Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “finite state machine”

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.

266 records · Page 15

Enhanced accuracy through ensembling of randomly initialized auto-regressive models for dynamical systems

Computational mechanics simulations using traditional finite element methods (FEM) require prohibitively expensive computational resources for real-time engineering applications, design optimization, and digital twin implementations. While machine learning (ML) surrogate models offer significant computational speedups, autoregressive ML models for time-dependent mechanical systems suffer from error accumulation that compromises long-term prediction reliability - a critical concern for engineering applications where accuracy over extended time horizons is essential for safety and performance assessments. Here, we propose a deep ensemble framework specifically designed to address this challenge in computational mechanics applications, where multiple ML surrogate models with random weight initializations are trained in parallel and their predictions aggregated during inference. This approach leverages statistical diversity to maximize information gain from a fixed set of training data and to mitigate error propagation, while maintaining the computational efficiency that makes ML surrogates attractive for engineering practice. We validate the framework on three representative problems spanning critical areas of computational mechanics: stress field evolution in heterogeneous microstructures under complex loading (relevant to advanced materials design and composite analysis), planetary-scale shallow water dynamics (applicable to environmental and geotechnical engineering), and Gray-Scott reaction-diffusion systems (relevant to mass transport and chemical process engineering). Across all test cases, the ensemble approach demonstrates consistent error reduction of 15-33% compared to individual models. The codes for this work are available on GitHub (https://github.com/Graham-Brady-Research-Group/AutoregressiveEnsemble_SpatioTemporal_Evolution).

autoregressive prediction↗

Feedback Control of Plasma Instabilities with Charged Particle Beams and Study of Plasma Turbulence

A new non-perturbing technique for feedback control of plasma instabilities has been developed in the Columbia Linear Machine (CLM). The feedback control scheme involves the injection of a feedback modulated ion beam as a remote suppressor. The ion beam was obtained from a compact ion beam source which was developed for this purpose. A Langmuir probe was used as the feedback sensor. The feedback controller consisted of a phase-shifter and amplifiers. This technique was demonstrated by stabilizing various plasma instabilities to the background noise level, like the trapped particle instability, the ExB instability and the ion-temperature-gradient (ITG) driven instability. An important feature of this scheme is that the injected ion beam is non-perturbing to the plasma equilibrium parameters. The robustness of this feedback stabilization scheme was also investigated. The principal result is that the scheme is fairly robust, tolerating about 100% variation about the nominal parameter values. Next, this scheme is extended to the unsolved general problem of controlling multimode plasma instabilities simultaneously with a single sensor-suppressor pair. A single sensor-suppressor pair of feedback probes is desirable to reduce the perturbation caused by the probes. Two plasma instabilities the ExB and the ITG modes, were simultaneously stabilized. A simple 'state' feedback type method was used where more state information was generated from the single sensor Langmuir probe by appropriate signal processing, in this case, by differentiation. This proof-of-principle experiment demonstrated for the first time that by designing a more sophisticated electronic feedback controller, many plasma instabilities may be simultaneously controlled. Simple theoretical models showed generally good agreement with the feedback experimental results. On a parallel research front, a better understanding of the saturated state of a plasma instability was sought partly with the help of feedback. A plasma instability is usually observed in its saturated state and appears as a single feature in the frequency spectrum with a single azimuthal and parallel wavenumbers. The physics of the non-zero spectral width was investigated in detail because the finite spectral width can cause "turbulent" transport. One aspect of the "turbulence" was investigated by obtaining the scaling of the linear growth rate of the instabilities with the fluctuation levels. The linear growth rates were measured with the established gated feedback technique. The research showed that the ExB instability evolves into a quasi-coherent state when the fluctuation level is high. The coherent aspects were studied with a bispectral analysis. Moreover, the single spectral feature was discovered to be actually composed of a few radial harmonics. The radial harmonics play a role in the nonlinear saturation of the instability via three-wave coupling.

Tham, Philip Kin-Wah↗

Surrogate models for linear response

Linear response theory is a well-established method in physics and chemistry for exploring excitations of many-body systems. In particular, the quasiparticle random-phase approximation (QRPA) provides a powerful microscopic framework by building excitations on top of the mean-field vacuum; however, its high computational cost limits model calibration and uncertainty quantification studies. Here, we present two complementary QRPA surrogate models and apply them to study response functions of finite nuclei. One is a reduced-order model that exploits the underlying QRPA structure, while the other utilizes the recently developed parametric matrix model algorithm to construct a map between the system’s Hamiltonian and observables. Our benchmark applications, the calculation of the electric dipole polarizability of 180 Yb and the 𝛽-decay half-life of 80 Ni, show that both emulators can achieve 0.1%–1% accuracy while offering a 6–7 orders of magnitude speedup compared to state-of-the-art QRPA solvers. These results demonstrate that the developed QRPA emulators are well positioned to enable Bayesian calibration and large-scale studies of computationally expensive physics models describing the properties of many-body systems.

Beta decay↗

Real-Time Bayesian Inference at Extreme Scale: A Digital Twin for Tsunami Early Warning Applied to the Cascadia Subduction Zone

We present a Bayesian inversion-based digital twin that employs acoustic pressure data from seafloor sensors, along with 3D coupled acoustic–gravity wave equations, to infer earthquake-induced spatiotemporal seafloor motion in real time and forecast tsunami propagation toward coastlines for early warning with quantified uncertainties. Our target is the Cascadia subduction zone, with one billion parameters. Computing the posterior mean alone would require 50 years on a 512 GPU machine. Instead, exploiting the shift invariance of the parameter-to-observable map and devising novel parallel algorithms, we induce a fast offline–online decomposition. The offline component requires just one adjoint wave propagation per sensor; using MFEM, we scale this part of the computation to the full El Capitan system (43,520 GPUs) with 92% weak parallel efficiency. Moreover, given real-time data, the online component exactly solves the Bayesian inverse and forecasting problems in 0.2 seconds on a modest GPU system, a ten-billion-fold speedup.

97 MATHEMATICS AND COMPUTING↗

NSTXU Diagnostic Disruption Dynamic Loading Represented by Response Spectra

This article presents the results of transient dynamic simulations of loads due to disruption eddy currents on the NSTXU vacuum vessel. Dynamic loading at diagnostic mounting locations is expressed as response spectra derived from the time history results of the dynamic structural simulations of a variety of disruption scenarios. The disruption simulations draw on a history of the project assessments of worst case disruptions for specific components. Major efforts to assess disruption loading have included the vacuum vessel which is the major structural support for the machine, as well as the passive plates (PPs), high harmonic fast wave (HHFW) antenna, and centerstack casing. Each one of these efforts included transient electromagnetic simulations producing time-dependent eddy current Lorentz loads (and in some cases halo loads) which then were applied to time-dependent structural dynamic analyses intended to obtain the proper dynamic amplification factors. In some instances, the EM model and structural model were identical allowing direct transfer of EM forces to the structural model. In other cases, the EM and structural model were not identical and the vector potential (VP) transfer method was used. The results files from these analyses were available (or re-run) to post process in ANSYS Classic time history postprocessor. In conclusion, the ANSYS command is used to create response spectra from time history data at desired points on the vessel.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A resolution independent neural operator

The Deep operator network (DeepONet) is a powerful yet simple neural operator architecture that utilizes two deep neural networks to learn mappings between infinite-dimensional function spaces. This architecture is highly flexible, allowing the evaluation of the solution field at any location within the desired domain. However, it imposes a strict constraint on the input space, requiring all input functions to be discretized at the same locations; this limits its practical applications. Here, in this work, we introduce a general framework for operator learning from input–output data with arbitrary number and locations of sensors. This begins by introducing a resolution-independent DeepONet (RI-DeepONet), enabling it to handle input functions that are arbitrarily, but sufficiently finely, discretized. To this end, we propose two dictionary learning algorithms to adaptively learn a set of appropriate continuous basis functions, parameterized as implicit neural representations (INRs), from correlated signals defined on arbitrary point cloud data. These basis functions are then used to project arbitrary input function data as a point cloud onto an embedding space (i.e., a vector space of finite dimensions) with dimensionality equal to the dictionary size, which can be directly used by DeepONet without any architectural changes. In particular, we utilize sinusoidal representation networks (SIRENs) as trainable INR basis functions. The introduced dictionary learning algorithms are then used in a similar way to learn an appropriate dictionary of basis functions for the output function data, which defines a new neural operator architecture referred to as the R esolution I ndependent N eural O perator (RINO). In the RINO, the operator learning task simplifies to learning a mapping from the coefficients of input basis functions to the coefficients of output basis functions. We demonstrate the robustness and applicability of RINO in handling arbitrarily (but sufficiently richly) sampled input and output functions during both training and inference through several numerical examples.

Deep operator network (DeepONet)↗

Process Design by FEM Simulation for Shape Ring Rolling of Large-Sized Ring

Ring rolling process is usually used to fabricate large-sized ring, such as, tower flange for wind power electric generator. Many kinds of seamless ring are used in wind power electric generator and manufactured by ring rolling process. In general, final part is machined after forming with shape of plain square section. Since interests for near net shaping of seamless ring have been increased gradually because of green energy, it is necessary to develop the technology for shape ring rolling with respect to the market demands and cost. Therefore, we studied the process and die design for shape ring rolling of large sized ring over 3,500 mm out diameter by experiment and FEM simulation. Ring rolling process is very difficult to solve by FEM method because of equilibrium state and size effect, etc. Moreover, shape ring rolling is more difficult to solve the problem that two plastic deformation zones are different each other, that is main roll and conical roll. Also since conical roll has a shape, deformation velocity field is very much complex and the deformed section passed axial roll is different section and velocity field. The FE simulations are performed to analyze process variables affected in forming of profiled ring. Therefore, the main features of used FE model are: (1) it adopts a transient or unsteady state full ring mesh to model the deformation processes and shape development; (2) the mandrel and conical rolls are modeled using coupled heat-transfer elements; (3) the model involves the full process from blank through perform to final profiled ring. From these calculated results, we have proposed the mechanisms of various tools, such as mandrel and conical roll. The calculated results are compared experimental results. Calculated results can predict the tilting of profiled ring and then process variables to form large sized ring.

36 MATERIALS SCIENCE↗

Numerical analysis of modified Central Solenoid insert design

The United States ITER Project Office (USIPO) is responsible for fabrication of the Central Solenoid (CS) for ITER project. The ITER machine is currently under construction by seven parties in Cadarache, France. The CS Insert (CSI) project should provide a verification of the conductor performance in relevant conditions of temperature, field, currents and mechanical strain. The US IPO designed the CSI that will be tested at the Central Solenoid Model Coil (CSMC) Test Facility at JAEA, Naka. To validate the modified design we performed three-dimensional numerical simulations using coupled solver for simultaneous structural, thermal and electromagnetic analysis. Thermal and electromagnetic simulations supported structural calculations providing necessary loads and strains. According to current analysis design of the modified coil satisfies ITER magnet structural design criteria for the following conditions: (1) room temperature, no current, (2) temperature 4K, no current, (3) temperature 4K, current 60 kA direct charge, and (4) temperature 4K, current 60 kA reverse charge. Fatigue life assessment analysis is performed for the alternating conditions of: temperature 4K, no current, and temperature 4K, current 45 kA direct charge. Results of fatigue analysis show that parts of the coil assembly can be qualified for up to 1 million cycles. Distributions of the Current Sharing Temperature (TCS) in the superconductor were obtained from numerical results using parameterization of the critical surface in the form similar to that proposed for ITER. Lastly, special ADPL scripts were developed for ANSYS allowing one-dimensional representation of TCS along the cable, as well as three-dimensional fields of TCS in superconductor material. Published by Elsevier B.V.

42 ENGINEERING↗

Application of European Large Component Technology in the United States - 12561

The European nuclear community currently has safer, more environmentally-friendly disposal options for large component disposal than the United States. The technology and innovation that makes these options viable will soon be available stateside. The removal and replacement of steam generators is a large and complicated task making volume reduction a vital service for the nuclear power industry. As plants age, the replacement of certain components is required in order to maintain plant efficiency and public safety. Currently in the United States, there are 3 options for the disposal of these items including: on-site storage, direct-disposal at a licensed (low-level radioactive waste) LLWR site, and volume reduction prior to disposal. If today's nuclear plants are to be used by future generations, waste stored on-site will eventually have to be disposed of to make room for newer, more advanced technology. Also, the space at LLWR disposal sites is finite, so all waste sent for disposal should be minimized whenever possible. Reducing the volume of decommissioned steam generators prior to final disposal shows the public that the nuclear industry is doing everything that it can to minimize its waste and conserve the environment. Additionally, many steam generators are massive and are not easily transported to a disposal site. When there is no practical transportation route, the size and weight of the component must be reduced. For various reasons including public safety, cost, and technical expertise, size reduction cannot be practically performed at the site of generation. However, since most facilities have some kind of access to a waterway, the components can be safely transported to another facility with barge access for size reduction. The size reduction process for steam generators has evolved over the past 5 years. Because of reductions to the amount of final waste, the dose to personnel, and the processing time, it is now an environmentally and economically friendly alternative to on-site storage and direct disposal. The major operations of this process include: - removal of the carbon steel steam generator dome; - removal of the channel head and majority of tube sheet to free the tubes from the tube sheet and gain access to all of the tubes for grit blasting; - grit blast decontamination of the steam generator tubes to allow disassembly; - removal of the U-Bend Section of the tubes to allow for tube pulling - tube pulling operations; - disassembly of the remainder of the steam generator shell and tube support assemblies; - size reduction of the steam generator resultant materials. All of the equipment for the American market is based on previous designs but has been adapted to meet the size requirements for processing of larger steam generators. This includes a large horizontal band saw, a tube pulling machine, tube blasting equipment, and a diamond wire saw frame. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A New Light Weight Structural Material for Nuclear Structures

Radiation shielding materials are commonly used in nuclear facilities to attenuate the background ionization radiations to a minimum level for creating a safer workplace, meeting regulatory requirements and maintaining high quality performance. The conventional radiation shielding materials have a number of drawbacks: heavy concrete contains a high amount of elements that are not desirable for an effective shielding such as oxygen, silicon, and calcium; a well known limitation of lead is its low machinability and toxicity, which is causing a major environmental concern. Therefore, an effective and environmentally friendly shielding material with increased attenuation and low mass density is desirable. Close-cell composite metal foams (CMFs) and open-cell Al foam with fillers are light-weight candidate materials that we have studied in this project. Close-cell CMFs possess several suitable properties that are unattainable by conventional radiation shielding materials such as low density and high strength for structural applications, high surface area to volume ratio for excellent thermal isolation with an extraordinary energy absorption capability. Open-cell foam is made up of a network of interconnected solid struts, which allows gas or fluid media to pass through it. This unique structure provided a further motive to investigate its application as radiation shields by infiltrating original empty pores with high hydrogen or boron compounds, which are well known for their excellent neutron shielding capability. The resulting open-cell foam with fillers will not only exhibit light weight and high specific surface area, but also possess excellent radiation shielding capability and good processability. In this study, all the foams were investigated for their radiation shielding efficiency in terms of X-ray, gamma ray and neutron. X-ray transmission measurements were carried out on a high-resolution microcomputed tomography (microCT) system. Gamma-emitting sources: 3.0mCi 60 Co, 1.8mCi 137 Cs, 13.5mCi 241 Am, and 5.0mCi 133 Ba were used for gamma-ray attenuation analysis. The evaluations of neutron transmission measurements were conducted at the Neutron Powder Diffractometer beam facility at North Carolina State University. The experimental results were verified theoretically through XCOM and Monte Carlo Z-particle Transport Code (MCNP). A mechanical investigation was performed by means of quasi-static compressive testing. Thermal characterizations were carried out through effective thermal conductivity and thermal expansion analyses in terms of high temperature guarded-comparative-longitudinal heat flow technique and thermomechanical analyzer (TMA), respectively. The experimental results were compared with analytical results obtained from, respectively, Brailsford and Major’s model and modified Turner’s model for verification. Flame test was performed in accordance with United States Nuclear Regulatory Commission (USNRC) standard. CMF sample and a 304L stainless steel control sample were subjected to a fully engulfing fire with an average flame temperature of 800°C for a period of 30 minutes. Finite Element Analysis was conducted to secure the credibility of the experimental results. This research indicates the potential of utilizing the light-weight close-cell CMFs and open-cell Al foam with fillers as shielding material replacing current heavy structures with additional advantage of high-energy absorption and excellent thermal characteristics.

36 MATERIALS SCIENCE↗

Implementation of AI/DEEP learning disruption predictor into a plasma control system

Abstract This paper reports on advances in the state‐of‐the‐art deep learning disruption prediction models based on the Fusion Recurrent Neural Network (FRNN) originally introduced in a 2019 NATURE publication [ https://doi.org/10.1038/s41586‐019‐1116‐4 ]. In particular, the predictor now features not only the “disruption score,” as an indicator of the probability of an imminent disruption, but also a “sensitivity score” in real time to indicate the underlying reasons for the imminent disruption. This adds valuable physics interpretability for the deep learning model and can provide helpful guidance for control actuators now implemented into a modern plasma control system (PCS). The advance is a significant step forward in moving from modern deep learning disruption prediction to real‐time control and brings novel AI‐enabled capabilities relevant for application to the future burning plasma ITER system. Our analyses use large amounts of data from JET and DIII‐D vetted in the earlier NATURE publication. In addition to “when” a shot is predicted to disrupt, this paper addresses reasons “why” by carrying out sensitivity studies. FRNN is accordingly extended to use more channels of information, including measured DIII‐D signals such as (i) the “n1rms” signal that is correlated with the n = 1 modes with finite frequency, including neoclassical tearing mode and sawtooth dynamics; (ii) the bolometer data indicative of plasma impurity control; and (iii) “q‐min”—the minimum value of the safety factor relevant to the key physics of kink modes. The additional channels and interpretability features expand the ability of the deep learning FRNN software to provide information about disruption subcategories as well as more precise and direct guidance for the actuators in a PCS.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

The Prospect of using Three-Dimensional Earth Models To Improve Nuclear Explosion Monitoring and Ground Motion Hazard Assessment

The last ten years have brought rapid growth in the development and use of three-dimensional (3D) seismic models of Earth structure at crustal, regional and global scales. In order to explore the potential for 3D seismic models to contribute to important societal applications, Lawrence Livermore National Laboratory (LLNL) hosted a 'Workshop on Multi-Resolution 3D Earth Models to Predict Key Observables in Seismic Monitoring and Related Fields' on June 6 and 7, 2007 in Berkeley, California. The workshop brought together academic, government and industry leaders in the research programs developing 3D seismic models and methods for the nuclear explosion monitoring and seismic ground motion hazard communities. The workshop was designed to assess the current state of work in 3D seismology and to discuss a path forward for determining if and how 3D Earth models and techniques can be used to achieve measurable increases in our capabilities for monitoring underground nuclear explosions and characterizing seismic ground motion hazards. This paper highlights some of the presentations, issues, and discussions at the workshop and proposes two specific paths by which to begin quantifying the potential contribution of progressively refined 3D seismic models in critical applied arenas. Seismic monitoring agencies are tasked with detection, location, and characterization of seismic activity in near real time. In the case of nuclear explosion monitoring or seismic hazard, decisions to further investigate a suspect event or to launch disaster relief efforts may rely heavily on real-time analysis and results. Because these are weighty decisions, monitoring agencies are regularly called upon to meticulously document and justify every aspect of their monitoring system. In order to meet this level of scrutiny and maintain operational robustness requirements, only mature technologies are considered for operational monitoring systems, and operational technology necessarily lags contemporary research. Current monitoring practice is to use relatively simple Earth models that generally afford analytical prediction of seismic observables (see Examples of Current Monitoring Practice below). Empirical relationships or corrections to predictions are often used to account for unmodeled phenomena, such as the generation of S-waves from explosions or the effect of 3-dimensional Earth structure on wave propagation. This approach produces fast and accurate predictions in areas where empirical observations are available. However, accuracy may diminish away from empirical data. Further, much of the physics is wrapped into an empirical relationship or correction, which limits the ability to fully understand the physical processes underlying the seismic observation. Every generation of seismology researchers works toward quantitative results, with leaders who are active at or near the forefront of what has been computationally possible. While recognizing that only a 3-dimensional model can capture the full physics of seismic wave generation and propagation in the Earth, computational seismology has, until recently, been limited to simplifying model parameterizations (e.g. 1D Earth models) that lead to efficient algorithms. What is different today is the fact that the largest and fastest machines are at last capable of evaluating the effects of generalized 3D Earth structure, at levels of detail that improve significantly over past efforts, with potentially wide application. Advances in numerical methods to compute travel times and complete seismograms for 3D models are enabling new ways to interpret available data. This includes algorithms such as the Fast Marching Method (Rawlison and Sambridge, 2004) for travel time calculations and full waveform methods such as the spectral element method (SEM; Komatitsch et al., 2002, Tromp et al., 2005), higher order Galerkin methods (Kaser and Dumbser, 2006; Dumbser and Kaser, 2006) and advances in more traditional Cartesian finite difference methods (e.g. Pitarka, 1999; Nilsson et al., 2007). The ability to compute seismic observables using a 3D model is only half of the challenge; models must be developed that accurately represent true Earth structure. Indeed, advances in seismic imaging have followed improvements in 3D computing capability (e.g. Tromp et al., 2005; Rawlinson and Urvoy, 2006). Advances in seismic imaging methods have been fueled in part by theoretical developments and the introduction of novel approaches for combining different seismological observables, both of which can increase the sensitivity of observations to Earth structure. Examples of such developments are finite-frequency sensitivity kernels for body-wave tomography (e.g. Marquering et al., 1998; Montelli et al., 2004) and joint inversion of receiver functions and surface wave group velocities (e.g. Julia et al., 2000).

58 GEOSCIENCES↗

Multiphysics Meshfree Degradation Modeling of Energy Storage Materials with Kernel Enrichment

Energy storage materials exhibit strong electro-chemo-mechanical coupling and highly anisotropic material properties, contributing to the formation and propagation of micro-cracking during charge/discharge cycling and ultimately diminishing performance and service life. With microstructural images supplied by the National Renewable Energy Laboratory (NREL), pixel-based meshfree model construction by the reproducing kernel particle method (RKPM) is used to represent the complex material microstructures that dictate the coupled physics of these systems. Traditional electro-chemo-mechanical models rely on mesh-based finite element methods, which can lead to difficulties in meshing such complex geometries and capturing crack propagation due to mesh dependency. The first kernel enrichment discussed will be the interface modified reproducing kernel (IM-RK) [1, 2], constructed by scaling a smooth kernel function with an interface-distance function to achieve strategic discontinuity types (i.e. weak discontinuities for strain discontinuities and strong discontinuities for cracks) and alleviate Gibbs oscillations near these transition zones. The IM-RK is especially useful for areas in which a known discontinuity-type is expected a priori. The second kernel enrichment to be discussed is a neural network-enhanced reproducing kernel (NN-RK) [3, 4], which is introduced to effectively model non-obvious damage and crack propagation in the material microstructures; the location, orientation, and solution transition near a localization are automatically captured by superimposed block-level NN optimizations. This NN enrichment approach allows for effective modeling of localizations via a fixed background discretization, relieving tedious efforts for adaptive refinement in traditional mesh-based methods. Applications to the heterogeneous microstructures of Li-ion battery cathodes will be presented to demonstrate the effectiveness of the proposed methods. NN-RK is additionally used to inform how crack opening and closure in turn affect the electro-chemo-mechanical responses in the material microstructure. Reference: [1] Wang, Y., Baek, J., Tang, Y. et al. "Support vector machine guided reproducing kernel particle method for image-based modeling of microstructures," Comput Mech 73, 907-942 (2024). https://doi.org/10.1007/s00466-023-02394-9. [2] Susuki, K., Allen, J. & Chen, J. S.. "Image-based modeling of coupled electro-chemo-mechanical behavior of Li-ion battery cathode using an interface-modified reproducing kernel particle method," Engineering with Computers (2024). https://doi.org/10.1007/s00366-024-02016-9. [3] Baek, J., Chen, J. S., Susuki, K., "Neural Network enhanced Reproducing Kernel Particle Method for Modeling Localizations," International Journal for Numerical Methods in Engineering, Vol. 123, 4422-4454 (2022). https://doi.org/10.1002/nme.7040.

25 ENERGY STORAGE↗

Kernel Enriched Meshfree Multiphysics Degradation Modeling of Energy Storage Materials

Energy storage materials exhibit strong electro-chemo-mechanical coupling and highly anisotropic material properties, contributing to the formation and propagation of micro-cracking during charge/discharge cycling and ultimately diminishing performance and service life. With microstructural images supplied by the National Laboratory of the Rockies (NLR), pixel-based meshfree model construction by the reproducing kernel particle method (RKPM) is used to represent the complex material microstructures that dictate the coupled physics of these systems. Traditional electro-chemo-mechanical models rely on mesh-based finite element methods, which can lead to difficulties in meshing such complex geometries and capturing crack propagation due to mesh dependency. The first kernel enrichment discussed will be the interface modified reproducing kernel (IM-RK) [1, 2], constructed by scaling a smooth kernel function with an interface-distance function to achieve strategic discontinuity types (i.e. weak discontinuities for strain discontinuities and strong discontinuities for cracks) and alleviate Gibbs oscillations near these transition zones. The IM-RK is especially useful for areas in which a known discontinuity-type is expected a priori. The second kernel enrichment to be discussed is a neural network-enhanced reproducing kernel (NN-RK) [3, 4], which is introduced to effectively model non-obvious damage and crack propagation in the material microstructures; the location, orientation, and solution transition near a localization are automatically captured by superimposed block-level NN optimizations. This NN enrichment approach allows for effective modeling of localizations via a fixed background discretization, relieving tedious efforts for adaptive refinement in traditional mesh-based methods. Applications to the heterogeneous microstructures of Li-ion battery cathodes will be presented to demonstrate the effectiveness of the proposed methods. NN-RK is additionally used to inform how crack opening and closure in turn affect the electro-chemo-mechanical responses in the material microstructure. References: [1] Wang, Y., Baek, J., Tang, Y. et al. "Support vector machine guided reproducing kernel particle method for image-based modeling of microstructures," Comput Mech 73, 907-942 (2024). https://doi.org/10.1007/s00466-023-02394-9. [2] Susuki, K., Allen, J. & Chen, J. S.. "Image-based modeling of coupled electro-chemo-mechanical behavior of Li-ion battery cathode using an interface-modified reproducing kernel particle method," Engineering with Computers (2024). https://doi.org/10.1007/s00366-024-02016-9. [3] Baek, J., Chen, J. S., Susuki, K., "Neural Network enhanced Reproducing Kernel Particle Method for Modeling Localizations," International Journal for Numerical Methods in Engineering, Vol. 123, 4422-4454 (2022). https://doi.org/10.1002/nme.7040.

97 MATHEMATICS AND COMPUTING↗