equation. The materials in Chapter 2 are based on a joint work with Zhongmin Qian (referenced in Chapter 2). Chapter 4 is devoted to the analytic study of some parabolic PDEs on the gasket. Using a new type of Sobolev inequality which involves singular measures developed in Section 4.2, we establish the existence and uniqueness of solutions to these PDEs, and derive the space-time regularity for solutions. As an interesting application of the results in Chapter 4 and the probabilistic representation developed in Chapter 2, we further study Burgers equations on the gasket, to which the space-time regularity for solutions is deduced. The materials in Chapter 4 are based on a joint work with Zhongmin Qian (referenced in Chapter 4). In Chapter 5, we consider a class of continuous stochastic processes which satisfy the conditional increment control condition. Typical examples include continuous martingales, fractional Brownian motions, and diffusions governed by SDEs. For such processes, we establish a Doob type maximal inequality. Under additional assumptions on the tail decays of their marginal distributions, we derive an estimate for the tail decay of the suprema (Theorem 5.3.2), which states that the suprema decays in a manner similar to the margins of the processes. In Section 5.4, as an application of Theorem 5.3.2, we derive the existence of strong solutions to a class of SDEs. The materials in this chapter is based on the work [44] by the author (Section 5.2 and Section 5.3) and an ongoing joint project with Guangyu Xi (Section 5.4).

The study of biological and genetic information, mostly DNA data, is an extremely important subject which can provide critical information in many areas, such as understanding human diseases or discovering new drugs. A huge number of computing algorithms are developed and available now to help with the study of these, and in order to solve these problems more efficiently and accurately, much attention has been paid in recent decades to developing new and better algorithms.

The stochastic unit commitment problem seeks to determine the day-ahead generating unit on/off schedule in a power system with significant renewable generation. In this context, this dissertation leverages data-driven approaches to balance solution efficiency and modeling accuracy. First, since solving the stochastic unit commitment problem is computationally very costly, we propose two practical data-driven approaches to reduce its computational burden. Second, we build a computationally efficient security-constrained stochastic unit commitment model to identify accurate reserves by considering uncertainties of renewable generation and demand, and generating unit failures. Third, we design two risk metrics to rank committed units based on the impact their failures have on system operation.Since the time available to solve the stochastic unit commitment problem is limited, the first research work in this thesis aims to shorten the solution time of the stochastic unit commitment problem by reducing its size via learning. In particular, learning from historical solved instances, we fix unchanged binary variables and eliminate inactive inequality constraints in the stochastic unit commitment problem. The numerical results show that the reduced problem generally requires significantly less time to solve while obtaining high-quality solutions, which are very close to or indistinguishable from those obtained by solving the original problem.The second research work in this thesis seeks to drastically reduce the computational burden of the network-constrained stochastic unit commitment problem by using a quasi-deterministic proxy. The proxy has the same structure as the original stochastic unit commitment problem but only includes two envelope scenarios, generated based on a large scenario set representing renewable generation uncertainty and demand uncertainty. The proxy can identify very similar commitment decisions as the one obtained by solving the original problem. Its computational performance, though, is close to that of a deterministic unit commitment problem.The third research work in this thesis develops a computational tractable security-constrained stochastic unit commitment model to determine the optimal reserve level. The proposed model characterizes the uncertainty of renewable generation and demand via scenarios and enforces security constraints per scenario to ensure an uneventful operation in case of failure of key thermal units. Once a contingency occurs, the reserve obtained by the proposed model can be effectively deployed to prevent unserved energy. The fourth research work in this thesis designs two metrics to rank committed units according to the risk their failures bring to the system operation. Considering an extreme scenario, the first metric provides multiple ranking criteria. The second metric ranks units based on the system flexibility reduction once the considered unit fails. Once units are committed, those two metrics make the system operator aware of next-day operation risks.

Sustainable production of clean energy carriers and chemicals have been regarded as promising strategies to meet the increasing global energy demand and mitigate the environmental impact of the fossil fuel utilization. Electrocatalytic water splitting to generate H2 is one of the attractive strategies for carbon-neutral energy needs, but often limited by the high-cost noble metal catalysts or the mediocre performance of nonprecious catalysts. Thus, we developed a series of low-cost electrocatalysts via general surface modification of the first-row transition metals to boost H2 evolution reaction (HER). The N decorated Ni framework only requires 64 mV overpotential to produce 10 mA cm-2 and shows superior activity compared to commercial Pt/C at the current density larger than 50 mA cm-2 in neutral electrolyte.Compared to the small overpotential requirement for HER, the other half reaction of water splitting, O2 evolution reaction (OER) requires higher overpotential because of the sluggish kinetics, which dominantly causes the high energy cost for electrocatalytic water splitting. Therefore, the development of low-cost and competent OER electrocatalysts plays an important role for overall water splitting on large scale. Stainless steel consists of many nonprecious metals that have potential capability for electrocatalytic water splitting. We reported a surface modification strategy to prepare stainless steel-based catalysts, which are active for both HER and OER, significantly simplifying the preparation of catalysts and reducing the complexity and the cost of electrolyzers. On account of the high energy requirement of OER and less economic attractiveness of O2, we introduced a new strategy to couple HER with more thermodynamically favorable organic oxidation reaction to reduce the energy input and produce more valuable chemicals. Biomass stands out in this circumstance, owing to its sustainability, tremendous annual production, and the versatility to produce fuels and value-added chemicals. In particular, we successfully demonstrated that 5-hydroxymethylfurfural (HMF), one of the most important biomass derivatives, can be converted electrochemically to more valuable biopolymer precursors, 2,5-furandicarboxylic acid (FDCA) and 2,5-bis(hydroxymethyl)furan (BHMF). Nevertheless, the poor yield from fructose, instability, and difficult isolation of HMF significantly inhibit its large-scale application. Thus, we reported a two-step process for the direct production of FDCA and BHMF from fructose, bypassing the isolation of HMF. The utilization of flow electrolyzer enabled us to run gram-scale electrolysis with >70% yields vs fructose for both reactions, showing the potential for large-scale applications.Achmatowicz reaction enables the conversion of furfuryl alcohol, another biomass derivative, to hydropyranone, which is highly useful in the synthesis of natural products and pharmaceuticals. The utilization of chemical oxidants in existing methods, such as Br2, NBS, and m-CPBA, are neither economically nor environmentally friendly. Therefore, we developed an electrocatalytic method to drive the Achmatowicz reaction with immobilized Ni-based catalysts under ambient condition. Theoretical calculation results indicate that the specially designed coordination unsaturated Ni sites are critical for the reaction. We further demonstrated the feasibility of the electrosynthesis of furopyranones with hydropyranone acetate and 1,3-diketones via electro-reductive cyclization, expanding the application of furfuryl alcohol and electrosynthesis.

Teaching and learning mathematical functions and graphs pose significant challenges for teachers and students. Students often have difficulty in understanding a functional relationship between two quantities such as distance and time, temperature and precipitation, and gas price and number of gallons. Teaching students to have quantitative thinking about functions can help them understand the rate of change for complicated functions and later succeed in learning Calculus. Traditional educational methods such as static graph images and some learning tools usually have some limitations. Teaching students the dynamic changes of quantities within the static picture has serious difficulties. Compared to the learning tools, the game-based learning increases interest when students are learning complicated functions. This thesis presents a game-based learning application called Functions Fun, which runs on iPad tablet computers. The game is created to teach / learn the following functions: Linear, Quadratic, Exponential, Logarithmic, Trigonometric, and Polynomial with degrees over four. Each function is covered under a game level. The game setting is a jungle environment. Each game level has its own scene, challenging the player to take an action while teaching a function and its graphical representation. Functions Fun enables students to play and learn functions and graphs in a more effective and entertaining manner.

Fuel cells are electrochemical devices that combine hydrogen and oxygen from air to produce electric current, with water and heat as the main co-products. The management of liquid water from either the internal chemical reactions or externally humidified reactants is an important design consideration for proton exchange membrane (PEM) fuel cells because of the effects on both cell performance and durability. To achieve proper water management, significant effort has been devoted to developing new fuel cell materials, hardware designs, and appropriate stack operating conditions. However, water management in the region of the channel-to-manifold interfaces has received limited attention. This region covers the ends of the bipolar plate from where liquid water exits the active area to the entrance of the stack exhaust manifolds where excess reactant flows from individual cells are combined and leave the stack. For practical applications, especially in the anode flow field, there is a small driving force to expel liquid water in this region. Under severe operating conditions such as freezing temperatures, the buildup of water may cause channel-scale blockage. This work investigated the water management of PEM fuel cells in the flow field by both ex-situ experiments and in-situ neutron imaging technique to provide a comprehensive two-phase transport model and propose a water mitigation strategy by flow field surface modification method. The results demonstrate the effects of small variations in cell temperature on water accumulation, which translate into significant changes in cell voltage under some conditions. This water can also influence the pressure drop across both anode and cathode flow fields, and it was found that a small amount of water flow can significantly affect the differential pressure, but further increases in water flux appeared to have an incrementally smaller influence. Additionally, the ex-situ experiments also investigated the water distribution of the inlet non-active area, active area, and outlet non-active area, which confirmed the significance of water management in the channel-to-manifold region. A new empirical correlation was developed to characterize the variation of two-phase friction multiplier (i.e., ratio of two-phase to single-phase ΔP) with gas and liquid flow rates. In cases where water accumulates in the non-active cell region downstream of the active area, it was determined that hydrophilic bipolar plate coating was effective in reducing or eliminating full-channel water blockages, thus minimizing the start-up time and energy under freezing conditions. The novel research contributions from this part of the dissertation research include: • Assessed PEM fuel cell water management behavior in the low non-freezing temperature range (200C to 400C), which significantly affects the reliability and durability of PEM fuel systems, but has received very little attention in the literature. • Analyzed water management in the non-active region of the bipolar plate, which not only affects the channel-to-channel water distribution within the fuel cell flow field, but also the cell-to-cell water distribution in a fuel cell stack. This research concluded that water management should focus on the anode side, especially in the outlet channel-to-manifold region. • Quantified the water content in a PEM fuel cell flow field using measurements of channel two-phase flow quality and differential pressure. The two-phase transport model developed in this research is capable of quantifying the water volume in PEM fuel cell flow field, and the results showed good agreement with neutron imaging data. • Evaluated the water mitigation effectiveness of PEM fuel cell for various surface energy modification locations, and concluded for the first time that only one hydrophilic coated channel in the anode channel-to-manifold transition could substantially facilitate the fuel cell cold start-up process. In addition, there is a significant global activity in assessing and optimizing distributed energy systems in so-called “microgrid” architectures, which in principle enable operation completely independent of the primary electrical grid. A shortcoming of such an approach is that many renewable energy systems are intermittent by nature, and thus supply and demand are often out of phase. This necessitates the implementation of energy storage, but few options exist for cost-effective, large-scale storage. One attractive alternative is to use hydrogen as an energy storage medium, because it offers the possibility for storage at relatively high volumetric density, and hydrogen is readily utilized in various energy applications of immediate interest in large product distribution centers. The dissertation work explored the economic impact of PEM fuel cell material handling equipment (MHE) with comparison to the conventional lead acid battery MHE and the emerging lithium-ion battery MHE. Using data obtained directly from large product distribution centers, it was determined that fuel cells are the low-cost option in installations with large MHE vehicle fleets, multi-shift facilities, and relatively high grid electricity costs. The novel contributions of this analysis stem from it being the first to consider lithium-ion batteries with lead acid batteries and fuel cells as competing MHE propulsion technologies. Moreover, it is the only known study to date to account for the time value of money in the economic analysis, and to consider the target facility fleet configuration using data acquired directly from large product distribution centers in various U.S. locations.

In superiority "exploratory" Phase II clinical trials, we often compare the efficacy of several doses of an experimental product versus a control group (often a placebo). We then use the results from the Phase II to design the subsequent confirmatory superiority Phase III trial, and we use statistical methods in the Phase III trial to demonstrate that the "best" (most efficacious) dose selected from the Phase II study is superior to the control. The two phases are usually separate and independent: the Phase III trial does not incorporate patient data from Phase II except, again, in designing the Phase III trial. If we can combine data from the two phases into one design and use data from both phases to assess efficacy of the most efficacious dose of the experimental treatment versus control, we can potentially shorten the overall time of clinical development by reducing the overall sample size across Phase II and Phase III combined. This kind of design is the so-called Adaptive Seamless Designs (ASD). In this dissertation, we first review two commonly used combination approaches for the adaptive seamless designs. These approaches combine the stagewise p-values and apply the closed testing procedure to control the familywise error rate at the nominal level. Due to their complexity in both understanding and implementation, we propose an approach that uses a standard statistical test to compare treatments on the endpoint at the final analysis; we derive the distribution of the final test statistic and the critical value required to maintain Type I error rate at the nominal level Our simulation studies show our approach is comparable to the combination approaches in terms of Type I error rate and power. An extension to Denne's sample size re-estimation method is applied in order to estimate the final Phase III sample size required to maintain desired power, conditioned on Phase II results, when using our proposed adaptive seamless design statistical test. Simulation results demonstrate that the type I error rate and power are maintained at the desired level.

Lateral ankle sprains most frequently occurs during sports. Individuals who experienced a first time ankle sprain had a high reoccurrence rate and residual symptoms and functional instability leading to chronic ankle instability (CAI). The purpose of this study was to investigate kinematic and kinetic differences between CAI individuals and healthy subjects in single-leg drop landing on a flat surface, an inverted surface and a combined surface of inversion and plantarflexion. A total of 17 subjects (6 subjects with chronic ankle instability, 11 healthy subjects) performed five trails in each of four dynamic movement conditions of drop landing from a height of 30 cm onto a force plat form: double leg landing, single-leg drop landing on flat surface, inversion surface of 25 degrees and combined surfaces of 25 degrees of inversion and 25 degrees of plantarflexion. A nine-camera motion analysis system was used to capture the movement of dynamic testing. A 2 × 4 (ankle stability × surfaces) repeated measures ANOVA was used to evaluate the variables for dynamic testing (p

Accurate real-time estimation of water demands in water distribution system (WDS) plays animportant role in accurate estimation of the system behavior, especially during emergency events.Traditionally, water demands are estimated based on population densities, customer billingrecords or census data. However, these methods are not appropriate for real-time modeling.This study describes and tests a Predictor-Corrector methodology for real-time WDS waterdemand estimation. Current water demands are predicted by a reasonable time-series model.Improved estimation results are generated by correcting the prediction using real-timemeasurements (i.e., nodal heads or pipe flow rates). Extended Kalman filter algorithm is used tooperate the prediction and correction process.Experiments to test the performance of the algorithm are carried out. The approach is applied to asample water distribution system comprised of 97 nodes and 119 pipes. The test resultsdemonstrate the impacts of measurement accuracy, sampling design and demand model forecasterror on water demand estimation.

Travel demand forecasting has been a key component of long range planning at Metropolitan Planning Organizations (MPOs) in the United States. Research advancements have led to incorporating transportation accessibility into household and business location choice analysis and forecasting. The dynamic feedback effects between transportation and land use have been studied using accessibility measures with mixed results. This dissertation examines multiple aspects of accessibility and their effects on residential location choice. First, while accessibility has been dichotomized into local and regional accessibility, this study suggests that a mid-range accessibility may have an independent and statistically significant effect on residential location choice. Second, accessibility metrics have traditionally been indifferent to the clustering of destinations. This dissertation tests the idea that, in addition to amount of activities, clustering of activities also contributes to accessibility. Models that explicitly incorporate clustering into measures of accessibility may show stronger explanatory power in predicting residential location choice than models that do not incorporate clustering into accessibility measures. Third, this study compares the effects of place-based accessibility measures and personal commute time on residential location choice. Finally, this dissertation develops alternative models for analyzing residential location choice in regard to accessibility for various socio-economic groups of population, particularly by race and ethnicity, as well as alternative models at three levels of geographic scale, which are metropolitan region, county, and city, for assessing the effects of scale on accessibility. Research hypotheses in this dissertation are tested using multinomial logit models estimated for Detroit metropolitan area based on data from 2004 to 2010. The results show that local, mid-range, and regional accessibility measures affect residential location choice significantly, while the effects of clustering need further study. Individual workers’ commute time has the biggest impact on residential location choice. This is found to be true at multiple geographic levels in Detroit region. The purpose of the study is to contribute to understanding the effects of accessibility in residential location choice, developing innovative tools for measuring accessibility that incorporate clustering and at multiple geographic scales, improving land use and transportation modeling practice, and eventually helping development of land use and transportation policies.