SEARCH · Engineering Papers
Results for “Numerical optimization”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Optimization and Parallelization of the Thermal-Hydraulic Sub-channel Code CTF for High-Fidelity Multi-physics Applications
This paper describes major improvements to the computational infrastructure of the CTF sub-channel code so that full-core sub-channel-resolved simulations can now be performed in much shorter run-times, either in stand-alone mode or as part of coupled-code multi-physics calculations. These improvements support the goals of the Department Of Energy (DOE) Consortium for Advanced Simulations of Light Water (CASL) Energy Innovation Hub to develop high fidelity multi-physics simulation tools for nuclear energy design and analysis. A set of serial code optimizations--including fixing computational inefficiencies, optimizing the numerical approach, and making smarter data storage choices--are first described and shown to reduce both execution time and memory usage by about a factor of ten. Next, a Single Program Multiple Data (SPMD) parallelization strategy targeting distributed memory Multiple Instruction Multiple Data (MIMD) platforms and utilizing domain-decomposition is presented. In this approach, data communication between processors is accomplished by inserting standard MPI calls at strategic points in the code. The domain decomposition approach implemented assigns one MPI process to each fuel assembly, with each domain being represented by its own CTF input file. The creation of CTF input files, both for serial and parallel runs, is also fully automated through use of a pre-processor utility that takes a greatly reduced set of user input over the traditional CTF input file. To run CTF in parallel, two additional libraries are currently needed; MPI, for inter-processor message passing, and the Parallel Extensible Toolkit for Scientific Computation (PETSc), which is leveraged to solve the global pressure matrix in parallel. Results presented include a set of testing and verification calculations and performance tests assessing parallel scaling characteristics up to a full core, sub-channel-resolved model of Watts Bar Unit 1 under hot full-power conditions (193 17x17 assemblies, ~56,000 pins, ~59,000 sub-channels, ~2.8 million thermal-hydraulic (TH) control volumes). Results demonstrate that CTF can now perform full core analysis (not previously possible due to excessively long runtimes and memory requirements) on the order of 20 minutes. This new capability is not only useful to standalone CTF users, but is also being leveraged in support of coupled code multi-physics calculations being done in the CASL program.
Numerical Approaches for Sequential Bayesian Optimal Experimental Design.
Abstract not provided.
Design of Helical Cooling Channel for Muon Collider
Fast muon beam six dimensional (6D) phase space cooling is essential for muon colliders. The Helical Cooling Channel (HCC) uses hydrogen-pressurized RF cavities imbedded in a magnet system with solenoid, helical dipole, and helical quadrupole components that provide the continuous dispersion needed for emittance exchange and effective 6D beam cooling. A series of HCC segments, each with sequentially smaller aperture, higher magnetic field, and higher RF frequency to match the beam size as it is cooled, has been optimized by numerical simulation to achieve a factor of 10{sup 5} emittance reduction in a 300 m long channel with only a 40% loss of beam. Conceptual designs of the hardware required for this HCC system and the status of the RF studies and HTS helical solenoid magnet prototypes are described.
Shape Optimization for Navier-Stokes Equations with Algebraic Turbulence Model: Numerical Analysis and Computation
We study the shape optimization problem for the paper machine headbox which distributes a mixture of water and wood fibers in the paper making process. The aim is to find a shape which a priori ensures the given velocity profile on the outlet part. The mathematical formulation leads to the optimal control problem in which the control variable is the shape of the domain representing the header, the state problem is represented by the generalized Navier-Stokes system with nontrivial boundary conditions. This paper deals with numerical aspects of the problem.
An algorithm for the empirical optimization of antenna arrays
A numerical technique is presented to optimize the performance of arbitrary antenna arrays under realistic conditions. An experimental-computational algorithm is formulated in which n-dimensional minimization methods are applied to measured data obtained from the antenna array. A numerical update formula is used to induce partial derivative information without requiring special perturbations of the array parameters. The algorithm provides a new design for the antenna array, and the method proceeds in an iterative fashion. Test case results are presented showing the effectiveness of the algorithm.
Scaled Aeroelastic Wind Tunnel Model Design with Topology Optimization
The goal of this work is to use numerical aeroelastic topology optimization to design flutter models to be tested in the Transonic Dynamics Tunnel, as part of an envisioned validation experiment for transonic flutter. The topological design develops the structural details of a variable-thickness aluminum plate located at the camber line of a scaled transport aircraft model, in order to satisfy the various conflicting design requirements, including a prescribed flutter boundary, flutter frequency, and static stress levels. Four designs are demonstrated here, where it is shown that the simultaneous design of a cylindrical tip store can help ease the conflict between low stiffness (to allow a wing to flutter in the tunnel envelope) and high strength (to ensure safe structural response). The tip store complicates the geometrical features of the wing, however, and is generally undesirable from the vantage point of a validation experiment.
Stability Analysis of Streaks Induced by Optimized Vortex Generators
Numerical computations are performed to investigate the potential for transition control in an axisymmetric boundary layer via fully realizable, streamwise stationary streaks induced by an azimuthally periodic array of surface mounted vortex generators (VGs). Previous work has shown that suitable streaks of this type can significantly reduce the growth of Mack’s second mode instabilities, but large streak amplitudes can make the flow susceptible to previously absent streak instabilities that can become the leading cause of transition. Here, we use the adjoint capabilities of the SU2 flow solver to optimize the VG shape to maximize the reduction in the growth of second-mode disturbances while also preventing the streak amplitudes from reaching large enough values to precipitate an earlier onset of transition via streak instabilities. The geometry and the freestream flow conditions are selected to match a relevant trajectory lo-cation from the HIFiRE-1 flight experiment. Results show that the optimized VGs can increase the mean streak amplitude by 117% with respect to a manually developed baseline design. The stability of this optimized basic state is analyzed via the plane-marching parabolized stability equations, predicting a fully laminar flow over the entire cone, or equivalently, yielding transition delay of 130% versus the 17% for the baseline VGs.
Stability Analysis of Streaks Induced By Optimized Vortex Generators
Numerical computations are performed to investigate the potential for transition control in an axisymmetric boundary layer via fully realizable, streamwise stationary streaks induced by an azimuthally periodic array of surface mounted vortex generators (VGs). Previous work has shown that suitable streaks of this type can significantly reduce the growth of Mack’s second mode instabilities, but large streak amplitudes can make the flow susceptible to previously absent streak instabilities that can become the leading cause of transition. Here, we use the adjoint capabilities of the SU2 flow solver to optimize the VG shape to maximize the reduction in the growth of second-mode disturbances while also preventing the streak amplitudes from reaching large enough values to precipitate an earlier onset of transition via streak instabilities. The geometry and the freestream flow conditions are selected to match a relevant trajectory lo-cation from the HIFiRE-1 flight experiment. Results show that the optimized VGs can increase the mean streak amplitude by 117% with respect to a manually developed baseline design. The stability of this optimized basic state is analyzed via the plane-marching parabolized stability equations, predicting a fully laminar flow over the entire cone, or equivalently, yielding transition delay of 130% versus the 17% for the baseline VGs.
Optimal discrete-time dynamic output-feedback design - A w-domain approach
An alternative method for optimal digital control design is described in this paper. The method is based on the usage of the w-transform and has many attractive design features. One of these is its immediate connection with frequency loop-shaping techniques that are now popular and effective for multivariable control synthesis in continuous-time domain. Furthermore, any design algorithms originally developed for continuous-time systems can now be immediately extended to the discrete-time domain. The main results presented in this paper are the exact problem formulation and solution of an optimal discrete-time dynamic output-feedback design in the w-domain involving a quadratic performance index to random disturbances. In addition, necessary conditions for optimality are obtained for the numerical solution of the optimal output-feedback compensator design. A numerical example is presented illustrating its application to the design of a low-order dynamic compensator in a stability augmentation system of a commercial transport.
Numerical approach of collision avoidance and optimal control on robotic manipulators
Collision-free optimal motion and trajectory planning for robotic manipulators are solved by a method of sequential gradient restoration algorithm. Numerical examples of a two degree-of-freedom (DOF) robotic manipulator are demonstrated to show the excellence of the optimization technique and obstacle avoidance scheme. The obstacle is put on the midway, or even further inward on purpose, of the previous no-obstacle optimal trajectory. For the minimum-time purpose, the trajectory grazes by the obstacle and the minimum-time motion successfully avoids the obstacle. The minimum-time is longer for the obstacle avoidance cases than the one without obstacle. The obstacle avoidance scheme can deal with multiple obstacles in any ellipsoid forms by using artificial potential fields as penalty functions via distance functions. The method is promising in solving collision-free optimal control problems for robotics and can be applied to any DOF robotic manipulators with any performance indices and mobile robots as well. Since this method generates optimum solution based on Pontryagin Extremum Principle, rather than based on assumptions, the results provide a benchmark against which any optimization techniques can be measured.
Ciel
Compiler optimizations can alter the numerical results of scientific computing applications. When numerical results differ significantly between compilers, optimization levels, and floating-point hardware, these numerical inconsistencies can impact programming productivity. Ciel is a framework that helps programmers identify locations in the source code that are affected by compiler optimizations in CPU and GPU code. Ciel uses a floating-point precision enhancement strategy, guided by a recursive bisection search algorithm with increasing search granularity, to identify the program expressions that induce numerical inconsistencies due to compiler optimizations.
Robust Sensor Faults Reconstruction for a Class of Uncertain Linear Systems Using a Sliding Mode Observer: An LMI Approach
This paper presents a design method of a Sliding Mode Observer (SMO) for robust sensor faults reconstruction of systems with matched uncertainty. This class of uncertainty requires a known upper bound. The basic idea is to use the H{sub {infinity}} concept to design the observer, which minimizes the effect of the uncertainty on the reconstruction of the sensor faults. Specifically, we applied the equivalent output error injection concept from previous work in Fault Detection and Isolation (FDI) scheme. Then, these two problems of design and reconstruction can be expressed and numerically formulate via Linear Matrix Inequalities (LMIs) optimization. Finally, a numerical example is given to illustrate the validity and the applicability of the proposed approach.
Optimization and supervised machine learning methods for fitting numerical physics models without derivatives
Here, we address the calibration of a computationally expensive nuclear physics model for which derivative information with respect to the fit parameters is not readily available. Of particular interest is the performance of optimization-based training algorithms when dozens, rather than millions or more, of training data are available and when the expense of the model places limitations on the number of concurrent model evaluations that can be performed. As a case study, we consider the Fayans energy density functional model, which has characteristics similar to many model fitting and calibration problems in nuclear physics. We analyze hyperparameter tuning considerations and variability associated with stochastic optimization algorithms and illustrate considerations for tuning in different computational settings.
Multi-solution nature of topology optimization and its application in design for additive manufacturing
The purpose of this paper is to introduce the multi-solution nature of topology optimization (TO) as a design tool for additive manufacturing (AM). The sensitivity of topologically optimized parts and manufacturing constraints to the initial starting point of the optimization process leading to structures with equivalent performance is explored. A modified bi-directional evolutionary structural optimization (BESO) code was used as the numerical approach to optimize a cantilever beam problem and reduce the mass by 50 per cent. Several optimized structures with relatively equivalent mechanical performance were generated by changing the initial starting point of the TO algorithm. These optimized structures were manufactured using fused deposition modeling (FDM). The equivalence of strain distribution in FDM parts was tested with the digital image correlation (DIC) technique and compared with that from the modified BESO code.The results confirm that TO could lead to a wide variety of non-unique solutions based on loading and manufacturability constraints. The modified BESO code was able to reduce the support structure needed to build the simple two-dimensional cantilever beam by 15 per cent while keeping the mechanical performance at the same level. Finally, the originality of this paper lies in introduction and application of the multi-solution nature of TO for AM as a design tool for optimizing structures with minimized features in the overhang condition and the need for support structures.
AI/OR computational model for integrating qualitative and quantitative design methods
A theoretical framework for integrating qualitative and numerical computational methods for optimally-directed design is described. The theory is presented as a computational model and features of implementations are summarized where appropriate. To demonstrate the versatility of the methodology we focus on four seemingly disparate aspects of the design process and their interaction: (1) conceptual design, (2) qualitative optimal design, (3) design innovation, and (4) numerical global optimization.
An accelerated gradient method for parameter optimization with non-linear constraints.
Accelerated gradient method for numerical solution of parameter optimization problems with nonlinear constraints
Optimization problems governed by systems of PDEs with uncertainties
This paper reviews current theoretical and numerical approaches to optimization problems governed by partial differential equations (PDEs) that depend on random variables or random fields. Such problems arise in many engineering, science, economics and societal decision-making tasks. This paper focuses on problems in which the governing PDEs are parametrized by the random variables/fields, and the decisions are made at the beginning and are not revised once uncertainty is revealed. Examples of such problems are presented to motivate the topic of this paper, and to illustrate the impact of different ways to model uncertainty in the formulations of the optimization problem and their impact on the solution. A linear–quadratic elliptic optimal control problem is used to provide a detailed discussion of the set-up for the risk-neutral optimization problem formulation, study the existence and characterization of its solution, and survey numerical methods for computing it. Different ways to model uncertainty in the PDE-constrained optimization problem are surveyed in an abstract setting, including risk measures, distributionally robust optimization formulations, probabilistic functions and chance constraints, and stochastic orders. Furthermore, approximation-based optimization approaches and stochastic methods for the solution of the large-scale PDE-constrained optimization problems under uncertainty are described. Some possible future research directions are outlined.