Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Adjoint Computation”

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.

At least 307 records · Page 17

Design of efficient stiffened shells of revolution

A method to produce efficient piecewise uniform stiffened shells of revolution is presented. The approach uses a first order differential equation formulation for the shell prebuckling and buckling analyses and the necessary conditions for an optimum design are derived by a variational approach. A variety of local yielding and buckling constraints and the general buckling constraint are included in the design process. The local constraints are treated by means of an interior penalty function and the general buckling load is treated by means of an exterior penalty function. This allows the general buckling constraint to be included in the design process only when it is violated. The self-adjoint nature of the prebuckling and buckling formulations is used to reduce the computational effort. Results for four conical shells and one spherical shell are given.

Majumder, D. K.↗

SCALE Sensitivity Calculations Using Contributon Theory

The SCALE TSUNAMI-3D sensitivity and uncertainty analysis sequence computes the sensitivity of k-eff to each constituent multigroup cross section using adjoint techniques with the KENO Monte Carlo codes. A new technique to simultaneously obtain the product of the forward and adjoint angular flux moments within a single Monte Carlo calculation has been developed and implemented in the SCALE TSUNAMI-3D analysis sequence. A new concept in Monte Carlo theory has been developed for this work, an eigenvalue contributon estimator, which is an extension of previously developed fixed source contributon estimators. A contributon is a particle for which the forward solution is accumulated, and its importance to the response, which is equivalent to the adjoint solution, is simultaneously accumulated. Thus, the contributon is a particle coupled with its contribution to the response, in this case keff. As implemented in SCALE, the contributon provides the importance of a particle exiting at any energy or direction for each location, energy and direction at which the forward flux solution is sampled. Although currently implemented for eigenvalue calculations in multigroup mode in KENO, this technique is directly applicable to continuous-energy calculations for many other responses such as fixed-source sensitivity analysis and quantification of reactor kinetics parameters. This paper provides the physical bases of eigenvalue contributon theory, provides details of implementation into TSUNAMI-3D, and provides results of sample calculations.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Aerodynamic Design Optimization for Natural Laminar Flow Airfoils

Natural laminar flow technology is a passive laminar flow control (LFC) strategy that seeks to delay the onset of boundary-layer transition (BLT) through shape optimization to reduce the drag of the aerodynamic vehicle. Adjoint-based design optimization for LFC is proposed in an integrated multidisciplinary framework, which includes the computational fluid dynamics (CFD), geometry and grid deformation, and linear stability analysis (LSA) for transition prediction. In particular, the BLT location is predicted using the dual N-factor method that is based on a linear stability theory (LST) eigenvalue problem. The dual N-factor criterion accounts for the amplification of planar Tollmien-Schlichting (TS) and stationary crossflow (CF) boundary-layer instabilities to predict the transition location in three-dimensional boundary-layer flows. The adjoint-based shape optimization procedure is based on an iteratively coupled CFD and LSA methodology to converge the transition location and flow solutions, as well as to calculate the sensitivities of the aerodynamic metrics of interest with respect to the flow and shape design parameters. The RAE 2822 airfoil at 0 and 30 degrees yaw angles, an angle of attack of 0.72 degrees, and subsonic conditions (M∞ = 0.19, Rec = 5.6 × 106 ) are used as baseline configurations for design optimization. The angle of attack and the vertical displacement of free-form-deformation control points are used as design variables to reduce the drag coefficient while reaching a specified lift coefficient. The optimized unswept airfoil designs achieve a 30% drag reduction accompanied by a downstream shift of the transition locations over both suction and pressure sides of the airfoil. The initial design iterations for the swept case also show a favorable trend in the drag reduction with transition delay over both sides.

Transition↗

Parameter estimation in a thermoelastic composite problem via adjoint formulation and model reduction

Advances in nondestructive material characterization are providing a wealth of information that could be exploited to gain insight into general aspects of material performance and, in particular, discover relationships between microstructure and thermo-mechanical properties in polycrystalline and other complex composite materials. Here, in order to facilitate the integration of such measurements into existing models, as well as inform new physics-based predictions, we developed a C++/MPI computational framework for sensitivity analysis and parameter estimation. The framework utilizes a micro-mechanical modeling based on fast Fourier transforms, direct and adjoint formulations, and Markov chain Monte Carlo sampling techniques. We illustrate the characteristics of this framework and demonstrate its utility by computing the residual stresses arising from thermal expansion of an elastic composite and using data from simulated experiments. Lastly, we show that the availability of nondestructive 3-D measurements is crucial to reduce the uncertainty in predictions, emphasizing the importance of an integrated experimental/modeling/data analysis approach for improved material characterization and design.

36 MATERIALS SCIENCE↗

Necessary Conditions for Optimal Control of Stochastic Evolution Equations in Hilbert Spaces

We consider a nonlinear stochastic optimal control problem associated with a stochastic evolution equation. This equation is driven by a continuous martingale in a separable Hilbert space and an unbounded time-dependent linear operator.We derive a stochastic maximum principle for this optimal control problem. Our results are achieved by using the adjoint backward stochastic partial differential equation.

97 MATHEMATICS AND COMPUTING↗

Variable-fidelity multipoint aerodynamic shape optimization with output-based adapted meshes

This work presents a method to control the discretization error in multipoint aerodynamic shape optimization using output-based adapted meshes. The meshes are adapted via adjoint-based error estimates, taking into account both the objective and constraint output errors. A multi-fidelity optimization framework is then developed by taking advantage of the variable fidelity offered by adaptive meshes. The objective functional and its sensitivity at each design point (operating condition) are first evaluated on the same initial coarse mesh, which is then subsequently adapted for each design point individually as the shape optimization proceeds. The effort to set up the optimization is minimal since the initial mesh can be fairly coarse and easy to generate. As the shape approaches the optimal design, the mesh at each design point becomes finer, in regions necessary for that particular operating condition. The multi-fidelity framework is tightly coupled with the objective error estimation to ensure the optimization accuracy at each fidelity. Computational savings arise from a reduction of the mesh size when the design is far from optimal and avoiding an exhaustive search on low-fidelity meshes. The proposed method is demonstrated on multipoint drag minimization problems of a transonic airfoil with lift and area constraints. Improved accuracy and efficiency are shown compared to traditional fixed-fidelity optimization with a fixed computational mesh.

42 ENGINEERING↗

OpenAD/F : a modular, open-source tool for automatic differentiation of Fortran codes.

The OpenAD/F tool allows the evaluation of derivatives of functions defined by a Fortran program. The derivative evaluation is performed by a Fortran code resulting from the analysis and transformation of the original program that defines the function of interest. OpenAD/F has been designed with a particular emphasis on modularity, flexibility, and the use of open source components. While the code transformation follows the basic principles of automatic differentiation, the tool implements new algorithmic approaches at various levels, for example, for basic block preaccumulation and call graph reversal. Unlike most other automatic differentiation tools, OpenAD/F uses components provided by the OpenAD framework, which supports a comparatively easy extension of the code transformations in a language-independent fashion. It uses code analysis results implemented in the OpenAnalysis component. The interface to the language-independent transformation engine is an XML-based format, specified through an XML schema. The implemented transformation algorithms allow efficient derivative computations utilizing locally optimized cross-country sequences of vertex, edge, and face elimination steps. Specifically, for the generation of adjoint codes, OpenAD/F supports various code reversal schemes with hierarchical checkpointing at the subroutine level. As an example from geophysical fluid dynamics a nonlinear time-dependent scalable, yet simple, barotropic ocean model is considered. OpenAD/F's reverse mode is applied to compute sensitivities of some of the model's transport properties with respect to gridded fields such as bottom topography as independent (control) variables.

54 ENVIRONMENTAL SCIENCES↗

THE ADJOINT METHOD APPLIED TO TIME-DISTANCE HELIOSEISMOLOGY

For a given misfit function, a specified optimality measure of a model, its gradient describes the manner in which one may alter properties of the system to march toward a stationary point. The adjoint method, arising from partial-differential-equation-constrained optimization, describes a means of extracting derivatives of a misfit function with respect to model parameters through finite computation. It relies on the accurate calculation of wavefields that are driven by two types of sources, namely, the average wave-excitation spectrum, resulting in the forward wavefield, and differences between predictions and observations, resulting in an adjoint wavefield. All sensitivity kernels relevant to a given measurement emerge directly from the evaluation of an interaction integral involving these wavefields. The technique facilitates computation of sensitivity kernels (Frechet derivatives) relative to three-dimensional heterogeneous background models, thereby paving the way for nonlinear iterative inversions. An algorithm to perform such inversions using as many observations as desired is discussed.

79 ASTRONOMY AND ASTROPHYSICS↗

Optimum design of ninety degree bends

An algorithm for the optimum design of an internal flow component to obtain the maximum pressure rise is presented. Maximum pressure rise in a duct with simultaneous turning and diffusion is shown to be related to the control of flow separation on the passage walls. Such a flow is usually associated with downstream conditions that are desirable in turbomachinery and propulsion applications to ensure low loss and stable performance. The algorithm requires the solution of an 'adjoint' problem in addition to the 'direct' equations governing the flow in a body, which in the present analysis are assumed to be the laminar Navier-Stokes equations. The theoretical framework and computational algorithms presented in this study are for the steady Navier-Stokes equations. A procedure is developed for the numerical solution of the adjoint equations. This procedure is coupled with a direct solver in a design iteration loop, that provides a new shape with a higher pressure rise. This procedure is first validated for the design of optimum plane diffusers in two-dimensional flow. The direct Navier-Stokes and the 'adjoint' equations are solved using a finite volume formulation for spatial discretization in an artificial compressibility framework. A simplified version of the above approach is then utilized to design ninety degree diffusing bends. Calculations were carried out for a mean radius ratio at inlet of 2.5 and Reynolds numbers varying from 100 to 500. While at this stage laminar flows is assumed, it is shown that a similar approach can be conceived for turbulent flows.

Modi, Vijay↗

Global adjoint tomography: first-generation model

We present the first-generation global tomographic model constructed based on adjoint tomography, an iterative full-waveform inversion technique. Synthetic seismograms were calculated using GPU-accelerated spectral-element simulations of global seismic wave propagation, accommodating effects due to 3-D anelastic crust & mantle structure, topography & bathymetry, the ocean load, ellipticity, rotation, and self-gravitation. Fréchet derivatives were calculated in 3-D anelastic models based on an adjoint-state method. The simulations were performed on the Cray XK7 named ‘Titan’, a computer with 18 688 GPU accelerators housed at Oak Ridge National Laboratory. The transversely isotropic global model is the result of 15 tomographic iterations, which systematically reduced differences between observed and simulated three-component seismograms. Our starting model combined 3-D mantle model S362ANI with 3-D crustal model Crust2.0. We simultaneously inverted for structure in the crust and mantle, thereby eliminating the need for widely used ‘crustal corrections’. We used data from 253 earthquakes in the magnitude range 5.8 ≤ M w ≤ 7.0. We started inversions by combining ~30 s body-wave data with ~60 s surface-wave data. The shortest period of the surface waves was gradually decreased, and in the last three iterations we combined ~17 s body waves with ~45 s surface waves. We started using 180 min long seismograms after the 12th iteration and assimilated minor- and major-arc body and surface waves. The 15th iteration model features enhancements of well-known slabs, an enhanced image of the Samoa/Tahiti plume, as well as various other plumes and hotspots, such as Caroline, Galapagos, Yellowstone and Erebus. Furthermore, we see clear improvements in slab resolution along the Hellenic and Japan Arcs, as well as subduction along the East of Scotia Plate, which does not exist in the starting model. Point-spread function tests demonstrate that we are approaching the resolution of continental-scale studies in some areas, for example, underneath Yellowstone. Here, this is a consequence of our multiscale smoothing strategy in which we define our smoothing operator as a function of the approximate Hessian kernel, thereby smoothing gradients less wherever we have good ray coverage, such as underneath North America.

58 GEOSCIENCES↗

Influence of the Perception Constraint, Installation Effects, and Broadband SelfNoise on Low-Fidelity, Multi-point, UAM Vehicle Perception-Influenced-Design(PID) Optimization

UAM vehicles represent a unique challenge in acoustics because there are a limited number of full-scale datasets that can be used to calibrate and validate prediction models. In addition, acoustics is considered more often in the early design loop as a higher priority concern in UAM vehicle design, whereas with traditional vehicles, acoustics is typically addressed with post-design constraints. This paper will present an approach in design optimization of a proprotor in two flight conditions, hover, and forward flight, with several acoustic objective functions that simulate community response. An adjoint-capable vortex-lattice method (VLM), implemented in VSPAERO, will be used to compute the blade forces and inflow properties required to compute tonal and broadband noise using the second-generation Aircraft NOise Prediction Program (ANOPP2). In addition, a wing is placed near the proprotor to simulate installation effects on a vehicle’s wing. The influence of rotation rate, chord distribution, twist angle, and sweep on installation, broadband noise, and choice of perception constraint from the design optimization will be shown.

acoustics↗

Influence of the Perception Constraint, Observer Position, and Broadband Self-Noise on Low-Fidelity UAM Vehicle Perception-Influenced-Design (PID) Optimization

This paper will present an approach in design optimization of a proprotor in a forward flight condition with several acoustic objective functions that simulate community response. An adjoint-capable blade element momentum theory (BEMT), implemented in CCBlade.jl, will be used to compute the blade forces and inflow properties required to compute tonal and broadband noise using the second-generation Aircraft NOise Prediction Program (ANOPP2). The influence of broadband noise, observer location, and choice of perception constraint on optimized values of rotation rate, chord distribution, and twist angle will be shown. It is determined that broadband self-noise is an influential source noise mechanism in the design optimization when frequency weighting is used to predict the noise at an out-of-plane observer position.

acoustics↗

Influence of the Perception Constraint, Observer Position, and Broadband Self-Noise on Low-Fidelity UAM Vehicle Perception-Influenced-Design (PID) Optimization

This paper will present an approach in design optimization of a proprotor in a forward flight condition with several acoustic objective functions that simulate community response. An adjoint-capable blade element momentum theory (BEMT), implemented in CCBlade.jl, will be used to compute the blade forces and inflow properties required to compute tonal and broadband noise using the second-generation Aircraft NOise Prediction Program (ANOPP2). The influence of broadband noise, observer location, and choice of perception constraint on optimized values of rotation rate, chord distribution, and twist angle will be shown. It is determined that broadband self-noise is an influential source noise mechanism in the design optimization when frequency weighting is used to predict the noise at an out-of-plane observer position.

acoustics↗

Sonic Boom Ground Noise Minimization Via the Adjoint Method

A multidisciplinary design optimization methodology to directly minimize the ground-level noise generated by the sonic boom of a high altitude supersonic body is presented. A Cartesian Euler flow solver is coupled with an atmospheric propagation tool to create a ground-noise analysis capability for supersonic bodies. Adjoint formulations for both the flow solver and propagation tool are also coupled to compute sensitivities of shape variations in the body in a highly efficient manner. A gradient-based optimizer is then introduced to forge a valuable design capability. Output-based mesh adaptation that is driven directly by ground-level noise is employed to increase accuracy and provide error estimation. The design method is demonstrated first on a simple axisymmetric body with few design variables to evaluate the efficacy of the optimization scheme. Guided by the results of this initial case, the problem is then repeated with somewhat different design variables to further demonstrate the capabilities of the design method. Finally, the method is applied to a real-world problem by optimizing control surface deflection settings of a low-boom aircraft to minimize ground noise while maintaining trimmed, level flight.

ARMD↗

Efficient, automated Monte Carlo methods for radiation transport

Monte Carlo simulations provide an indispensible model for solving radiative transport problems, but their slow convergence inhibits their use as an everyday computational tool. In this paper, we present two new ideas for accelerating the convergence of Monte Carlo algorithms based upon an efficient algorithm that couples simulations of forward and adjoint transport equations. Forward random walks are first processed in stages, each using a fixed sample size, and information from stage k is used to alter the sampling and weighting procedure in stage k+1. This produces rapid geometric convergence and accounts for dramatic gains in the efficiency of the forward computation. In case still greater accuracy is required in the forward solution, information from an adjoint simulation can be added to extend the geometric learning of the forward solution. The resulting new approach should find widespread use when fast, accurate simulations of the transport equation are needed.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Development and Applications of the FV3 GEOS-5 Adjoint Modeling System

GMAO has developed a highly sophisticated adjoint modeling system based on the most recent version of the finite volume cubed sphere (FV3) dynamical core. This provides a mechanism for investigating sensitivity to initial conditions and examining observation impacts. It also allows for the computation of singular vectors and for the implementation of hybrid 4DVAR. In this work we will present the scientific assessment of the new adjoint system and show results from a number of research application of the adjoint system.

Holdaway, Daniel↗

The strip method for shape derivatives

Abstract A major challenge in shape optimization is the coupling of finite element method (FEM) codes in a way that facilitates efficient computation of shape derivatives. This is particularly difficult with multiphysics problems involving legacy codes, where the costs of implementing and maintaining shape derivative capabilities are prohibitive. The volume and boundary methods are two approaches to computing shape derivatives. Each has a major drawback: the boundary method is less accurate, while the volume method is more invasive to the FEM code. We introduce the strip method , which computes shape derivatives on a strip adjacent to the boundary. The strip method makes code coupling simple. Like the boundary method, it queries the state and adjoint solutions at quadrature nodes, but requires no knowledge of the FEM code implementations. At the same time, it exhibits the higher accuracy of the volume method. As an added benefit, its computational complexity is comparable to that of the boundary method, that is, it is faster than the volume method. We illustrate the benefits of the strip method with numerical examples.

Hardesty, Sean↗

Accelerating full waveform inversion via source stacking and cross-correlations

SUMMARY Accurate synthetic seismic wavefields can now be computed in 3-D earth models using the spectral element method (SEM), which helps improve resolution in full waveform global tomography. However, computational costs are still a challenge. These costs can be reduced by implementing a source stacking method, in which multiple earthquake sources are simultaneously triggered in only one teleseismic SEM simulation. One drawback of this approach is the perceived loss of resolution at depth, in particular because high-amplitude fundamental mode surface waves dominate the summed waveforms, without the possibility of windowing and weighting as in conventional waveform tomography. This can be addressed by redefining the cost-function and computing the cross-correlation wavefield between pairs of stations before each inversion iteration. While the Green’s function between the two stations is not reconstructed as well as in the case of ambient noise tomography, where sources are distributed more uniformly around the globe, this is not a drawback, since the same processing is applied to the 3-D synthetics and to the data, and the source parameters are known to a good approximation. By doing so, we can separate time windows with large energy arrivals corresponding to fundamental mode surface waves. This opens the possibility of designing a weighting scheme to bring out the contribution of overtones and body waves. It also makes it possible to balance the contributions of frequently sampled paths versus rarely sampled ones, as in more conventional tomography. Here we present the results of proof of concept testing of such an approach for a synthetic 3-component long period waveform data set (periods longer than 60 s), computed for 273 globally distributed events in a simple toy 3-D radially anisotropic upper mantle model which contains shear wave anomalies at different scales. We compare the results of inversion of 10 000 s long stacked time-series, starting from a 1-D model, using source stacked waveforms and station-pair cross-correlations of these stacked waveforms in the definition of the cost function. We compute the gradient and the Hessian using normal mode perturbation theory, which avoids the problem of cross-talk encountered when forming the gradient using an adjoint approach. We perform inversions with and without realistic noise added and show that the model can be recovered equally well using one or the other cost function. The proposed approach is computationally very efficient. While application to more realistic synthetic data sets is beyond the scope of this paper, as well as to real data, since that requires additional steps to account for such issues as missing data, we illustrate how this methodology can help inform first order questions such as model resolution in the presence of noise, and trade-offs between different physical parameters (anisotropy, attenuation, crustal structure, etc.) that would be computationally very costly to address adequately, when using conventional full waveform tomography based on single-event wavefield computations.

Romanowicz, Barbara↗