Search NASA⌕ Search

SEARCH · Search NASA

Results for “hybrid algorithm”

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 685 records · Page 38

Physics-Based Methods of Failure Analysis and Diagnostics in Human Space Flight

The Integrated Health Management (IHM) for the future aerospace systems requires to interface models of multiple subsystems in an efficient and accurate information environment at the earlier stages of system design. The complexity of modern aeronautic and aircraft systems (including e.g. the power distribution, flight control, solid and liquid motors) dictates employment of hybrid models and high-level reasoners for analysing mixed continuous and discrete information flow involving multiple modes of operation in uncertain environments, unknown state variables, heterogeneous software and hardware components. To provide the information link between key design/performance parameters and high-level reasoners we rely on development of multi-physics performance models, distributed sensors networks, and fault diagnostic and prognostic (FD&P) technologies in close collaboration with system designers. The main challenges of our research are related to the in-flight assessment of the structural stability, engine performance, and trajectory control. The main goal is to develop an intelligent IHM that not only enhances components and system reliability, but also provides a post-flight feedback helping to optimize design of the next generation of aerospace systems. Our efforts are concentrated on several directions of the research. One of the key components of our strategy is an innovative approach to the diagnostics/prognostics based on the real time dynamical inference (DI) technologies extended to encompass hybrid systems with hidden state trajectories. The major investments are into the multiphysics performance modelling that provides an access of the FD&P technologies to the main performance parameters of e.g. solid and liquid rocket motors and composite materials of the nozzle and case. Some of the recent results of our research are discussed in this chapter. We begin by introducing the problem of dynamical inference of stochastic nonlinear models and reviewing earlier results. Next, we present our analytical approach to the solution of this problem based on the path integral formulation. The resulting algorithm does not require an extensive global search for the model parameters, provides optimal compensation for the effects of dynamical noise, and is robust for a broad range of dynamical models. In the following Section the strengths of the algorithm are illustrated illustrated by inferring the parameters of the stochastic Lorenz system and comparing the results with those of earlier research. Next, we discuss a number of recent results in application to the development of the IHM for aerospace system. Firstly, we apply dynamical inference approach to a solution of classical three tank problems with mixed unknown continuous and binary parameters. The problem is considered in the context of ground support system for filling fuel tanks of liquid rocket motors. It is shown that the DI algorithm is well suited for successful solution of a hybrid version of this benchmark problem even in the presence of additional periodic and stochastic perturbation of unknown strength. Secondly, we illustrate our approach by its application to an analysis of the nozzle fault in a solid rocket motor (SRM). The internal ballistics of the SRM is modelled as a set of one-dimensional partial differential equations coupled to the dynamics of the propellant regression. In this example we are specifically focussed on the inference of discrete and continuous parameters of the nozzle blocking fault and on the possibility of an application of the DI algorithm to reducing the probability of "misses" of an on-board FD&P for SRM. In the next section re-contact problem caused by first stage/upper stage separation failure is discussed. The reaction forces imposed on the nozzle of the upper stage during the re-contact and their connection to the nozzle damage and to the thrust vector control (TVC) signal are obtained. It is shown that transient impact induced torquean be modelled as a response of an effective damped oscillator. A possible application of the DI algorithm to the inference of damage parameters and predicting fault dynamics ahead of time using the actuator signal is discussed. Finally, we formulate Bayesian inferential framework for development of the IHM system for in-flight structural health monitoring (SHM) of composite materials. We consider the signal generated by piezoelectric actuator mounted on composite structure generating elastic waves in it. The signal received by the sensor is than compared with the baseline signal. The possibility of damage inference is discussed in the context of development of the SHM.

Smelyanskiy, Vadim N.↗

Hybrid Models of Reactive Transport in Porous and Fractured Media

Darcy-scale models of flow and transport in porous media often fail to describe experimentally observed phenomena, while their pore-scale counterparts are accu- rate but can be computationally prohibitive. Most numerical multi-scale models, which seek to combine these two descriptions, require empirical closures and/or assumptions on the behavior of pore-scale quantities at the continuum (Darcy) scale. We present a general formulation of an iterative hybrid numerical method that links these two scales without resorting to such approximations. The algorithm treats the fluxes exchanged at the internal boundaries between the pore- and continuum-scale domains as unknown, and allows for iteratively determined boundary conditions to be applied at the pore-scale in order to guarantee their continuity. While the algorithm proposed is general, we use it to model Taylor dispersion in a fracture with chemically reactive walls. Results show significant improvement upon standard continuum-scale formulations.

54 ENVIRONMENTAL SCIENCES↗

Energy-Efficient Cruise Control Using Optimal Control for a Hybrid Electric Vehicle

Increasing connectivity in passenger vehicles provides for a large amount of look-ahead information about driving conditions. An intelligent control algorithm is presented that takes advantage of this information to obtain the operation strategy for the powertrain of a parallel hybrid electric vehicle, in an uncongested highway cruising situation. In order to guarantee sufficient computational efficiency to meet future online requirements, the algorithm is based on Pontryagin’s Minimum Principle. The whole driving/operation strategy is composed of a series of solutions to the optimal control sub-problem for each separate route segment. The control sequence computed offline is then evaluated in Autonomie. The simulation result shows 6% fuel savings compared to a baseline rule-based controller with no speed optimization.

Shen, Daliang↗

Adapting Wave-front Algorithms to Efficiently Utilize Systems with Deep Communication Hierarchies

Large-scale systems increasingly exhibit a differential between intra-chip and inter-chip communication performance especially in hybrid systems using accelerators. Processorcores on the same socket are able to communicate at lower latencies, and with higher bandwidths, than cores on different sockets either within the same node or between nodes. A key challenge is to efficiently use this communication hierarchy and hence optimize performance. We consider here the class of applications that contains wavefront processing. In these applications data can only be processed after their upstream neighbors have been processed. Similar dependencies result between processors in which communication is required to pass boundary data downstream and whose cost is typically impacted by the slowest communication channel in use. In this work we develop a novel hierarchical wave-front approach that reduces the use of slower communications in the hierarchy but at the cost of additional steps in the parallel computation and higher use of on-chip communications. This tradeoff is explored using a performance model. An implementation using the Reverse-acceleration programming model on the petascale Roadrunner system demonstrates a 27% performance improvement at full system-scale on a kernel application. The approach is generally applicable to large-scale multi-core and accelerated systems where a differential in system communication performance exists.

43 PARTICLE ACCELERATORS↗

Efficient Hybrid Attack Graph Generation for Cyber-Physical System Resilience Experimentation (Final Project Report)

HAGEN project has developed theory, algorithms, and capabilities to assist cyber physical system modelers and operators to perform system and device-level vulnerability assessment, risk assessment, impact assessment, and mitigation planning. The project generates hybrid attack graphs for Cyber-Physical System (CPS) resilience experimentation at desired scale and speed. The project will produce composite attack datasets, algorithms, and demonstrable prototypical tools, and a library of high-impact attack sequences for a given CPS of interest. This report provided overall summary of research and development performed between FY22-24.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

A hybrid technique for spacecraft attitude interpolation with arbitrary attitude data gaps

An algorithm for obtaining sufficiently accurate estimates of a system parameter Y, given a data record of the parameter in which at arbitrary times no direct observations of Y exist, and describing the variance of these estimates, is proposed. The method provides estimates at all requested times, and the variance of the estimates is maintained as close to the variance of directly measured data as the quality of input information will allow. Three classes of information are used: (1) the measured data record of Y and the statistical extension of this record into time intervals of no direct measurement, obtained by a modification of standard autocorrelation techniques; (2) an a priori indicator of the behavior of Y; and (3) one or more indirect indicators of Y based on observation of other system parameters. The application of the algorithm to the SEASAT-A attitude determination under conditions of sensor field-of-view limitations and sun occultations is presented.

Treder, A. J.↗

Pattern recognition analysis of polar clouds during summer and winter

A pattern recognition algorithm is demonstrated which classifies eighteen surface and cloud types in high-latitude AVHRR imagery based on several spectral and textural features, then estimates the cloud properties (fractional coverage, albedo, and brightness temperature) using a hybrid histogram and spatial coherence technique. The summertime version of the algorithm uses both visible and infrared data (AVHRR channels 1-4), while the wintertime version uses only infrared data (AVHRR channels 3-5). Three days of low-resolution AVHRR imagery from the Arctic and Antarctic during January and July 1984 were analyzed for cloud type and fractional coverage. The analysis showed significant amounts of high cloudiness in the Arctic during one day in winter. The Antarctic summer scene was characterized by heavy cloud cover in the southern ocean and relatively clear conditions in the continental interior. A large region of extremely low brightness temperatures in East Antarctica during winter suggests the presence of polar stratospheric cloud.

Ebert, Elizabeth E.↗

Domain Compilation for Embedded Real-Time Planning

A recently conceived approach to automated real-time control of the actions of a robotic system enables an embedded real-time planning algorithm to develop plans that are more robust than they would otherwise be, without imposing an excessive computational burden. This approach occupies a middle ground between two prior approaches known in the art as the universal-plan and hybrid approaches. Ever since discovering the performance limitations of taking a sense-plan-act approach to controlling robots, the robotics community has endeavored to follow a behavior-based approach in which a behavior includes a rapid feedback loop between state estimation and motor control. Heretofore, system architectures following this approach have been based, variously, on algorithms that implement universal plans or algorithms that function as hybrids of planners and executives. In a typical universal-plan case, a set of behaviors is merged into the plan, but the system must be restricted to relatively small problem domains to avoid having to reason about too many states and represent them in the plan. In the hybrid approach, one implements actions as small sets of behaviors, each applicable to a limited set of circumstances. Each action is intended to bring the system to a subgoal state. A planning algorithm is used to string these actions together into a sequence to traverse the state space from an initial or current state to a goal state. The hybrid approach works well in a static environment, but it is inherently brittle in a dynamic environment because a failure can occur when the environment strays beyond the region of applicability of the current activity. In the present approach, a system can vary from the hybrid approach to the universal-plan approach, depending on a single integer parameter, denoted n, which can range from 1 to a maximum domain-dependent value of M. As illustrated in the figure, n = 1 represents the hybrid approach, in which each linked action covers a small part of the state space of the system. As n increases, the portion of state space associated with each action and its subgoal grows. When n reaches M, coverage extends over the full state space, so that the system contains a universal plan.

Barrett, Anthony↗

Scalable Programming Workflows for Validation of Quantum Computers

Hybrid quantum-classical workflows have become standard methods for executing variational algorithms and other quantum simulation techniques, which are key applications for noisy intermediate scale quantum (NISQ) computers. Validating these simulations is an important task which helps gauge the progress of quantum computer development, and classical simulation can serve as a tool to this end. Both exact and more scalable approximate methods with quantifiable error bounds can be used in validation tasks where the applicable metrics include the distance from a calculable ground truth, the quality of an error model fit to data, etc. Here we present a library extension that includes methods for validation of quantum simulations based on scalable hybrid workflows executable on high performance computers. We provide examples that use approximate methods based on tensor networks and stabilizer simulators to bound the error of quantum simulations on NISQ hardware.

Nguyen, Thien↗

Modeling in the State Flow Environment to Support Launch Vehicle Verification Testing for Mission and Fault Management Algorithms in the NASA Space Launch System

Analysis methods and testing processes are essential activities in the engineering development and verification of the National Aeronautics and Space Administration's (NASA) new Space Launch System (SLS). Central to mission success is reliable verification of the Mission and Fault Management (M&FM) algorithms for the SLS launch vehicle (LV) flight software. This is particularly difficult because M&FM algorithms integrate and operate LV subsystems, which consist of diverse forms of hardware and software themselves, with equally diverse integration from the engineering disciplines of LV subsystems. M&FM operation of SLS requires a changing mix of LV automation. During pre-launch the LV is primarily operated by the Kennedy Space Center (KSC) Ground Systems Development and Operations (GSDO) organization with some LV automation of time-critical functions, and much more autonomous LV operations during ascent that have crucial interactions with the Orion crew capsule, its astronauts, and with mission controllers at the Johnson Space Center. M&FM algorithms must perform all nominal mission commanding via the flight computer to control LV states from pre-launch through disposal and also address failure conditions by initiating autonomous or commanded aborts (crew capsule escape from the failing LV), redundancy management of failing subsystems and components, and safing actions to reduce or prevent threats to ground systems and crew. To address the criticality of the verification testing of these algorithms, the NASA M&FM team has utilized the State Flow environment6 (SFE) with its existing Vehicle Management End-to-End Testbed (VMET) platform which also hosts vendor-supplied physics-based LV subsystem models. The human-derived M&FM algorithms are designed and vetted in Integrated Development Teams composed of design and development disciplines such as Systems Engineering, Flight Software (FSW), Safety and Mission Assurance (S&MA) and major subsystems and vehicle elements such as Main Propulsion Systems (MPS), boosters, avionics, Guidance, Navigation, and Control (GN&C), Thrust Vector Control (TVC), liquid engines, and the astronaut crew office. Since the algorithms are realized using model-based engineering (MBE) methods from a hybrid of the Unified Modeling Language (UML) and Systems Modeling Language (SysML), SFE methods are a natural fit to provide an in depth analysis of the interactive behavior of these algorithms with the SLS LV subsystem models. For this, the M&FM algorithms and the SLS LV subsystem models are modeled using constructs provided by Matlab which also enables modeling of the accompanying interfaces providing greater flexibility for integrated testing and analysis, which helps forecast expected behavior in forward VMET integrated testing activities. In VMET, the M&FM algorithms are prototyped and implemented using the same C++ programming language and similar state machine architectural concepts used by the FSW group. Due to the interactive complexity of the algorithms, VMET testing thus far has verified all the individual M&FM subsystem algorithms with select subsystem vendor models but is steadily progressing to assessing the interactive behavior of these algorithms with LV subsystems, as represented by subsystem models. The novel SFE applications has proven to be useful for quick look analysis into early integrated system behavior and assessment of the M&FM algorithms with the modeled LV subsystems. This early MBE analysis generates vital insight into the integrated system behaviors, algorithm sensitivities, design issues, and has aided in the debugging of the M&FM algorithms well before full testing can begin in more expensive, higher fidelity but more arduous environments such as VMET, FSW testing, and the Systems Integration Lab7 (SIL). SFE has exhibited both expected and unexpected behaviors in nominal and off nominal test cases prior to full VMET testing. In many findings, these behavioral characteristics were used to correct the M&FM algorithms, enable better test coverage, and develop more effective test cases for each of the LV subsystems. This has improved the fidelity of testing and planning for the next generation of M&FM algorithms as the SLS program evolves from non-crewed to crewed flight, impacting subsystem configurations and the M&FM algorithms that control them. SFE analysis has improved robustness and reliability of the M&FM algorithms by revealing implementation errors and documentation inconsistencies. It is also improving planning efficiency for future VMET testing of the M&FM algorithms hosted in the LV flight computers, further reducing risk for the SLS launch infrastructure, the SLS LV, and most importantly the crew.

Trevino, Luis↗

Cloud-Precipitation Hybrid Regimes and their Projection onto IMERG Precipitation Data

We extend and enhance the concept of the Cloud Regimes (CRs) developed from two-dimensional joint histograms of cloud optical thickness and cloud top pressure from the Moderate Resolution Imaging Spectroradiometer (MODIS), by adding precipitation information in order to better understand cloud-precipitation relationships. Taking advantage of the high-resolution Integrated Multi-satellitE Retrievals for GPM (IMERG) precipitation dataset, cloud-precipitation “hybrid” regimes are derived by implementing the k-means clustering algorithm with advanced initialization and objective measures to determine the most optimal clusters. By expressing precipitation rates within 1-degree grid cell as histograms and making choices on the relative weight of cloud and precipitation, we could obtain several editions of hybrid cloud-precipitation regimes (CPRs), and examine their characteristics. In the deep tropics, when precipitation is weighted weakly, the cloud part of the hybrid entroids resembles the centroid of cloud-only regimes, but still tightens the cloud-precipitation relationship by decreasing the precipitation variability of each regime. As precipitation weight progressively increases, the shape of the cloudy part of the hybrid centroids becomes blunter, while the precipitation part of the centroids sharpens. In the case where cloud and precipitation are weighted equally, the CPRs representing high clouds with intermediate to heavy precipitation exhibit distinct features in the precipitation parts of the centroids, which allows us to project them onto the 30-minly IMERG domain. Such a projection can be used to overcome the temporal sparseness of MODIS cloud observations, which leads to great application potential for various convection-focused studies, including diurnal cycle analysis.

cloud-precipitation↗

Optimizing Non-Terrestrial Hybrid RF/FSO Links With Reinforcement Learning: Navigating Through Clouds

In the pursuit of ubiquitous broadband connectivity, there has been a significant shift towards the vertical expansion of communication networks into space, particularly through the exploitation of low Earth orbit (LEO) satellite constellations, which are favored for their relatively low latency. However, this approach faces many challenges that need to be addressed, including atmospheric turbulence, high path loss, and dynamic cloud formations. High-altitude pseudo-satellites (HAPS) have emerged as promising relaying layers between LEO satellites and ground stations, enhancing coverage, latency, and direct terrestrial user connectivity. While radio frequency (RF) bands suffer from congestion and limited bandwidth, free space optical (FSO) communications offer higher data rates, but are susceptible to misalignment and weather-induced signal degradation. To address these challenges, a hybrid RF/FSO approach has been proposed to take advantage of both technologies by dynamic switching between RF and FSO based on propagation channel conditions. This paper introduces a reinforcement learning-based algorithm designed to optimize the trajectory of HAPS, maneuver around cloudy areas, and seamlessly switch between the RF and FSO communication modes to maximize the achievable capacity. The proposed approach aims to maximize system performance by intelligently adapting to environmental conditions and offering a promising solution for next-generation space communication networks.

actor-critic algorithm↗

Perspectives on the integration between first-principles and data-driven modeling

Efficiently embedding and/or integrating mechanistic information with data-driven models is essential if it is desired to simultaneously take advantage of both engineering principles and data-science. Further the opportunity for hybridization occurs in many scenarios, such as the development of a faster model of an accurate high-fidelity computer model; the correction of a mechanistic model that does not fully-capture the physical phenomena of the system; or the integration of a data-driven component approximating an unknown correlation within a mechanistic model. At the same time, different techniques have been proposed and applied in different literatures to achieve this hybridization, such as hybrid modeling, physics-informed Machine Learning (ML) and model calibration. In this paper we review the methods, challenges, applications and algorithms of these three research areas and discuss them in the context of the different hybridization scenarios. Moreover, we provide a comprehensive comparison of the hybridization techniques with respect to their differences and similarities, as well as advantages and limitations and future perspectives. Finally, we apply and illustrate hybrid modeling, physics-informed ML and model calibration via a chemical reactor case study.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Decoherence-protected quantum gates for a hybrid solid-state spin register

Protecting the dynamics of coupled quantum systems from decoherence by the environment is a key challenge for solid-state quantum information processing. An idle quantum bit (qubit) can be efficiently insulated from the outside world by dynamical decoupling, as has recently been demonstrated for individual solid-state qubits. However, protecting qubit coherence during a multi-qubit gate is a non-trivial problem in general, the decoupling disrupts the interqubit dynamics and hence conflicts with gate operation. This problem is particularly salient for hybrid systems, in which different types of qubit evolve and decohere at very different rates. Here we present the integration of dynamical decoupling into quantum gates for a standard hybrid system, the electron-nuclear spin register. Our design harnesses the internal resonance in the coupled-spin system to resolve the conflict between gate operation and decoupling. We experimentally demonstrate these gates using a two-qubit register in diamond operating at room temperature. Quantum tomography reveals that the qubits involved in the gate operation are protected as accurately as idle qubits. We also perform Grover's quantum search algorithm, and achieve fidelities of more than 90% even though the algorithm run-time exceeds the electron spin dephasing time by two orders of magnitude. Our results directly allow decoherence-protected interface gates between different types of solid-state qubit. Ultimately, quantum gates with integrated decoupling may reach the accuracy threshold for fault-tolerant quantum information processing with solid-state devices.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Minimum deltaV Burn Planning for the International Space Station Using a Hybrid Optimization Technique, Level 1

The International Space Station's (ISS) trajectory is coordinated and executed by the Trajectory Operations and Planning (TOPO) group at NASA's Johnson Space Center. TOPO group personnel routinely generate look-ahead trajectories for the ISS that incorporate translation burns needed to maintain its orbit over the next three to twelve months. The burns are modeled as in-plane, horizontal burns, and must meet operational trajectory constraints imposed by both NASA and the Russian Space Agency. In generating these trajectories, TOPO personnel must determine the number of burns to model, each burn's Time of Ignition (TIG), and magnitude (i.e. deltaV) that meet these constraints. The current process for targeting these burns is manually intensive, and does not take advantage of more modern techniques that can reduce the workload needed to find feasible burn solutions, i.e. solutions that simply meet the constraints, or provide optimal burn solutions that minimize the total DeltaV while simultaneously meeting the constraints. A two-level, hybrid optimization technique is proposed to find both feasible and globally optimal burn solutions for ISS trajectory planning. For optimal solutions, the technique breaks the optimization problem into two distinct sub-problems, one for choosing the optimal number of burns and each burn's optimal TIG, and the other for computing the minimum total deltaV burn solution that satisfies the trajectory constraints. Each of the two aforementioned levels uses a different optimization algorithm to solve one of the sub-problems, giving rise to a hybrid technique. Level 2, or the outer level, uses a genetic algorithm to select the number of burns and each burn's TIG. Level 1, or the inner level, uses the burn TIGs from Level 2 in a sequential quadratic programming (SQP) algorithm to compute a minimum total deltaV burn solution subject to the trajectory constraints. The total deltaV from Level 1 is then used as a fitness function by the genetic algorithm in Level 2 to select the number of burns and their TIGs for the next generation. In this manner, the two levels solve their respective sub-problems separately but collaboratively until a burn solution is found that globally minimizes the deltaV across the entire trajectory. Feasible solutions can also be found by simply using the SQP algorithm in Level 1 with a zero cost function. This paper discusses the formulation of the Level 1 sub-problem and the development of a prototype software tool to solve it. The Level 2 sub-problem will be discussed in a future work. Following the Level 1 formulation and solution, several look-ahead trajectory examples for the ISS are explored. In each case, the burn targeting results using the current process are compared against a feasible solution found using Level 1 in the proposed technique. Level 1 is then used to find a minimum deltaV solution given the fixed number of burns and burn TIGs. The optimal solution is compared with the previously found feasible solution to determine the deltaV (and therefore propellant) savings. The proposed technique seeks to both improve the current process for targeting ISS burns, and to add the capability to optimize ISS burns in a novel fashion. The optimal solutions found using this technique can potentially save hundreds of kilograms of propellant over the course of the ISS mission compared to feasible solutions alone. While the software tool being developed to implement this technique is specific to ISS, the concept is extensible to other long-duration, central-body orbiting missions that must perform orbit maintenance burns to meet operational trajectory constraints.

Brown, Aaron J.↗

Conic Sector Analysis of Hybrid Control Systems

A hybrid control system contains an analog plant and a hybrid (or sampled-data) compensator. In this thesis a new conic sector is determined which is constructive and can be used to: (1) determine closed loop stability, (2) analyze robustness with respect to modelling uncertainties, (3) analyze steady state response to commands, and (4) select the sample rate. The use of conic sectors allows the designer to treat hybrid control systems as though they were analog control systems. The center of the conic sector can be used as a rigorous linear time invariant approximation of the hybrid control system, and the radius places a bound on the errors of this approximation. The hybrid feedback system can be multivariable, and the sampler is assumed to be synchronous. Algorithms to compute the conic sector are presented. Several examples demonstrate how the conic sector analysis techniques are applied. Extensions to single loop multirate hybrid feedback systems are presented. Further extensions are proposed for multiloop multirate hybrid feedback system and for single rate systems with asynchronous sampling.

Thompson, P. M.↗

Hybrid Model Based Approaches for Systems Health Management and Prognostics

This is a previously approved and published presentation. To accurately predict the future state of any system, it is required to possess knowledge of its current health state and future operational conditions. Present achievements of data-driven algorithms in regression of complex nonlinear functions and classification tasks have generated a growing interest in artificial intelligence for industrial applications. Complex multi-physics models as well as digital twins, once purely built on physics and corresponding simplified lumped parameter iterations, can now benefit from machine learning algorithms to mitigate the lack of understanding of some complex behavior. Given models of the current and future system behavior, a general approach of model-based prognostics can solve the prediction problem and further decision-making. In principle, data-driven approaches can replace expensive experimental test-setups as well as reduce the number of simulations needed to explore, e.g., the parametric space of a multi-parameter model. Nonetheless, the limitations of pure data-driven methods came to light rather quickly, at least for some industries. In many industrial applications, data acquisition is costly, and the volume of data that can be collected does not satisfy the requirements for effective model training and cross-validation. Therefore, some recent works in the area of machine learning is focusing on blending physics with data-driven algorithms, thus mitigating the drawbacks of the two approaches and emphasizing respective advantages. Partial physical knowledge of the problem can aid the learning process by “guiding” the algorithm towards efficient solutions that satisfy the physics driving the system behavior. The result is a hybrid modeling approach combining physical knowledge as well data-driven methods to develop a unified hybrid approach. A hybrid framework for fusing information from physics-based performance models along with deep learning algorithms for prognostics of complex safety-critical systems is presented. In this framework, physics-based performance models infer unobservable model parameters related to the system's components health solving a calibration problem in the deep learning approach.

Prognostics↗

Hybrid-PIC modeling of laser-plasma interactions and hot electron generation in gold hohlraum walls

The walls of the hohlraum used in experiments at the national ignition facility are heated by laser beams with intensities ∼1015 W/cm2, a wavelength of ∼1/3 μm, and pulse lengths on the order of a ns, with collisional absorption believed to be the primary heating mechanism. X-rays generated by the hot ablated plasma at the gold walls are then used to implode a target in the hohlraum interior. In addition to the collisional absorption of laser energy at the walls, non-linear laser-plasma interactions (LPI), such as stimulated Raman scattering and two plasmon decay, are believed to generate a population of supra-thermal electrons which, if present in the hohlraum, can have a deleterious effect on target implosion. We describe results of hohlraum modeling using a hybrid particle-in-cell code. To enable this work, new particle-based algorithms for a multiple-ion magneto-hydrodynamic (MHD) treatment, and a particle-based ray-tracing model were developed. The use of such hybrid methods relaxes the requirement to resolve the laser wavelength, and allows for relatively large-scale hohlraum simulations with a reasonable number of cells. But the non-linear effects which are believed to be the cause of hot electron generation can only be captured by fully kinetic simulations with good resolution of the laser wavelength. For this reason, we employ a two-tiered approach to hohlraum modeling. Large-scale simulations of the collisional absorption process can be conducted using the fast quasi-neutral MHD algorithm with fluid particle species. From these simulations, we can observe the time evolution of the hohlraum walls and characterize the density and temperature profiles. From these results, we can transition to smaller-scale highly resolved simulations using traditional kinetic particle-in-cell methods, from which we can fully model all of the non-linear laser-plasma interactions, as well as assess the details of the electron distribution function. We find that vacuum hohlraums should be stable to both two plasmon decay and stimulated Raman scattering instabilities for intensities ≤1015 W/cm2. In gas-filled hohlraums, shocks may be induced in the blowoff gold plasma, which leads to more complex density and temperatures profiles. The resulting effect on LPI stability depends strongly on the details of the profile, and it is possible for the gas-filled hohlraum to become unstable to two plasmon decay at 1015 W/cm2 if the quarter-critical surface reaches temperatures exceeding 1 keV.

Thoma, C.↗