Optimal boundary geometry in an elasticity problem: a systematic adjoint approach

In different problems of Elasticity the definition of the optimal gcometry of the boundary, according to a given objective function, is an issue of great interest. Finding the shape of a hole in the middle of a plate subjected to an arbitrary loading such that the stresses along the hole minimizes some functional or the optimal middle curved concrete vault for a tunnel along which a uniform minimum compression are two typical examples. In these two examples the objective functional depends on the geometry of the boundary that can be either a curve (in case of 2D problems) or a surface boundary (in 3D problems). Typically, optimization is achieved by means of an iterative process which requires the computation of gradients of the objective function with respect to design variables. Gradients can by computed in a variety of ways, although adjoint methods either continuous or discrete ones are the more efficient ones when they are applied in different technical branches. In this paper the adjoint continuous method is introduced in a systematic way to this type of problems and an illustrative simple example, namely the finding of an optimal shape tunnel vault immersed in a linearly elastic terrain, is presented.

Optimization of prestressed concrete bridge decks

In this paper a summary of the methods presently used for optimization of prestressed concrete bridge decks is given. By means of linear optimization the sizes of the prestressing cables with a given fixed geometry are obtained. This simple procedure of linear optimization is also used to obtain the ‘best’ cable profile, by combining a series of feasible cable profiles. The results are compared with the ones obtained by other researchers. A step ahead in the field of optimization of prestressed bridge decks is the simultaneous search of the geometry and size of the prestressing cables. A non-linear programming for optimization is used, namely, ‘the steepest gradient method’. The results obtained are compared with the ones computed previously by means of linear programming techniques. Finally, the general problem of structural optimization is considered. This problem consists in finding the sizes and geometries of the prestressing cables as well as the longitudinal variation of the concrete section.

Development of Shape-Optimization Tools for the Aerodynamic Design of Turbomachinery Blades

The aim of this work is to develop an automated tool for the optimization of turbomachinery blades founded on an evolutionary strategy. This optimization scheme will serve to deal with supersonic blades cascades for application to Organic Rankine Cycle (ORC) turbines. The blade geometry is defined using parameterization techniques based on B-Splines curves, that allow to have a local control of the shape. The location in space of the control points of the B-Spline curve define the design variables of the optimization problem. In the present work, the performance of the blade shape is assessed by means of fully-turbulent flow simulations performed with a CFD package, in which a look-up table method is applied to ensure an accurate thermodynamic treatment. The solver is set along with the optimization tool to determine the optimal shape of the blade. As only blade-to-blade effects are of interest in this study, quasi-3D calculations are performed, and a single-objective evolutionary strategy is applied to the optimization. As a result, a non-intrusive tool, with no need for gradients definition, is developed. The computational cost is reduced by the use of surrogate models. A Gaussian interpolation scheme (Kriging model) is applied for the estimated n-dimensional function, and a surrogate-based local optimization strategy is proved to yield an accurate way for optimization. In particular, the present optimization scheme has been applied to the re-design of a supersonic stator cascade of an axial-flow turbine. In this design exercise very strong shock waves are generated in the rear blade suction side and shock-boundary layer interaction mechanisms occur. A significant efficiency improvement as a consequence of a more uniform flow at the blade outlet section of the stator is achieved. This is also expected to provide beneficial effects on the design of a subsequent downstream rotor. The method provides an improvement to gradient-based methods and an optimized blade geometry is easily achieved using the genetic algorithm.

Aerodynamic Optimization of a High-speed Train using Genetic Algorithms

Genetic algorithms (GA) have been used for the minimization of the aerodynamic drag of a train subject to front wind. The significant importance of the external aerodynamic drag on the total resistance a train experiments as the cruise speed is increased highlights the interest of this study. A complete description of the methodology required for this optimization method is introduced here, where the parameterization of the geometry to be optimized and the metamodel used to speed up the optimization process are detailed. A reduction of about a 25% of the initial aerodynamic drag is obtained in this study, what confirms GA as a proper method for this optimization problem. The evolution of the nose shape is consistent with the literature. The advantage of using metamodels is stressed thanks to the information of the whole design space extracted from it. The influence of each design variable on the objective function is analyzed by means of an ANOVA test.

Real time optimization in chemical process: evaluation of strategies, improvements and industrial application.

The increasing economic competition drives the industry to implement tools that improve their processes efficiencies. The process automation is one of these tools, and the Real Time Optimization (RTO) is an automation methodology that considers economic aspects to update the process control in accordance with market prices and disturbances. Basically, RTO uses a steady-state phenomenological model to predict the process behavior, and then, optimizes an economic objective function subject to this model. Although largely implemented in industry, there is not a general agreement about the benefits of implementing RTO due to some limitations discussed in the present work: structural plant/model mismatch, identifiability issues and low frequency of set points update. Some alternative RTO approaches have been proposed in literature to handle the problem of structural plant/model mismatch. However, there is not a sensible comparison evaluating the scope and limitations of these RTO approaches under different aspects. For this reason, the classical two-step method is compared to more recently derivative-based methods (Modifier Adaptation, Integrated System Optimization and Parameter estimation, and Sufficient Conditions of Feasibility and Optimality) using a Monte Carlo methodology. The results of this comparison show that the classical RTO method is consistent, providing a model flexible enough to represent the process topology, a parameter estimation method appropriate to handle measurement noise characteristics and a method to improve the sample information quality. At each iteration, the RTO methodology updates some key parameter of the model, where it is possible to observe identifiability issues caused by lack of measurements and measurement noise, resulting in bad prediction ability. Therefore, four different parameter estimation approaches (Rotational Discrimination, Automatic Selection and Parameter estimation, Reparametrization via Differential Geometry and classical nonlinear Least Square) are evaluated with respect to their prediction accuracy, robustness and speed. The results show that the Rotational Discrimination method is the most suitable to be implemented in a RTO framework, since it requires less a priori information, it is simple to be implemented and avoid the overfitting caused by the Least Square method. The third RTO drawback discussed in the present thesis is the low frequency of set points update, this problem increases the period in which the process operates at suboptimum conditions. An alternative to handle this problem is proposed in this thesis, by integrating the classic RTO and Self-Optimizing control (SOC) using a new Model Predictive Control strategy. The new approach demonstrates that it is possible to reduce the problem of low frequency of set points updates, improving the economic performance. Finally, the practical aspects of the RTO implementation are carried out in an industrial case study, a Vapor Recompression Distillation (VRD) process located in Paulínea refinery from Petrobras. The conclusions of this study suggest that the model parameters are successfully estimated by the Rotational Discrimination method; the RTO is able to improve the process profit in about 3%, equivalent to 2 million dollars per year; and the integration of SOC and RTO may be an interesting control alternative for the VRD process.

A theoretical comparison of two optimization methods for radiofrequency drive schemes in high frequency MRI resonators

In this paper, numerical simulations are used in an attempt to find optimal Source profiles for high frequency radiofrequency (RF) volume coils. Biologically loaded, shielded/unshielded circular and elliptical birdcage coils operating at 170 MHz, 300 MHz and 470 MHz are modelled using the FDTD method for both 2D and 3D cases. Taking advantage of the fact that some aspects of the electromagnetic system are linear, two approaches have been proposed for the determination of the drives for individual elements in the RF resonator. The first method is an iterative optimization technique with a kernel for the evaluation of RF fields inside an imaging plane of a human head model using pre-characterized sensitivity profiles of the individual rungs of a resonator; the second method is a regularization-based technique. In the second approach, a sensitivity matrix is explicitly constructed and a regularization procedure is employed to solve the ill-posed problem. Test simulations show that both methods can improve the B-1-field homogeneity in both focused and non-focused scenarios. While the regularization-based method is more efficient, the first optimization method is more flexible as it can take into account other issues such as controlling SAR or reshaping the resonator structures. It is hoped that these schemes and their extensions will be useful for the determination of multi-element RF drives in a variety of applications.

Laser tracker position optimization

This article presents a laser tracker position optimization code based on the tracker uncertainty model developed by the National Physical Laboratory (NPL). The code is able to find the optimal tracker positions for generic measurements involving one or a network of many trackers, and an arbitrary set of targets. The optimization is performed using pattern search or optionally, genetic algorithm (GA) or particle swarm optimization (PSO). Different objective function weightings for the uncertainties of individual points, distance uncertainties between point pairs, and the angular uncertainties between three points can be defined. Constraints for tracker position limits and minimum measurement distances have also been implemented. Furthermore, position optimization taking into account of lines-of-sight (LOS) within complex CAD geometry have also been demonstrated. The code is simple to use and can be a valuable measurement planning tool.

Laser tracker position optimization

This article presents a laser tracker position optimization code based on the tracker uncertainty model developed by the National Physical Laboratory (NPL). The code is able to find the optimal tracker positions for generic measurements involving one or a network of many trackers, and an arbitrary set of targets. The optimization is performed using pattern search or optionally, genetic algorithm (GA) or particle swarm optimization (PSO). Different objective function weightings for the uncertainties of individual points, distance uncertainties between point pairs, and the angular uncertainties between three points can be defined. Constraints for tracker position limits and minimum measurement distances have also been implemented. Furthermore, position optimization taking into account of lines-of-sight (LOS) within complex CAD geometry have also been demonstrated. The code is simple to use and can be a valuable measurement planning tool.

PC Points and their Application to Vector Optimization

AMS subject classification: 90C29, 90C48

Learning from Geometry

Subspaces and manifolds are two powerful models for high dimensional signals. Subspaces model linear correlation and are a good fit to signals generated by physical systems, such as frontal images of human faces and multiple sources impinging at an antenna array. Manifolds model sources that are not linearly correlated, but where signals are determined by a small number of parameters. Examples are images of human faces under different poses or expressions, and handwritten digits with varying styles. However, there will always be some degree of model mismatch between the subspace or manifold model and the true statistics of the source. This dissertation exploits subspace and manifold models as prior information in various signal processing and machine learning tasks.

A near-low-rank Gaussian mixture model measures proximity to a union of linear or affine subspaces. This simple model can effectively capture the signal distribution when each class is near a subspace. This dissertation studies how the pairwise geometry between these subspaces affects classification performance. When model mismatch is vanishingly small, the probability of misclassification is determined by the product of the sines of the principal angles between subspaces. When the model mismatch is more significant, the probability of misclassification is determined by the sum of the squares of the sines of the principal angles. Reliability of classification is derived in terms of the distribution of signal energy across principal vectors. Larger principal angles lead to smaller classification error, motivating a linear transform that optimizes principal angles. This linear transformation, termed TRAIT, also preserves some specific features in each class, being complementary to a recently developed Low Rank Transform (LRT). Moreover, when the model mismatch is more significant, TRAIT shows superior performance compared to LRT.

The manifold model enforces a constraint on the freedom of data variation. Learning features that are robust to data variation is very important, especially when the size of the training set is small. A learning machine with large numbers of parameters, e.g., deep neural network, can well describe a very complicated data distribution. However, it is also more likely to be sensitive to small perturbations of the data, and to suffer from suffer from degraded performance when generalizing to unseen (test) data.

From the perspective of complexity of function classes, such a learning machine has a huge capacity (complexity), which tends to overfit. The manifold model provides us with a way of regularizing the learning machine, so as to reduce the generalization error, therefore mitigate overfiting. Two different overfiting-preventing approaches are proposed, one from the perspective of data variation, the other from capacity/complexity control. In the first approach, the learning machine is encouraged to make decisions that vary smoothly for data points in local neighborhoods on the manifold. In the second approach, a graph adjacency matrix is derived for the manifold, and the learned features are encouraged to be aligned with the principal components of this adjacency matrix. Experimental results on benchmark datasets are demonstrated, showing an obvious advantage of the proposed approaches when the training set is small.

Stochastic optimization makes it possible to track a slowly varying subspace underlying streaming data. By approximating local neighborhoods using affine subspaces, a slowly varying manifold can be efficiently tracked as well, even with corrupted and noisy data. The more the local neighborhoods, the better the approximation, but the higher the computational complexity. A multiscale approximation scheme is proposed, where the local approximating subspaces are organized in a tree structure. Splitting and merging of the tree nodes then allows efficient control of the number of neighbourhoods. Deviation (of each datum) from the learned model is estimated, yielding a series of statistics for anomaly detection. This framework extends the classical {\em changepoint detection} technique, which only works for one dimensional signals. Simulations and experiments highlight the robustness and efficacy of the proposed approach in detecting an abrupt change in an otherwise slowly varying low-dimensional manifold.

Parametric CAD model based shape optimization using adjoint functions

Adjoint methods have proven to be an efficient way of calculating the gradient of an objective function with respect to a shape parameter for optimisation, with a computational cost nearly independent of the number of the design variables [1]. The approach in this paper links the adjoint surface sensitivities (gradient of objective function with respect to the surface movement) with the parametric design velocities (movement of the surface due to a CAD parameter perturbation) in order to compute the gradient of the objective function with respect to CAD variables.
For a successful implementation of shape optimization strategies in practical industrial cases, the choice of design variables or parameterisation scheme used for the model to be optimized plays a vital role. Where the goal is to base the optimization on a CAD model the choices are to use a NURBS geometry generated from CAD modelling software, where the position of the NURBS control points are the optimisation variables [2] or to use the feature based CAD model with all of the construction history to preserve the design intent [3]. The main advantage of using the feature based model is that the optimized model produced can be directly used for the downstream applications including manufacturing and process planning.
This paper presents an approach for optimization based on the feature based CAD model, which uses CAD parameters defining the features in the model geometry as the design variables. In order to capture the CAD surface movement with respect to the change in design variable, the “Parametric Design Velocity” is calculated, which is defined as the movement of the CAD model boundary in the normal direction due to a change in the parameter value.
The approach presented here for calculating the design velocities represents an advancement in terms of capability and robustness of that described by Robinson et al. [3]. The process can be easily integrated to most industrial optimisation workflows and is immune to the topology and labelling issues highlighted by other CAD based optimisation processes. It considers every continuous (“real value”) parameter type as an optimisation variable, and it can be adapted to work with any CAD modelling software, as long as it has an API which provides access to the values of the parameters which control the model shape and allows the model geometry to be exported. To calculate the movement of the boundary the methodology employs finite differences on the shape of the 3D CAD models before and after the parameter perturbation. The implementation procedure includes calculating the geometrical movement along a normal direction between two discrete representations of the original and perturbed geometry respectively. Parametric design velocities can then be directly linked with adjoint surface sensitivities to extract the gradients to use in a gradient-based optimization algorithm.
The optimisation of a flow optimisation problem is presented, in which the power dissipation of the flow in an automotive air duct is to be reduced by changing the parameters of the CAD geometry created in CATIA V5. The flow sensitivities are computed with the continuous adjoint method for a laminar and turbulent flow [4] and are combined with the parametric design velocities to compute the cost function gradients. A line-search algorithm is then used to update the design variables and proceed further with optimisation process.

Coupled CFD Shape Optimization for aerodynamic profiles

The present document deals with the optimization of shape of aerodynamic profiles -- The objective is to reduce the drag coefficient on a given profile without penalising the lift coefficient -- A set of control points defining the geometry are passed and parameterized as a B-Spline curve -- These points are modified automatically by means of CFD analysis -- A given shape is defined by an user and a valid volumetric CFD domain is constructed from this planar data and a set of user-defined parameters -- The construction process involves the usage of 2D and 3D meshing algorithms that were coupled into own- code -- The volume of air surrounding the airfoil and mesh quality are also parametrically defined -- Some standard NACA profiles were used by obtaining first its control points in order to test the algorithm -- Navier-Stokes equations were solved for turbulent, steady-state ow of compressible uids using the k-epsilon model and SIMPLE algorithm -- In order to obtain data for the optimization process an utility to extract drag and lift data from the CFD simulation was added -- After a simulation is run drag and lift data are passed to the optimization process -- A gradient-based method using the steepest descent was implemented in order to define the magnitude and direction of the displacement of each control point -- The control points and other parameters defined as the design variables are iteratively modified in order to achieve an optimum -- Preliminary results on conceptual examples show a decrease in drag and a change in geometry that obeys to aerodynamic behavior principles

Introduction to geodesics in sub-Riemannian geometry

International audience

Surface acoustic wave streaming in a PDMS microfluidic system: effect of frequency and fluid geometry & A remote ultrasonic glucose sensor

This thesis describes two separate projects. The first is a theoretical and experimental investigation of surface acoustic wave streaming in microfluidics. The second is the development of a novel acoustic glucose sensor. A separate abstract is given for each here. Optimization of acoustic streaming in microfluidic channels by SAWs Surface Acoustic Waves, (SAWs) actuated on flat piezoelectric substrates constitute a convenient and versatile tool for microfluidic manipulation due to the easy and versatile interfacing with microfluidic droplets and channels. The acoustic streaming effect can be exploited to drive fast streaming and pumping of fluids in microchannels and droplets (Shilton et al. 2014; Schmid et al. 2011), as well as size dependant sorting of particles in centrifugal flows and vortices (Franke et al. 2009; Rogers et al. 2010). Although the theory describing acoustic streaming by SAWs is well understood, very little attention has been paid to the optimisation of SAW streaming by the correct selection of frequency. In this thesis a finite element simulation of the fluid streaming in a microfluidic chamber due to a SAW beam was constructed and verified against micro-PIV measurements of the fluid flow in a fabricated device. It was found that there is an optimum frequency that generates the fastest streaming dependent on the height and width of the chamber. It is hoped this will serve as a design tool for those who want to optimally match SAW frequency with a particular microfluidic design. An acoustic glucose sensor Diabetes mellitus is a disease characterised by an inability to properly regulate blood glucose levels. In order to keep glucose levels under control some diabetics require regular injections of insulin. Continuous monitoring of glucose has been demonstrated to improve the management of diabetes (Zick et al. 2007; Heinemann & DeVries 2014), however there is a low patient uptake of continuous glucose monitoring systems due to the invasive nature of the current technology (Ramchandani et al. 2011). In this thesis a novel way of monitoring glucose levels is proposed which would use ultrasonic waves to ‘read’ a subcutaneous glucose sensitive-implant, which is only minimally invasive. The implant is an acoustic analogy of a Bragg stack with a ‘defect’ layer that acts as the sensing layer. A numerical study was performed on how the physical changes in the sensing layer can be deduced by monitoring the reflection amplitude spectrum of ultrasonic waves reflected from the implant. Coupled modes between the skin and the sensing layer were found to be a potential source of error and drift in the measurement. It was found that by increasing the number of layers in the stack that this could be minimized. A laboratory proof of concept system was developed using a glucose sensitive hydrogel as the sensing layer. It was possible to monitor the changing thickness and speed of sound of the hydrogel due to physiological relevant changes in glucose concentration.

Optimization of Pattern Matching Algorithms for Multi- and Many-Core Platforms

Image and video compression play a major role in the world today, allowing the storage and transmission of large multimedia content volumes. However, the processing of this information requires high computational resources, hence the improvement of the computational performance of these compression algorithms is very important. The Multidimensional Multiscale Parser (MMP) is a pattern-matching-based compression algorithm for multimedia contents, namely images, achieving high compression ratios, maintaining good image quality, Rodrigues et al. [2008]. However, in comparison with other existing algorithms, this algorithm takes some time to execute. Therefore, two parallel implementations for GPUs were proposed by Ribeiro [2016] and Silva [2015] in CUDA and OpenCL-GPU, respectively. In this dissertation, to complement the referred work, we propose two parallel versions that run the MMP algorithm in CPU: one resorting to OpenMP and another that converts the existing OpenCL-GPU into OpenCL-CPU. The proposed solutions are able to improve the computational performance of MMP by 3 and 2:7 , respectively. The High Efficiency Video Coding (HEVC/H.265) is the most recent standard for compression of image and video. Its impressive compression performance, makes it a target for many adaptations, particularly for holoscopic image/video processing (or light field). Some of the proposed modifications to encode this new multimedia content are based on geometry-based disparity compensations (SS), developed by Conti et al. [2014], and a Geometric Transformations (GT) module, proposed by Monteiro et al. [2015]. These compression algorithms for holoscopic images based on HEVC present an implementation of specific search for similar micro-images that is more efficient than the one performed by HEVC, but its implementation is considerably slower than HEVC. In order to enable better execution times, we choose to use the OpenCL API as the GPU enabling language in order to increase the module performance. With its most costly setting, we are able to reduce the GT module execution time from 6.9 days to less then 4 hours, effectively attaining a speedup of 45 .