Search NASA⌕ Search

SEARCH · Search NASA

Results for “Adaptive finite elements”

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 757 records · Page 42

Higher harmonic control analysis for vibration reduction of helicopter rotor systems

An advanced higher harmonic control (HHC) analysis has been developed and applied to investigate its effect on vibration reduction levels, blade and control system fatigue loads, rotor performance, and power requirements of servo-actuators. The analysis is based on a finite element method in space and time. A nonlinear time domain unsteady aerodynamic model, based on the indicial response formulation, is used to calculate the airloads. The rotor induced inflow is computed using a free wake model. The vehicle trim controls and blade steady responses are solved as one coupled solution using a modified Newton method. A linear frequency-domain quasi-steady transfer matrix is used to relate the harmonics of the vibratory hub loads to the harmonics of the HHC inputs. Optimal HHC is calculated from the minimization of the vibratory hub loads expressed in term of a quadratic performance index. Predicted vibratory hub shears are correlated with wind tunnel data. The fixed-gain HHC controller suppresses completely the vibratory hub shears for most of steady or quasi-steady flight conditions. HHC actuator amplitudes and power increase significantly at high forward speeds (above 100 knots). Due to the applied HHC, the blade torsional stresses and control loads are increased substantially. For flight conditions where the blades are stalled considerably, the HHC input-output model is quite nonlinear. For such cases, the adaptive-gain controller is effective in suppressing vibratory hub loads, even though HHC may actually increase stall areas on the rotor disk. The fixed-gain controller performs poorly for such flight conditions. Comparison study of different rotor systems indicates that a soft-inplane hingeless rotor requires less actuator power at high speeds (above 130 knots) than an articulated rotor, and a stiff-inplane hingeless rotor generally requires more actuator power than an articulated or a soft-inplane hingeless rotor. Parametric studies for a hingeless rotor operating in a transition flight regime and for an articulated rotor operating at the level-flight boundary (high speed and high thrust conditions) indicate that blade parameters including flap, lag, torsion stiffness distributions, linear pretwist, chordwise offset of center-of-mass from elastic axis and chordwise offset of elastic axis from aerodynamic center can be selected to minimize the actuator power requirements for HHC.

Nguyen, Khanh Q.↗

Development of Multiobjective Optimization Techniques for Sonic Boom Minimization

A discrete, semi-analytical sensitivity analysis procedure has been developed for calculating aerodynamic design sensitivities. The sensitivities of the flow variables and the grid coordinates are numerically calculated using direct differentiation of the respective discretized governing equations. The sensitivity analysis techniques are adapted within a parabolized Navier Stokes equations solver. Aerodynamic design sensitivities for high speed wing-body configurations are calculated using the semi-analytical sensitivity analysis procedures. Representative results obtained compare well with those obtained using the finite difference approach and establish the computational efficiency and accuracy of the semi-analytical procedures. Multidisciplinary design optimization procedures have been developed for aerospace applications namely, gas turbine blades and high speed wing-body configurations. In complex applications, the coupled optimization problems are decomposed into sublevels using multilevel decomposition techniques. In cases with multiple objective functions, formal multiobjective formulation such as the Kreisselmeier-Steinhauser function approach and the modified global criteria approach have been used. Nonlinear programming techniques for continuous design variables and a hybrid optimization technique, based on a simulated annealing algorithm, for discrete design variables have been used for solving the optimization problems. The optimization procedure for gas turbine blades improves the aerodynamic and heat transfer characteristics of the blades. The two-dimensional, blade-to-blade aerodynamic analysis is performed using a panel code. The blade heat transfer analysis is performed using an in-house developed finite element procedure. The optimization procedure yields blade shapes with significantly improved velocity and temperature distributions. The multidisciplinary design optimization procedures for high speed wing-body configurations simultaneously improve the aerodynamic, the sonic boom and the structural characteristics of the aircraft. The flow solution is obtained using a comprehensive parabolized Navier Stokes solver. Sonic boom analysis is performed using an extrapolation procedure. The aircraft wing load carrying member is modeled as either an isotropic or a composite box beam. The isotropic box beam is analyzed using thin wall theory. The composite box beam is analyzed using a finite element procedure. The developed optimization procedures yield significant improvements in all the performance criteria and provide interesting design trade-offs. The semi-analytical sensitivity analysis techniques offer significant computational savings and allow the use of comprehensive analysis procedures within design optimization studies.

Chattopadhyay, Aditi↗

Detection of Cracks at Welds in Steel Tubing Using Flux Focusing Electromagnetic Probe

The inspection of weldments in critical pressure vessel joints is a major concern in the nuclear power industry. Corrosive environments can speed the fatigue process and access to the critical area is often limited. Eddy current techniques have begun to be used to help overcome these obstacles [1]. As direct contact and couplants are not required, remote areas can be inspected by simply snaking an eddy current coil into the intake tube of the vessel. The drawback of the eddy current method has been the high sensitivity to small changes in the conductivity and permeability of the test piece which are known to vary at weldments [1]. The flaw detection mechanism of the flux focusing electromagnetic probe can help alleviate these difficulties and provide a unique capability for detecting longitudinal fatigue cracks in critical tube structures. The Flux Focusing Electromagnetic Flaw Detector, originally invented for the detection of fatigue and corrosion damage in aluminum plates [2-3], has been adapted for use in testing steel tubing for longitudinal fatigue cracks. The modified design allows for the probe to be placed axisymmetrically into the tubing, inducing eddy currents in the tube wall. The pickup coil of the probe is fixed slightly below the primary windings and is rotated 90 so that its axis is normal to the tube wall. The magnetic flux of the primary coil is focused through the use of ferromagnetic material so that in the absence of fatigue damage there will be no flux linkage with the pickup coil. The presence of a longitudinal fatigue crack will cause the eddy currents induced in the tube wall to flow around the flaw and directly under the pickup coil. The magnetic field associated with these currents will then link the pickup coil and an unambiguous increase in the output voltage of the probe will be measured. The use of the flux focusing electromagnetic probe is especially suited for the detection of flaws originating at or near tube welds. The probe is shown to discriminate against signals due solely to the weld joint so that flaw signals are not hidden in the background in these locations. Experimental and finite element modeling results are presented for the flaw detection capabilities of the probe in stainless steel tubes.

Wincheski, Buzz↗

Test-Anchored Vibration Response Predictions for an Acoustically Energized Curved Orthogrid Panel with Mounted Components

A rich body of vibroacoustic test data was recently generated at Marshall Space Flight Center for a curved orthogrid panel typical of launch vehicle skin structures. Several test article configurations were produced by adding component equipment of differing weights to the flight-like vehicle panel. The test data were used to anchor computational predictions of a variety of spatially distributed responses including acceleration, strain and component interface force. Transfer functions relating the responses to the input pressure field were generated from finite element based modal solutions and test-derived damping estimates. A diffuse acoustic field model was employed to describe the assumed correlation of phased input sound pressures across the energized panel. This application demonstrates the ability to quickly and accurately predict a variety of responses to acoustically energized skin panels with mounted components. Favorable comparisons between the measured and predicted responses were established. The validated models were used to examine vibration response sensitivities to relevant modeling parameters such as pressure patch density, mesh density, weight of the mounted component and model form. Convergence metrics include spectral densities and cumulative root-mean squared (RMS) functions for acceleration, velocity, displacement, strain and interface force. Minimum frequencies for response convergence were established as well as recommendations for modeling techniques, particularly in the early stages of a component design when accurate structural vibration requirements are needed relatively quickly. The results were compared with long-established guidelines for modeling accuracy of component-loaded panels. A theoretical basis for the Response/Pressure Transfer Function (RPTF) approach provides insight into trends observed in the response predictions and confirmed in the test data. The software modules developed for the RPTF method can be easily adapted for quick replacement of the diffuse acoustic field with other pressure field models; for example a turbulent boundary layer (TBL) model suitable for vehicle ascent. Wind tunnel tests have been proposed to anchor the predictions and provide new insight into modeling approaches for this type of environment. Finally, component vibration environments for design were developed from the measured and predicted responses and compared with those derived from traditional techniques such as Barrett scaling methods for unloaded and component-loaded panels.

Frady, Gregory P.↗

Identifying high-impact and high-uncertainty parameters in MiniFuel model predictions

The MiniFuel irradiation platform at Oak Ridge National Laboratory's High Flux Isotope Reactor (HFIR) is a flexible, high-throughput separate effects test capability. Finite element thermal models are relied upon to design MiniFuel experiments and to achieve experimental objectives. Recent reports show good agreement in the model prediction of target fuel temperatures, but as the capability of the experiments is extended to higher temperatures, the uncertainty in the model predictions must be quantified. To that end, high-impact, high-uncertainty parameters that contribute the most uncertainty to the model are identified. The uncertainty quantification was accomplished through a series of screening and sensitivity analyses. The first analysis utilizes the method of Morris to perform a computationally efficient preliminary screening that considers uncertainty in a large number of the model inputs. The most important parameters identified in the Morris screening study were then considered in a Sobol sensitivity analysis that more robustly ranks and quantifies the uncertainty associated with each parameter. From these analyses, it was determined that thermal contact conductance between components is the parameter that contributes the highest uncertainty. The estimated uncertainty of the MiniFuel model fuel temperature predictions is ±80 °C in the removable beryllium and ±40 °C in the vertical experiment facilities. In conclusion, the framework established by the series of sensitivity analyses presented herein could easily be adapted to fit the needs of accelerated fuel qualification processes.

Fuel, Irradiation↗

A global/local analysis method for treating details in structural design

A method for analyzing global/local behavior of plate and shell structures is described. In this approach, a detailed finite element model of the local region is incorporated within a coarser global finite element model. The local model need not be nodally compatible (i.e., need not have a one-to-one nodal correspondence) with the global model at their common boundary; therefore, the two models may be constructed independently. The nodal incompatibility of the models is accounted for by introducing appropriate constraint conditions into the potential energy in a hybrid variational formulation. The primary advantage of this method is that the need for transition modeling between global and local models is eliminated. Eliminating transition modeling has two benefits. First, modeling efforts are reduced since tedious and complex transitioning need not be performed. Second, errors due to the mesh distortion, often unavoidable in mesh transitioning, are minimized by avoiding distorted elements beyond what is needed to represent the geometry of the component. The method is applied reduced to a plate loaded in tension and transverse bending. The plate has a central hole, and various hole sixes and shapes are studied. The method is also applied to a composite laminated fuselage panel with a crack emanating from a window in the panel. While this method is applied herein to global/local problems, it is also applicable to the coupled analysis of independently modeled components as well as adaptive refinement.

Aminpour, Mohammad A.↗

Unstructured Mesh Methods for the Simulation of Hypersonic Flows

This report describes the research work undertaken at the Massachusetts Institute of Technology. The aim of this research is to identify effective algorithms and methodologies for the efficient and routine solution of hypersonic viscous flows about re-entry vehicles. For over ten years we have received support from NASA to develop unstructured mesh methods for Computational Fluid Dynamics. As a result of this effort a methodology based on the use, of unstructured adapted meshes of tetrahedra and finite volume flow solvers has been developed. A number of gridding algorithms flow solvers, and adaptive strategies have been proposed. The most successful algorithms developed from the basis of the unstructured mesh system FELISA. The FELISA system has been extensively for the analysis of transonic and hypersonic flows about complete vehicle configurations. The system is highly automatic and allows for the routine aerodynamic analysis of complex configurations starting from CAD data. The code has been parallelized and utilizes efficient solution algorithms. For hypersonic flows, a version of the, code which incorporates real gas effects, has been produced. One of the latest developments before the start of this grant was to extend the system to include viscous effects. This required the development of viscous generators, capable of generating the anisotropic grids required to represent boundary layers, and viscous flow solvers. In figures I and 2, we show some sample hypersonic viscous computations using the developed viscous generators and solvers. Although these initial results were encouraging, it became apparent that in order to develop a fully functional capability for viscous flows, several advances in gridding, solution accuracy, robustness and efficiency were required. As part of this research we have developed: 1) automatic meshing techniques and the corresponding computer codes have been delivered to NASA and implemented into the GridEx system, 2) a finite element algorithm for the solution of the viscous compressible flow equations which can solve flows all the way down to the incompressible limit and that can use higher order (quadratic) approximations leading to highly accurate answers, and 3) and iterative algebraic multigrid solution techniques.

Peraire, Jaime↗

Strain Gage Load Calibration of the Wing Interface Fittings for the Adaptive Compliant Trailing Edge Flap Flight Test

This is the presentation to follow conference paper of the same name. The adaptive compliant trailing edge (ACTE) flap experiment safety of flight requires that the flap to wing interface loads be sensed and monitored in real time to ensure that the wing structural load limits are not exceeded. This paper discusses the strain gage load calibration testing and load equation derivation methodology for the ACTE interface fittings. Both the left and right wing flap interfaces will be monitored and each contains four uniquely designed and instrumented flap interface fittings. The interface hardware design and instrumentation layout are discussed. Twenty one applied test load cases were developed using the predicted in-flight loads for the ACTE experiment.

strain gage calibration↗

Investigation of Liner Characteristics in the NASA Langley Curved Duct Test Rig

The Curved Duct Test Rig (CDTR), which is designed to investigate propagation of sound in a duct with flow, has been developed at NASA Langley Research Center. The duct incorporates an adaptive control system to generate a tone in the duct at a specific frequency with a target Sound Pressure Level and a target mode shape. The size of the duct, the ability to isolate higher order modes, and the ability to modify the duct configuration make this rig unique among experimental duct acoustics facilities. An experiment is described in which the facility performance is evaluated by measuring the sound attenuation by a sample duct liner. The liner sample comprises one wall of the liner test section. Sound in tones from 500 to 2400 Hz, with modes that are parallel to the liner surface of order 0 to 5, and that are normal to the liner surface of order 0 to 2, can be generated incident on the liner test section. Tests are performed in which sound is generated without axial flow in the duct and with flow at a Mach number of 0.275. The attenuation of the liner is determined by comparing the sound power in a hard wall section downstream of the liner test section to the sound power in a hard wall section upstream of the liner test section. These experimentally determined attenuations are compared to numerically determined attenuations calculated by means of a finite element analysis code. The code incorporates liner impedance values educed from measured data from the NASA Langley Grazing Incidence Tube, a test rig that is used for investigating liner performance with flow and with (0,0) mode incident grazing. The analytical and experimental results compare favorably, indicating the validity of the finite element method and demonstrating that finite element prediction tools can be used together with experiment to characterize the liner attenuation.

Gerhold, Carl H.↗

A decentralized linear quadratic control design method for flexible structures

A decentralized suboptimal linear quadratic control design procedure which combines substructural synthesis, model reduction, decentralized control design, subcontroller synthesis, and controller reduction is proposed for the design of reduced-order controllers for flexible structures. The procedure starts with a definition of the continuum structure to be controlled. An evaluation model of finite dimension is obtained by the finite element method. Then, the finite element model is decomposed into several substructures by using a natural decomposition called substructuring decomposition. Each substructure, at this point, still has too large a dimension and must be reduced to a size that is Riccati-solvable. Model reduction of each substructure can be performed by using any existing model reduction method, e.g., modal truncation, balanced reduction, Krylov model reduction, or mixed-mode method. Then, based on the reduced substructure model, a subcontroller is designed by an LQ optimal control method for each substructure independently. After all subcontrollers are designed, a controller synthesis method called substructural controller synthesis is employed to synthesize all subcontrollers into a global controller. The assembling scheme used is the same as that employed for the structure matrices. Finally, a controller reduction scheme, called the equivalent impulse response energy controller (EIREC) reduction algorithm, is used to reduce the global controller to a reasonable size for implementation. The EIREC reduced controller preserves the impulse response energy of the full-order controller and has the property of matching low-frequency moments and low-frequency power moments. An advantage of the substructural controller synthesis method is that it relieves the computational burden associated with dimensionality. Besides that, the SCS design scheme is also a highly adaptable controller synthesis method for structures with varying configuration, or varying mass and stiffness properties.

Su, Tzu-Jeng↗

Development of an Unstructured Mesh Code for Flows About Complete Vehicles

This report describes the research work undertaken at the Massachusetts Institute of Technology, under NASA Research Grant NAG4-157. The aim of this research is to identify effective algorithms and methodologies for the efficient and routine solution of flow simulations about complete vehicle configurations. For over ten years we have received support from NASA to develop unstructured mesh methods for Computational Fluid Dynamics. As a result of this effort a methodology based on the use of unstructured adapted meshes of tetrahedra and finite volume flow solvers has been developed. A number of gridding algorithms, flow solvers, and adaptive strategies have been proposed. The most successful algorithms developed from the basis of the unstructured mesh system FELISA. The FELISA system has been extensively for the analysis of transonic and hypersonic flows about complete vehicle configurations. The system is highly automatic and allows for the routine aerodynamic analysis of complex configurations starting from CAD data. The code has been parallelized and utilizes efficient solution algorithms. For hypersonic flows, a version of the code which incorporates real gas effects, has been produced. The FELISA system is also a component of the STARS aeroservoelastic system developed at NASA Dryden. One of the latest developments before the start of this grant was to extend the system to include viscous effects. This required the development of viscous generators, capable of generating the anisotropic grids required to represent boundary layers, and viscous flow solvers. We show some sample hypersonic viscous computations using the developed viscous generators and solvers. Although this initial results were encouraging it became apparent that in order to develop a fully functional capability for viscous flows, several advances in solution accuracy, robustness and efficiency were required. In this grant we set out to investigate some novel methodologies that could lead to the required improvements. In particular we focused on two fronts: (1) finite element methods and (2) iterative algebraic multigrid solution techniques.

Peraire, Jaime↗

Unstructured grid large eddy simulation of wall bounded turbulent flows

Historically, large eddy simulations (LES) have been restricted to simple geometries where spectral or finite difference methods have dominated due to their efficient use of structured grids. Structured grids, however, not only have difficulty representing complex domains and adapting to complicated flow features, but also are rather inefficient for simulating flows at high Reynolds numbers. The lack of efficiency stems from the need to resolve the viscous sub layer which requires very fine resolution in all three directions near the wall. Structured grids make use of a stretching to reduce the normal grid spacing but must carry the fine resolution in the streamwise and spanwise directions throughout the domain. The unnecessarily fine grid for much of the domain leads to disturbingly high grid estimates. Chapman (1979), and later Moin & Jimenez (1993), pointed out that, in order to advance the technology to airfoils at flight Reynolds numbers, structured grids must be abandoned in lieu of what are known as nested or unstructured grids. We illustrate the ability of an unstructured mesh to refine only the near wall region. Note the large number of points near the wall (where the fine vortical features need better resolution) and the coarseness in all directions away from the wall (where the scales are much larger). The important difference between this approach and the usual structured grid stretching is that the number of elements used to discretize the spanwise and streamwise features of the flow is reduced in each successive layer coming off the wall. This is due to the fact that the elements not only grow in the normal direction, but in the other directions as well. This greatly reduces the total number of points or elements required for a given Reynolds number flow.

Jansen, Kenneth↗

Fish-Friendly Hydropower Turbine Development & Deployment: Alden Turbine Preliminary Engineering and Model Testing

The Alden turbine was developed through the U.S. Department of Energy's (DOE's) former Advanced Hydro Turbine Systems Program (1994-2006) and, more recently, through the Electric Power Research Institute (EPRI) and the DOE's Wind & Water Power Program. The primary goal of the engineering study described here was to provide a commercially competitive turbine design that would yield fish passage survival rates comparable to or better than the survival rates of bypassing or spilling flow. Although the turbine design was performed for site conditions corresponding to 92 ft (28 m) net head and a discharge of 1500 cfs (42.5 cms), the design can be modified for additional sites with differing operating conditions. During the turbine development, design modifications were identified for the spiral case, distributor (stay vanes and wicket gates), runner, and draft tube to improve turbine performance while maintaining features for high fish passage survival. Computational results for pressure change rates and shear within the runner passage were similar in the original and final turbine geometries, while predicted minimum pressures were higher for the final turbine. The final turbine geometry and resulting flow environments are expected to further enhance the fish passage characteristics of the turbine. Computational results for the final design were shown to improve turbine efficiencies by over 6% at the selected operating condition when compared to the original concept. Prior to the release of the hydraulic components for model fabrication, finite element analysis calculations were conducted for the stay vanes, wicket gates, and runner to verify that structural design criteria for stress and deflections were met. A physical model of the turbine was manufactured and tested with data collected for power and efficiency, cavitation limits, runaway speed, axial and radial thrust, pressure pulsations, and wicket gate torque. All parameters were observed to fall within ranges expected for conventional radial flow machines. Based on these measurements, the expected efficiency peak for prototype application is 93.64%. These data were used in the final sizing of the supporting mechanical and balance of plant equipment. The preliminary equipment cost for the design specification is $1450/kW with a total supply schedule of 28 months. This equipment supply includes turbine, generator, unit controls, limited balance of plant equipment, field installation, and commissioning. Based on the selected head and flow design conditions, fish passage survival through the final turbine is estimated to be approximately 98% for 7.9-inch (200-mm) fish, and the predicted survival reaches 100% for fish 3.9 inches (100 mm) and less in length. Note that fish up to 7.9- inches (200 mm) in length make up more than 90% of fish entrained at hydro projects in the United States. Completion of these efforts provides a mechanical and electrical design that can be readily adapted to site-specific conditions with additional engineering development comparable to costs associated with conventional turbine designs.

13 HYDRO ENERGY↗

Efficient Reformulation of the Thermoelastic Higher-order Theory for Fgms

Functionally graded materials (FGMs) are characterized by spatially variable microstructures which are introduced to satisfy given performance requirements. The microstructural gradation gives rise to continuously or discretely changing material properties which complicate FGM analysis. Various techniques have been developed during the past several decades for analyzing traditional composites and many of these have been adapted for the analysis of FGMs. Most of the available techniques use the so-called uncoupled approach in order to analyze graded structures. These techniques ignore the effect of microstructural gradation by employing specific spatial material property variations that are either assumed or obtained by local homogenization. The higher-order theory for functionally graded materials (HOTFGM) is a coupled approach developed by Aboudi et al. (1999) which takes the effect of microstructural gradation into consideration and does not ignore the local-global interaction of the spatially variable inclusion phase(s). Despite its demonstrated utility, however, the original formulation of the higher-order theory is computationally intensive. Herein, an efficient reformulation of the original higher-order theory for two-dimensional elastic problems is developed and validated. The use of the local-global conductivity and local-global stiffness matrix approach is made in order to reduce the number of equations involved. In this approach, surface-averaged quantities are the primary variables which replace volume-averaged quantities employed in the original formulation. The reformulation decreases the size of the global conductivity and stiffness matrices by approximately sixty percent. Various thermal, mechanical, and combined thermomechanical problems are analyzed in order to validate the accuracy of the reformulated theory through comparison with analytical and finite-element solutions. The presented results illustrate the efficiency of the reformulation and its advantages in analyzing functionally graded materials.

Bansal, Yogesh↗

Dynamic equation-based thermo-hydraulic pipe model for district heating and cooling systems

Simulation and optimisation of district heating and cooling networks requires efficient and realistic models of the individual network elements in order to correctly represent heat losses or gains, temperature propagation and pressure drops. Due to more recent thermal networks incorporating meshing decentralised heat and cold sources, the system often has to deal with variable temperatures and mass flow rates, with flow reversal occurring more frequently. This paper presents the mathematical derivation and software implementation in Modelica of a thermo-hydraulic model for thermal networks that meets the above requirements and compares it to both experimental data and a commonly used model. Good correspondence between experimental data from a controlled test set-up and simulations using the presented model was found. Compared to measurement data from a real district heating network, the simulation results led to a larger error than in the controlled test set-up, but the general trend is still approximated closely and the model yields results similar to a pipe model from the Modelica Standard Library. However, the presented model simulates 1.7 (for low number of volumes) to 68 (for highly discretized pipes) times faster than a conventional model for a realistic test case. A working implementation of the presented model is made openly available within the IBPSA Modelica Library. The model is robust in the sense that grid size and time step do not need to be adapted to the flow rate, as is the case in finite volume models.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Toward large-scale simulation of residual stress and distortion in wire and arc additive manufacturing

In this study, we aim to advance the structural analysis of wire and arc additive manufacturing (WAAM) by considering the thermomechanical features inherent in direct energy deposition. Simulation approaches including the iterative substructure method (ISM), dynamic mesh refining method (DMRM), and graphics processing unit (GPU) based explicit finite element method (FEM) were developed for accelerating additive manufacturing stress analysis that is very time consuming by conventional numerical methods. ISM and DMRM take advantage of the strong nonlinearity phenomenon near the moving heat source by reducing the global iterations and refining the local mesh, respectively. In addition to the use of GPUs, the explicit FEM is accelerated by a time scaling technique based on the inherent strain concept. The residual stress and distortion of two large builds were analyzed, showing very consistent numerical results and good agreement with experiments. Compared with the commercial software Abaqus, the novel approaches reduced the computational cost substantially without compromising accuracy. The GPU code showed the highest computational efficiency (30~70×), while DMRM and ISM had acceleration factors of 9× and 3×, respectively. Finally, such high-fidelity modeling approaches will be very useful for building up a digital twin of WAAM to reduce development time and cost.

36 MATERIALS SCIENCE↗

Adaptive linear predictor FIR filter based on the Cyclone V FPGA with HPS to reduce narrow band RFI in AERA radio detection of cosmic rays

We present the new approach to a filtering of radio frequency interferences (RFI) in the Auger Engineering Radio Array (AERA) which study the electromagnetic part of the Extensive Air Showers. The radio stations can observe radio signals caused by coherent emissions due to geomagnetic radiation and charge excess processes. AERA observes frequency band from 30 to 80 MHz. This range is highly contaminated by human-made RFI. In order to improve the signal to noise ratio RFI filters are used in AERA to suppress this contamination. The first kind of filter used by AERA was the Median one, based on the Fast Fourier Transform (FFT) technique. The second one, which is currently in use, is the infinite impulse response (IIR) notch filter. The proposed new filter is a finite impulse response (FIR) filter based on a linear prediction (LP). A periodic contamination hidden in a registered signal (digitized in the ADC) can be extracted and next subtracted to make signal cleaner. The FIR filter requires a calculation of n=32, 64 or even 128 coefficients (dependent on a required speed or accuracy) by solving of n linear equations with coefficients built from the covariance Toeplitz matrix. This matrix can be solved by the Levinson recursion, which is much faster than the Gauss procedure. The filter has been already tested in the real AERA radio stations on Argentinean pampas with a very successful results. The linear equations were solved either in the virtual soft-core NIOSR processor (implemented in the FPGA chip as a net of logic elements) or in the external Voipac PXA270M ARM processor. The NIOS processor is relatively slow (50 MHz internal clock), calculations performed in an external processor consume a significant amount of time for data exchange between the FPGA and the processor. Test showed a very good efficiency of the RFI suppression for stationary (long-term) contaminations. However, we observed a short-time contaminations, which could not be suppressed either by the IIR-notch filter or by the FIR filter based on the linear predictions. For the LP FIR filter the refreshment time of the filter coefficients was to long and filter did not keep up with the changes of a contamination structure, mainly due to a long calculation time in a slow processors. We propose to use the Cyclone V SE chip with embedded micro-controller operating with 925 MHz internal clock to significantly reduce a refreshment time of the FIR coefficients. The lab results are promising. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Bayesian sequential optimal experimental design for nonlinear models using policy gradient reinforcement learning

We present a mathematical framework and computational methods for optimally designing a finite sequence of experiments. This sequential optimal experimental design (sOED) problem is formulated as a finite-horizon partially observable Markov decision process (POMDP) under a Bayesian setting and with information-theoretic utilities. The formulation is general and may accommodate continuous random variables, non-Gaussian posteriors, and nonlinear forward models. The sOED design policy incorporates elements of feedback and lookahead simultaneously, and we show it to generalize the commonly-used batch and greedy design strategies. We solve for the sOED policy using the policy gradient (PG) method from reinforcement learning, and provide a derivation for the PG expression in the sOED context. Adopting an actor-critic approach, the policy and value functions are parameterized using deep neural networks and improved via PG estimates produced from simulated episodes of designs and observations. The new PG-sOED algorithm is first validated on a linear-Gaussian benchmark, and then compared against other design baselines on a sensor movement problem for contaminant source inversion in a convection-diffusion field. As a result, we provide explanation for the policy behaviors using knowledge of the underlying physical process.

97 MATHEMATICS AND COMPUTING↗