Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Time step adaptivity”

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 199 records · Page 11

Implementation of an Open-Scenario, Long-Term Space Debris Simulation Approach

This paper provides a status update on the implementation of a flexible, long-term space debris simulation approach. The motivation is to build a tool that can assess the long-term impact of various options for debris-remediation, including the LightForce space debris collision avoidance concept that diverts objects using photon pressure [9]. State-of-the-art simulation approaches that assess the long-term development of the debris environment use either completely statistical approaches, or they rely on large time steps on the order of several days if they simulate the positions of single objects over time. They cannot be easily adapted to investigate the impact of specific collision avoidance schemes or de-orbit schemes, because the efficiency of a collision avoidance maneuver can depend on various input parameters, including ground station positions and orbital and physical parameters of the objects involved in close encounters (conjunctions). Furthermore, maneuvers take place on timescales much smaller than days. For example, LightForce only changes the orbit of a certain object (aiming to reduce the probability of collision), but it does not remove entire objects or groups of objects. In the same sense, it is also not straightforward to compare specific de-orbit methods in regard to potential collision risks during a de-orbit maneuver. To gain flexibility in assessing interactions with objects, we implement a simulation that includes every tracked space object in Low Earth Orbit (LEO) and propagates all objects with high precision and variable time-steps as small as one second. It allows the assessment of the (potential) impact of physical or orbital changes to any object. The final goal is to employ a Monte Carlo approach to assess the debris evolution during the simulation time-frame of 100 years and to compare a baseline scenario to debris remediation scenarios or other scenarios of interest. To populate the initial simulation, we use the entire space-track object catalog in LEO. We then use a high precision propagator to propagate all objects over the entire simulation duration. If collisions are detected, the appropriate number of debris objects are created and inserted into the simulation framework. Depending on the scenario, further objects, e.g. due to new launches, can be added. At the end of the simulation, the total number of objects above a cut-off size and the number of detected collisions provide benchmark parameters for the comparison between scenarios. The simulation approach is computationally intensive as it involves tens of thousands of objects; hence we use a highly parallel approach employing up to a thousand cores on the NASA Pleiades supercomputer for a single run. This paper describes our simulation approach, the status of its implementation, the approach to developing scenarios and examples of first test runs.

laser↗

Adaptive workflow for simulation of RF heaters

Accurate RF (Radio Frequency) simulations of fusion systems like ITER require the definition of high-fidelity analysis geometries that include detailed antenna, reactor wall, and physics regions. Here, this paper will describe a workflow for the execution of adaptive high-performance simulations of RF fusion systems. In this workflow, the simulation input consists of a CAD model attributed with the needed analysis attributes. The analysis mesh is automatically generated and the analysis steps are executed using the time-harmonic Maxwell's equations solved using high-order Nédélec finite elements. A patch recovery-based error estimator is used to drive a conforming mesh adaptation procedure.

97 MATHEMATICS AND COMPUTING↗

Flame acceleration and transition to detonation in a pre-/main-chamber combustion system

Numerical simulations are performed to study the mechanism of deflagration to detonation transition (DDT) in a pre-/main-chamber combustion system with a stoichiometric ethylene–oxygen mixture. A Godunov algorithm, fifth-order in space, and third-order in time, is used to solve the fully compressible Navier–Stokes equations on a dynamically adapting mesh. A single-step, calibrated chemical diffusive model described by Arrhenius kinetics is used for energy release and conservation between the fuel and the product. The two-dimensional simulation shows that a laminar flame grows in the pre-chamber and then develops into a jet flame as it passes through the orifice. A strong shock forms immediately ahead of the flame, reflecting off the walls and interacting with the flame front. The shock–flame interactions are crucial for the development of flame instabilities, which trigger the subsequent flame development. The DDT arises due to a shock-focusing mechanism, where multiple shocks collide at the flame front. A chemical explosive mode analysis (CEMA) criterion is developed to study the DDT ignition mode. Preliminary one-dimensional computations for a laminar propagating flame, a fast flame deflagration, and a Chapman–Jouguet detonation are conducted to demonstrate the validity of CEMA on the chemical-diffusive model, as well as to determine the proper conditioning value for CEMA diagnostic. The two-dimensional analysis with CEMA indicates that the DDT initiated by the shock-focusing mechanism can form a strong thermal expansion region at the flame front that features large positive eigenvalues for the chemical explosive mode and dominance of the local autoignition mode. Thus, the CEMA criterion proposed in this study provides a robust diagnostic for identifying autoignition-supported DDT, of which the emergence of excessive local autoignition mode is found to be a precursor. The effect of grid size, initial temperature, and orifice size are then evaluated, and results show that although the close-chamber DDT is highly stochastic, the detonation initiation mechanism remains robust.

42 ENGINEERING↗

Work Experience Report

The NASA Platform for Autonomous Systems (NPAS) toolkit is currently being used at the NASA John C. Stennis Space Center (SSC) to develop the INSIGHT program, which will autonomously monitor and control the Nitrogen System of the High Pressure Gas Facility (HPGF) on site. The INSIGHT program is in need of generic timing capabilities in order to perform timing based actions such as pump usage timing and sequence step timing. The purpose of this project was to develop a timing module that could fulfill these requirements and be adaptable for expanded use in the future. The code was written in Gensym G2 software platform, the same as INSIGHT, and was written generically to ensure compatibility with any G2 program. Currently, the module has two timing capabilities, a stopwatch function and a countdown function. Although the module has gone through some functionality testing, actual integration of the module into NPAS and the INSIGHT program is contingent on the module passing later checks.

Guo, Daniel↗

Adaptive Data Processing Technique for Lidar-Assisted Control to Bridge the Gap between Lidar Systems and Wind Turbines: Preprint

This paper presents first steps toward an adaptive lidar data processing technique crucial for lidar-assisted control in wind turbines. The prediction time and the quality of the wind preview from lidar measurements depend on several factors and are not constant. If the data processing is not continually adjusted, the benefit of lidar-assisted control cannot be fully exploited, or can even result in harmful control action. An online analysis of the lidar and turbine data are necessary to continually reassess the prediction time and lidar data quality. In this work, a structured process to develop an analysis tool for the prediction time and a new hardware setup for lidar-assisted control are presented. The tool consists of an online estimation of the rotor effective wind speed from lidar and turbine data and the implementation of an online cross correlation to determine the time shift between both signals. Further, initial results from an ongoing campaign in which this system was employed for providing lidar preview for feed-forward pitch control are presented.

17 WIND ENERGY↗

Adaptive X-ray imaging with reinforcement learning

X-ray imaging is a powerful technique to scan samples in a variety of contexts including biological, environmental and materials science, but commonly requires a synchrotron light source to produce X-rays at sufficient intensity. As these facilities are expensive to operate, the available beam time is limited and always in high demand. Particularly if the illuminated samples are sparse, standard raster scanning methods can be time-consuming, with a majority of that time being spent on areas of the image that carry little information. To increase the efficiency and maximize the information gain for a given time budget, we split the scanning process into a series of steps where previous measurements are used to inform the decision making and adapt the exposure distribution at later stages of the sequence. We formulate this task as a reinforcement learning problem where the goal is to produce a sequence of exposure maps that maximize a predefined scalar metric. We demonstrate the potential of this approach in simulations where the adaptive illumination can accelerate the measurement process by up to an order of magnitude compared with standard raster scanning. Finally, we present the first results from deploying the trained agents on an X-ray fluorescence beamline at the Stanford Synchrotron Radiation Lightsource.

Reinforcement Learning↗

Optimal checkpointing for adjoint multistage time-stepping schemes

Here, we consider checkpointing strategies that minimize the number of recomputations needed when performing discrete adjoint computations using multistage time-stepping schemes that require computing several substeps within one complete time step. Specifically, we propose two algorithms that can generate optimal checkpoint-ing schedules under weak assumptions. The first is an extension of the seminal Revolve algorithm adapted to multistage schemes. The second algorithm, named CAMS, is developed based on dynamic programming, and it requires the least number of recomputations when compared with other algorithms. The CAMS algorithm is made publicly available in a library with bindings to C and Python. Numerical results show that the proposed algorithms can deliver up to two times the speedup compared with that of classical Revolve. Moreover, we discuss the utilization of the CAMS library in mature scientific computing libraries and demonstrate the ease of using it in an adjoint workflow. The proposed algorithms have been adopted by the PETSc TSAdjoint library. Their performance has been demonstrated with a large-scale PDE-constrained optimization problem on a leadership-class supercomputer. This work is a significant extension of the authors' conference paper.

97 MATHEMATICS AND COMPUTING↗

Computational diagnostics for flame acceleration and transition to detonation in a hydrogen/air mixture

A new computational diagnostic method for pressure-induced compressibility is proposed by projecting its local contribution to the chemical explosive mode (CEM) in the chemical explosive mode analysis (CEMA) framework. The new method is validated for the study of detonation development during the deflagration-to-detonation transition (DDT) process. The flame characteristics are identified through the quantification of individual CEM contributions of chemical reaction, diffusion, and pressure-induced compressibility. Numerical simulations are performed to investigate the DDT processes in a stoichiometric hydrogen-air mixture. A Godunov algorithm, fifth-order in space, and third-order in time are used to solve the fully compressible Navier-Stokes equations on a dynamically adapting mesh. A single-step, calibrated chemical diffusive model (CDM) described by Arrhenius kinetics is used for energy release and conservation between the fuel and the product. The new diagnostic method is first applied to onedimensional (1D) canonical flame configurations followed by two-dimensional (2D) simulations of DDT in an obstructed channel where different detonation initiation scenarios are examined using the new CEMA projection formulation. Detailed examinations of the idealized configuration of detonation initiation through shock focusing mechanism at a flame front are also studied using the new formulation. A comparison of the currently proposed CEMA projection and the original formulation by the authors suggests that including the pressure-induced compressibility is essential for the use of CEMA in DDT process. The results also show that the new formulation of CEMA projection can successively capture the detonation initiation through either a gradient mechanism or a direct initiation mechanism, and therefore can be used as an effective local analytical tool for the computational diagnostics of detonation initiation in a DDT process. It was found that detonation development is characterized by a strong contribution of chemistry role to the CEM which is pivotal to the initiation of detonation. The role of compressibility is found enhanced at the edge of the detonation front where diffusion was found to have minimal effects on detonation development.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Numerical Computation of a Continuous-thrust State Transition Matrix Incorporating Accurate Hardware and Ephemeris Models

A significant body of work exists showing that providing a nonlinear programming (NLP) solver with expressions for the problem constraint gradient substantially increases the speed of program execution and can also improve the robustness of convergence, especially for local optimizers. Calculation of these derivatives is often accomplished through the computation of spacecraft's state transition matrix (STM). If the two-body gravitational model is employed as is often done in the context of preliminary design, closed form expressions for these derivatives may be provided. If a high fidelity dynamics model, that might include perturbing forces such as the gravitational effect from multiple third bodies and solar radiation pressure is used then these STM's must be computed numerically. We present a method for the power hardward model and a full ephemeris model. An adaptive-step embedded eight order Dormand-Prince numerical integrator is discussed and a method for the computation of the time of flight derivatives in this framework is presented. The use of these numerically calculated derivatieves offer a substantial improvement over finite differencing in the context of a global optimizer. Specifically the inclusion of these STM's into the low thrust missiondesign tool chain in use at NASA Goddard Spaceflight Center allows for an increased preliminary mission design cadence.

Low Thrust↗

An adaptive finite element procedure for compressible high speed flows

A practical finite element-based solution procedure for high-speed inviscid compressible flow problems is described. The method provides time-accurate solutions to the coompressible Euler equations, and is computationally more efficient than the one-step Taylor-Galerkin approach and better suited for implementation on the modern generation of vector computers. The method is coupled to an adaptive mesh refinement process that enables steady state solutions of improved quality to be obtained.

Loehner, R.↗

Using Tests Designed to Measure Individual Sensorimotor Subsystem Perfomance to Predict Locomotor Adaptability

Astronauts experience sensorimotor disturbances during the initial exposure to microgravity and during the readapation phase following a return to a gravitational environment. These alterations may lead to disruption in the ability to perform mission critical functions during and after these gravitational transitions. Astronauts show significant inter-subject variation in adaptive capability following gravitational transitions. The way each individual's brain synthesizes the available visual, vestibular and somatosensory information is likely the basis for much of the variation. Identifying the presence of biases in each person's use of information available from these sensorimotor subsystems and relating it to their ability to adapt to a novel locomotor task will allow us to customize a training program designed to enhance sensorimotor adaptability. Eight tests are being used to measure sensorimotor subsystem performance. Three of these use measures of body sway to characterize balance during varying sensorimotor challenges. The effect of vision is assessed by repeating conditions with eyes open and eyes closed. Standing on foam, or on a support surface that pitches to maintain a constant ankle angle provide somatosensory challenges. Information from the vestibular system is isolated when vision is removed and the support surface is compromised, and it is challenged when the tasks are done while the head is in motion. The integration and dominance of visual information is assessed in three additional tests. The Rod & Frame Test measures the degree to which a subject's perception of the visual vertical is affected by the orientation of a tilted frame in the periphery. Locomotor visual dependence is determined by assessing how much an oscillating virtual visual world affects a treadmill-walking subject. In the third of the visual manipulation tests, subjects walk an obstacle course while wearing up-down reversing prisms. The two remaining tests include direct measures of knee and ankle proprioception and a functional movement assessment that screens for movement restrictions and asymmetries. To assess each subject's locomotor adaptability subjects walk for twenty minutes on a treadmill that oscillates laterally at 0.3 Hz. Throughout the test metabolic cost provides a measure of exertion and step frequency provides a measure of stability. Additionally, at four points during the perturbation period, reaction time tests are used to probe changes in the amount of mental effort being used to perform the task. As with the adaptive capability observed in astronauts during gravitational transitions, our data shows significant variability between subjects. To aid in the analysis of the results, custom software tools have been developed to enhance in the visualization of the large number of output variables. Preliminary analyses of the data collected to date do not show a strong relationship between adaptability and any single predictor variable. Analysis continues to identify a multifactorial predictor outcome "signature" that do inform us of locomotor adaptability.

Peters, B. T.↗

Adaptive Stress Testing: Finding Likely Failure Events with Reinforcement Learning

Finding the most likely path to a set of failure states is important to the analysis of safety-critical systems that operate over a sequence of time steps, such as aircraft collision avoidance systems and autonomous cars. In many applications such as autonomous driving, failures cannot be completely eliminated due to the complex stochastic environment in which the system operates.As a result, safety validation is not only concerned about whether a failure can occur, but also discovering which failures are most likely to occur. This article presents adaptive stress testing (AST), a framework for finding the most likely path to a failure event in simulation. We consider a general black box setting for partially observable and continuous-valued systems operating in an environment with stochastic disturbances. We formulate the problem as a Markov decision process and use reinforcement learning to optimize it. The approach is simulation-based and does not require internal knowledge of the system, making it suitable for black-box testing of large systems. We present different formulations depending on whether the state is fully observable or partially observable. In the latter case, we present a modified Monte Carlo tree search algorithm that only requires access to the pseudorandom number generator of the simulator to overcome partial observability. We also present an extension of the framework, called differential adaptive stress testing (DAST), that can find failures that occur in one system but not in another. This type of differential analysis is useful in applications such as regression testing, where we are concerned with finding areas of relative weakness compared to a baseline. We demonstrate the effectiveness of the approach on an aircraft collision avoidance application, where a prototype aircraft collision avoidance system is stress tested to find the most likely scenarios of near mid-air collision.

Verification and Validation↗

Finite element implementation of state variable-based viscoplasticity models

The implementation of state variable-based viscoplasticity models is made in a general purpose finite element code for structural applications of metals deformed at elevated temperatures. Two constitutive models, Walker's and Robinson's models, are studied in conjunction with two implicit integration methods: the trapezoidal rule with Newton-Raphson iterations and an asymptotic integration algorithm. A comparison is made between the two integration methods, and the latter method appears to be computationally more appealing in terms of numerical accuracy and CPU time. However, in order to make the asymptotic algorithm robust, it is necessary to include a self adaptive scheme with subincremental step control and error checking of the Jacobian matrix at the integration points. Three examples are given to illustrate the numerical aspects of the integration methods tested.

Iskovitz, I.↗

Finite element implementation of state variable-based viscoplasticity models

The implementation of state variable-based viscoplasticity models is made in a general purpose finite element code for structural applications of metals deformed at elevated temperatures. Two constitutive models, Walker's and Robinson's models, are studied in conjunction with two implicit integration methods: the trapezoidal rule with Newton-Raphson iterations and an asymptotic integration algorithm. A comparison is made between the two integration methods, and the latter method appears to be computationally more appealing in terms of numerical accuracy and CPU time. However, in order to make the asymptotic algorithm robust, it is necessary to include a self adaptive scheme with subincremental step control and error checking of the Jacobian matrix at the integration points. Three examples are given to illustrate the numerical aspects of the integration methods tested.

Chang, T. Y. P.↗

Toward fully self-consistent simulation of the interaction of E-Clouds and beams with WARP-POSINST

To predict the evolution of electron clouds and their effect on the beam, the high energy physics community has relied so far on the complementary use of 'buildup' and 'single/multi-bunch instability' reduced descriptions. The former describes the evolution of electron clouds at a given location in the ring, or 'station', under the influence of prescribed beams and external fields [1], while the latter (sometimes also referred as the 'quasi-static' approximation [2]) follows the interaction between the beams and the electron clouds around the accelerator with prescribed initial distributions of electrons, assumed to be concentrated at a number of discrete 'stations' around the ring. Examples of single bunch instability codes include HEADTAIL [3], QuickPIC [4, 5], and PEHTS [6]. By contrast, a fully self-consistent approach, in which both the electron cloud and beam distributions evolve simultaneously under their mutual influence without any restriction on their relative motion, is required for modeling the interaction of high-intensity beams with electron clouds for heavy-ion beam-driven fusion and warm-dense matter science. This community has relied on the use of Particle-In-Cell (PIC) methods through the development and use of the WARP-POSINST code suite [1, 7, 8]. The development of novel numerical techniques (including adaptive mesh refinement, and a new 'drift-Lorentz' particle mover for tracking charged particles in magnetic fields using large time steps) has enabled the first application of WARP-POSINST to the fully self-consistent modeling of beams and electron clouds in high energy accelerators [9], albeit for only a few betatron oscillations. It was recently observed [10] that there exists a preferred frame of reference which minimizes the number of computer operations needed to simulate the interaction of relativistic objects. This opens the possibility of reducing the cost of fully self-consistent simulations for the interaction of ultrarelativistic beams with electron cloud by orders of magnitude. The computational cost of the fully self-consistent mode is then predicted to be comparable to that of the quasi-static mode, assuming that several stations per betatron period are needed. During the workshop, there was some debate about the number of stations per betatron period that are needed when using the quasi-static mode. The argument was made that if there is less than one station per betatron period, then artificial resonances can be triggered and the resulting emittance growth provides an upper bound. The emittance growth thus obtained will fall either above or below the operational requirements of the machine. In the latter case, one can conclude that the electron effect that has been simulated is of no concern. However, if the emittance growth that was obtained is above the threshold, then the results become inconclusive, and simulations which resolve the betatron motion are then needed. In this case, according to [10], the fully self-consistent approach becomes an option. The aim of this paper is to investigate whether this option is indeed practical.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Four-dimensional dose distributions of step-and-shoot IMRT delivered with real-time tumor tracking for patients with irregular breathing: Constant dose rate vs dose rate regulation

Purpose: Dose-rate-regulated tracking (DRRT) is a tumor tracking strategy that programs the MLC to track the tumor under regular breathing and adapts to breathing irregularities during delivery using dose rate regulation. Constant-dose-rate tracking (CDRT) is a strategy that dynamically repositions the beam to account for intrafractional 3D target motion according to real-time information of target location obtained from an independent position monitoring system. The purpose of this study is to illustrate the differences in the effectiveness and delivery accuracy between these two tracking methods in the presence of breathing irregularities. Methods: Step-and-shoot IMRT plans optimized at a reference phase were extended to remaining phases to generate 10-phased 4D-IMRT plans using segment aperture morphing (SAM) algorithm, where both tumor displacement and deformation were considered. A SAM-based 4D plan has been demonstrated to provide better plan quality than plans not considering target deformation. However, delivering such a plan requires preprogramming of the MLC aperture sequence. Deliveries of the 4D plans using DRRT and CDRT tracking approaches were simulated assuming the breathing period is either shorter or longer than the planning day, for 4 IMRT cases: two lung and two pancreatic cases with maximum GTV centroid motion greater than 1 cm were selected. In DRRT, dose rate was regulated to speed up or slow down delivery as needed such that each planned segment is delivered at the planned breathing phase. In CDRT, MLC is separately controlled to follow the tumor motion, but dose rate was kept constant. In addition to breathing period change, effect of breathing amplitude variation on target and critical tissue dose distribution is also evaluated. Results: Delivery of preprogrammed 4D plans by the CDRT method resulted in an average of 5% increase in target dose and noticeable increase in organs at risk (OAR) dose when patient breathing is either 10% faster or slower than the planning day. In contrast, DRRT method showed less than 1% reduction in target dose and no noticeable change in OAR dose under the same breathing period irregularities. When {+-}20% variation of target motion amplitude was present as breathing irregularity, the two delivery methods show compatible plan quality if the dose distribution of CDRT delivery is renormalized. Conclusions: Delivery of 4D-IMRT treatment plans, stemmed from 3D step-and-shoot IMRT and preprogrammed using SAM algorithm, is simulated for two dynamic MLC-based real-time tumor tracking strategies: with and without dose-rate regulation. Comparison of cumulative dose distribution indicates that the preprogrammed 4D plan is more accurately and efficiently conformed using the DRRT strategy, as it compensates the interplay between patient breathing irregularity and tracking delivery without compromising the segment-weight modulation.

60 APPLIED LIFE SCIENCES↗

Dual adaptive control: Design principles and applications

The design of an actively adaptive dual controller based on an approximation of the stochastic dynamic programming equation for a multi-step horizon is presented. A dual controller that can enhance identification of the system while controlling it at the same time is derived for multi-dimensional problems. This dual controller uses sensitivity functions of the expected future cost with respect to the parameter uncertainties. A passively adaptive cautious controller and the actively adaptive dual controller are examined. In many instances, the cautious controller is seen to turn off while the latter avoids the turn-off of the control and the slow convergence of the parameter estimates, characteristic of the cautious controller. The algorithms have been applied to a multi-variable static model which represents a simplified linear version of the relationship between the vibration output and the higher harmonic control input for a helicopter. Monte Carlo comparisons based on parametric and nonparametric statistical analysis indicate the superiority of the dual controller over the baseline controller.

Mookerjee, Purusottam↗

Quantum simulations of hadron dynamics in the Schwinger model using 112 qubits

Hadron wave packets are prepared and time evolved in the Schwinger model using 112 qubits of IBM’s 133-qubit Heron quantum computer ibm_torino. The initialization of the hadron wave packet is performed in two steps. First, the vacuum is prepared across the whole lattice using the recently developed SC-ADAPT-VQE algorithm and workflow. SC-ADAPT-VQE is then extended to the preparation of localized states, and used to establish a hadron wave packet on top of the vacuum. This is done by adaptively constructing low-depth circuits that maximize the overlap with an adiabatically prepared hadron wave packet. Due to the localized nature of the wavepacket, these circuits can be determined on a sequence of small lattices using classical computers, and then robustly scaled to prepare wave packets on large lattices for simulations using quantum computers. Time evolution is implemented with a second-order Trotterization. To reduce both the required qubit connectivity and circuit depth, an approximate quasilocal interaction is introduced. This approximation is made possible by the emergence of confinement at long distances, and converges exponentially with increasing distance of the interactions. Using multiple error-mitigation strategies, up to 14 Trotter steps of time evolution are performed, employing 13,858 two-qubit gates (with a CNOT depth of 370). The propagation of hadrons is clearly identified, with results that compare favorably with Matrix Product State simulations. Finally, prospects for a near-term quantum advantage in simulations of hadron scattering are discussed.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗