Methods for optimal design of different means of control aredeveloped in this thesis. The main purpose is to maintain thelaminar flow on wings at a chord Reynolds number beyond what isusually transitional or turbulent. Linear stability analysis isused to compute the exponential amplification of infinitesimaldisturbances, which can be used to predict the location oflaminar-turbulent transition. The controls are computed usinggradient-based optimization techniques where the aim is tominimize an objective function based upon, or related to, thedisturbance growth. The gradients of the objective functionswith respect to the controls are evaluated from the solutionsof adjoint equations. Sensitivity analysis using the gradients of the disturbancekinetic energy with respect to different periodic forcing showwhere and by what means control is most efficiently made. Theresults are presented for flat plate boundary layer flows withdifferent free stream Mach numbers. A method to compute optimal steady suction distributions tominimize the disturbance kinetic energy is presented for bothincompressible and compressible boundary layer flows. It isshown how to formulate an objective function in order tominimize simultaneously different types of disturbances whichmight exist in two, and three-dimensional boundary layer flows.The problem formulation also includes control by means ofrealistic pressure chambers, and results are presented wherethe method is applied on a swept wing designed for commercialaircraft. Optimal temperature distributions for disturbance controlare presented for flat plate boundary layer flows. It is shownthat the efficiency of the control depends both on the freestream Mach number, and whether the wall downstream of thecontrol domain is insulated, or heat transfer occurs. Shape optimization is presented with the aim of reducing theaerodynamic drag, while maintaining operational properties.Results of optimized airfoils are presented for cases whereboth the disturbance kinetic energy, and wave drag are reducedsimultaneously while lift, and pitch-moment coefficients aswell as the volume are kept at desired values. <b>Keywords:</b>fluid mechanics, laminar-turbulent transition,boundary layer, laminar flow control, natural laminar flow,adjoint equations, optimal control, objective function, PSE,APSE, ABLE, HLFC, eN-method, Euler equations

Industrial facilities are characterized by the significant usage of natural resources and the massive discharge of waste materials. An effective strategy towards the sustainability of industrial processes is the conservation of natural resources through waste reclamation and recycles. Because of the numerous number of design alternatives, systematic procedures must be developed for the effective synthesis and screening of reclamation and recycle options. The objective of this work is to develop systematic and generally applicable procedures for the synthesis, design, and optimization of resource conservation networks. Focus is given to two important applications: material utilities (with water as an example) and spent products (with lube oil as an example). Traditionally, most of the previous research efforts in the area of designing direct-recycle water networks have considered the chemical composition as the basis for process constraints. However, there are many design problems that are not component-based; instead, they are property-based (e.g., pH, density, viscosity, chemical oxygen demand (COD), basic oxygen demand (BOD), toxicity). Additionally, thermal constraints (e.g., stream temperature) may be required to identify acceptable recycles. In this work, a novel approach is introduced to design material-utility (e.g., water) recycle networks that allows the simultaneous consideration of mass, thermal, and property constraints. Furthermore, the devised approach accounts for the heat of mixing and for the interdependence of properties. An optimization formulation is developed to embed all potential configurations of interest and to model the mass, thermal, and property characteristics of the targeted streams and units. Solution strategies are developed to identify stream allocation and targets for minimum fresh usage and waste discharge. A case study on water management is solved to illustrate the concept of the proposed approach and its computational aspects. Next, a systematic approach is developed for the selection of solvents, solvent blends, and system design in in extraction-based reclamation processes of spent lube oil Property-integration tools are employed for the systematic screening of solvents and solvent blends. The proposed approach identifies the main physical properties that influence solvent(s) performance in extracting additives and contaminants from used lubricating oils (i.e. solubility parameter (delta), viscosity (v), and vapor pressure (p)). The results of the theoretical approach are validated through comparison with experimental data for single solvents and for solvent blends. Next, an optimization formulation is developed and solved to identify system design and extraction solvent(s) by including techno-economic criteria. Two case studies are solved for identification of feasible blends and for the cost optimization of the system.

A computer simulation program was previously developed by the researcher which determines a theoretically optimal movement pattern for the free throw in wheelchair basketball. The purpose of this study was to evaluate the external validity of the optimization program by examining whether the knowledge of the optimal movement pattern facilitates performance of the free throw in wheelchair basketball. </p><p>In a pilot study, four able-bodied players from the Saskatchewan Wheelchair Basketball Mens Team were invited to participate on one occasion. These participants were videotaped shooting free throws to provide knowledge of an expert wheelchair free throw movement pattern. Using video analysis, it was found that the release conditions used by this group were very similar to those predicted to be optimal. This lent support to the predicted optimal movement pattern being an actual optimal movement pattern for the free throw in wheelchair basketball. In the primary study, thirty-three able-bodied male participants were randomly assigned to three groups: a no-feedback group; a video-feedback group; and an optimal pattern feedback group. The participants performed wheelchair basketball free throw training for three days over one week. The no-feedback group simply shot free throws from a wheelchair, whereas the video-feedback group viewed video of their previous free throws, and the optimal pattern group viewed video of their previous free throws with an optimal free throw pattern superimposed. The participants also completed a pretest one week before and a retention test one week after the training period. </p> <p>A repeated measures ANOVA was used to test for significant differences between the three training groups in free throw success in wheelchair basketball over each testing occasion. The statistical analyses indicated that there were no differences in free throw success between the group that had knowledge of their personalized optimal movement pattern when compared to the groups that received either no-feedback or video-feedback (p<0.05). </p> <p>Video analysis revealed that the wheelchair free throw movement pattern of participants in the optimal pattern group changed substantially from the pretest to the post-test. This suggests that the participants in the optimal pattern group were making progress towards their optimal movement patterns, but had not yet mastered the movement pattern.

Over the past decade, the constraint-based approach to metabolic modeling has become an important tool for understanding and controlling biology. Unfortunately, the application of this novel approach to systems biology in biotechnology has been limited by three significant technical issues: existing metabolic modeling methods cannot completely model the overflow metabolism, cannot model the metabolism of microbial communities, and cannot design strains optimized for productivity and titer. Three computational methods – the Flux Balance Analysis with Membrane Economics (FBAME) method, the Dynamic Multi-species Metabolic Modeling (DyMMM) framework, and the Dynamic Strain Scanning Optimization (DySScO) strategy – have been developed to resolve these issues respectively. First, the FBAME method, which adopts the membrane occupancy limitation hypothesis, was used to explain and predict the phenomenon of overflow metabolism, an important metabolic phenomenon in industrial fermentation, in Escherichia coli. Then, the DyMMM framework was used to investigate the community metabolism during uranium bioremediation, and demonstrated that the simultaneous addition of acetate and Fe(III) may be a theoretically viable uranium bioremediation strategy. Lastly, the DySScO strategy, which combines the DyMMM framework with existing strain design algorithms, was used to design commodity-chemical producing E. coli optimized for a balanced product yield, titer, and volumetric productivity. These novel computational methods allow for broader applications of constraint-based metabolic models in biotechnology settings.

Hard combinatorial optimization (CO) problems pose challenges to traditional algorithmic solutions. The search space usually contains a large number of local optimal points and the computational cost to reach a global optimum may be too high for practical use. In this work, we conduct a comparative study of several state-of-the-art metaheuristic algorithms for hard CO problems solving. Our study is motivated by an industrial application called the Fertilizer Blends Optimization. We focus our study on a number of local search metaheuristics and analyze their performance in terms of both runtime efficiency and solution quality. We show that local search granularity (move step size) and the downhill move probability are two major factors that affect algorithm performance, and we demonstrate how experimental tuning work can be applied to obtain good performance of the algorithms. <p>Our empirical result suggests that the well-known Simulated Annealing (SA) algorithm showed the best performance on the fertilizer problem. The simple Iterated Improvement Algorithm (IIA) also performed surprisingly well by combining strict uphill move and random neighborhood selection. A novel approach, called Delivery Network Model (DNM) algorithm, was also shown to be competitive, but it has the disadvantage of being very sensitive to local search granularity. The constructive local search method (GRASP), which combines heuristic space sampling and local search, outperformed IIA without a construction phase; however, the improvement in performance is limited and generally speaking, local search performance is not sensitive to initial search positions in our studied fertilizer problem.

Structural topology optimization methods have existing and been improving theoretically since 1980s; however, in industry, with respect to the certain conditions, proper modification is always desired. This study develops a specific method to utilize topology optimization for gearbox housing design. Gearbox housing maintains the position of the shafts to ensure the precision of gear engagement in all operational states (Naunheimer, et al., 2010). The current housing design processing used in Vicura AB, a Swedish powertrain company, is able to achieve stiff optimal housing material distribution, but difficult to fulfil gear misalignment requirement. This work overcomes the above shortages to develop a new methodology for gearbox housing topology optimization concerning the gear misalignment as well. The paper is starting with an introduction of the previous method and its defects, followed by a discussion of three possible improvements. Only one of them is feasible and two main difficulties need to be resolved to make it applicable. One of the difficulties is finding a linear assumption of the non-linear components and the other is deriving an approach for topology optimization involving both external forces and non-zero prescribed displacements. The corresponding solutions are described subsequently in detail both theoretically and practically. Then the results by implementing the new method and also the comparison with the results getting from the old method are presented. Finally, a validation of the new method is discussed and the conclusions and comments are given.

Creating controllers for physics-based characters is a long-standing open problem in animation and robotics. Such controllers would have numerous applications while potentially yielding insight into human motion. Creating controllers remains difficult: current approaches are either constrained to track motion capture data, are not robust, or provide limited control over style. This thesis presents an approach to control of physics-based characters based on high-level features of human movement, such as center-of-mass, angular momentum, and end-effector motion. Objective terms are used to control each feature, and are combined via optimization. We show how locomotion can be expressed in terms of a small number of features that control balance and end-effectors. This approach is used to build controllers for biped balancing, jumping, walking, and jogging. These controllers provide numerous benefits: human-like qualities such as arm-swing, heel-off, and hip-shoulder counter-rotation emerge automatically during walking; controllers are robust to changes in body parameters; control parameters apply to intuitive properties; and controller may be mapped onto entirely new bipeds with different topology and mass distribution, without controller modifications. Transitions between multiple types of gaits, including walking, jumping, and jogging, emerge automatically. Controllers can traverse challenging terrain while following high-level user commands at interactive rates. This approach uses no motion capture or off-line optimization process. Although we focus on the challenging case of bipedal locomotion, many other types of controllers stand to benefit from our approach.

Creating controllers for physics-based characters is a long-standing open problem in animation and robotics. Such controllers would have numerous applications while potentially yielding insight into human motion. Creating controllers remains difficult: current approaches are either constrained to track motion capture data, are not robust, or provide limited control over style. This thesis presents an approach to control of physics-based characters based on high-level features of human movement, such as center-of-mass, angular momentum, and end-effector motion. Objective terms are used to control each feature, and are combined via optimization. We show how locomotion can be expressed in terms of a small number of features that control balance and end-effectors. This approach is used to build controllers for biped balancing, jumping, walking, and jogging. These controllers provide numerous benefits: human-like qualities such as arm-swing, heel-off, and hip-shoulder counter-rotation emerge automatically during walking; controllers are robust to changes in body parameters; control parameters apply to intuitive properties; and controller may be mapped onto entirely new bipeds with different topology and mass distribution, without controller modifications. Transitions between multiple types of gaits, including walking, jumping, and jogging, emerge automatically. Controllers can traverse challenging terrain while following high-level user commands at interactive rates. This approach uses no motion capture or off-line optimization process. Although we focus on the challenging case of bipedal locomotion, many other types of controllers stand to benefit from our approach.

Informally, a rooted map is a topologically pointed embedding of a graph in a surface. This thesis examines two problems in the enumerative theory of rooted maps. The b-Conjecture, due to Goulden and Jackson, predicts that structural similarities between the generating series for rooted orientable maps with respect to vertex-degree sequence, face-degree sequence, and number of edges, and the corresponding generating series for rooted locally orientable maps, can be explained by a unified enumerative theory. Both series specialize M(x,y,z;b), a series defined algebraically in terms of Jack symmetric functions, and the unified theory should be based on the existence of an appropriate integer valued invariant of rooted maps with respect to which M(x,y,z;b) is the generating series for locally orientable maps. The conjectured invariant should take the value zero when evaluated on orientable maps, and should take positive values when evaluated on non-orientable maps, but since it must also depend on rooting, it cannot be directly related to genus. A new family of candidate invariants, η, is described recursively in terms of root-edge deletion. Both the generating series for rooted maps with respect to η and an appropriate specialization of M satisfy the same differential equation with a unique solution. This shows that η gives the appropriate enumerative theory when vertex degrees are ignored, which is precisely the setting required by Goulden, Harer, and Jackson for an application to algebraic geometry. A functional equation satisfied by M and the existence of a bijection between rooted maps on the torus and a restricted set of rooted maps on the Klein bottle show that η has additional structural properties that are required of the conjectured invariant. The q-Conjecture, due to Jackson and Visentin, posits a natural combinatorial explanation, for a functional relationship between a generating series for rooted orientable maps and the corresponding generating series for 4-regular rooted orientable maps. The explanation should take the form of a bijection, ϕ, between appropriately decorated rooted orientable maps and 4-regular rooted orientable maps, and its restriction to undecorated maps is expected to be related to the medial construction. Previous attempts to identify ϕ have suffered from the fact that the existing derivations of the functional relationship involve inherently non-combinatorial steps, but the techniques used to analyze η suggest the possibility of a new derivation of the relationship that may be more suitable to combinatorial analysis. An examination of automorphisms that must be induced by ϕ gives evidence for a refinement of the functional relationship, and this leads to a more combinatorially refined conjecture. The refined conjecture is then reformulated algebraically so that its predictions can be tested numerically.

Throughout a very long piping network crude oil in Saudi Arabia is sent to Gas Oil Separation Plant called GOSP. The main objectives of the GOSP are: - Separation of the associated gas through pressure drop in two series stages one to 120 psig and the other to 50 psig. - Separation of water by gravity separators called High Pressure Production Trap (HPPT), Dehydrator, Desalter and Water Oil Separator (WOSEP). - Reducing salt concentration to less than 10 PTB utilizing wash water and demulsifier. During the desalting process, the challenge is to overcome the existence of an emulsion layer at the interface between oil and water. In petroleum industry normally emulsions encountered are some kind of water droplets dispersed in a continuous phase of oil. In crude oil emulsions, emulsifying agents are present at the oil-water interface, hindering this coalescence process. Such agents include scale and clay particles, added chemicals or indigenous crude oil components like asphaltenes, resins, waxes and naphthenic acids. Many techniques made available to gas oil separation plant operators to minimize the effect of tight emulsions. These techniques include injection of demulsifier, increasing oil temperature, gravity separation in large vessels with high retention time as well as electrostatic voltage. From experience and studies these variables have been already optimized to a good extent; however, from the believe that knowledge never stop, this study is conducted targeting enhancing the demulsifier control and optimizing the wash water rate. The objective of this study is to design an Artificial Neural Network (ANN) trained on data set to cover wide operating range of all parameters effecting demulsifier dosage. This network will be used to work as a control black box inside the controller in which all effecting parameters are inputs and the demulsifier dosage is the controller output. Testing this control scheme showed an effective reduction in demulsifier consumption rate compared to the existing linear method. Results also, showed that the existing control strategy is highly conservative to prevent the salt from exceeding the limit. The generated function from the ANN was used also to optimize the amount of fresh water added to wash the salty crude oil. Finally, another ANN was developed to generate an online estimate of the salt content in the produced oil.