Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “parameter calibration”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Uncertainty Quantification for a Multi-phase Carbon Equation of State Model

Many physics models have tunable parameters that are calibrated by matching the model output to experimental or calculated data. However, given that calibration data often contain uncertainty and that different model parameter sets might result in a very similar simulated output for a finite calibration data set, it is advantageous to provide an ensemble of parameter sets that are consistent with the calibration data. Uncertainty quantification (UQ) provides a means to generate such an ensemble in a statistically rigorous fashion. In this work, we perform UQ for a multi-phase equation of state (EOS) model for carbon containing the diamond, graphite, and liquid phases. We use a Bayesian framework for the UQ and introduce a novel strategy for including phase diagram information in the calibration. The method is highly general and accurately reproduces the calibration data without any material-specific prior knowledge of the EOS model parameters.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Estimation Matrix Calibration of PMU Data-driven State Estimation Using Neural Network

Linear state estimation (LSE) is a phasor measurement unit (PMU) data-based power system state estimation that incorporates a linear measurement model in rectangular coordinates. Due to the high computational efficiency and high observational time-resolution, LSE can act as a supplementary state estimation in a wide-area monitoring system (WAMS). The performance of LSE is relatively sensitive to noises in measurements. Therefore, the estimation accuracy relies heavily on the accuracy of the estimation matrix, which is directly influenced by the measurement weight matrix. This paper proposes two novel calibration method of the estimation matrix using neural networks. One is based on the minimum absolute network loss (ANL), and the other is based on the minimum average squared network loss (ASNL). Both methods are tested and compared with LSE algorithms on the IEEE 14-bus system

neural network↗
DOE OSTI2021

Real-World Driving Features for Identifying Intelligent Driver Model Parameters

Driver behavior models play a significant role in representing different driving styles and the associated relationships with traffic patterns and vehicle energy consumption in simulation studies. The models often serve as a proxy for baseline human driving when assessing energy-saving strategies that alter vehicle velocity. Such models are especially important in connectivity-enabled energy-saving strategy research because they can easily adapt to changing driving conditions like posted speed limits or change in traffic light state. While numerous driver models exist, parametric driver models provide the flexibility required to represent variability in real-world driving through different combinations of model parameters. These model parameters must be informed by a representative set of parameter values for the driver model to adequately represent a real-world driver. It stands to reason that determining the parameter values from real-world driving data would serve the purpose of representing a real-world driver. Although the published literature is replete with techniques and consequences of using real-world driving data to determine parameter values for parametric driver models, none have explored them in the context of using shorter driving features where the parameter values may change over the course of a single trip for the same driver. In this study we consider the “intelligent driver model” (IDM) as our driver behavior model and use real-world driving data from the Transportation Secure Data Center (TSDC) maintained at the National Renewable Energy Laboratory (NREL). The TSDC includes real-world travel data from across the United States, from which NREL has created a wide range of driving routes consisting of road features such as speed limits, stop locations, and turn locations. The real-world driving data are categorized into different driving regimes and extracted into driving features. The driving features are then used to calibrate the parameter values for the IDM. The distribution of the parameters and the relationships among them are reported. The insights obtained from this study enable judicious usage of IDM or similar parametric driver models to represent baseline human driver behavior in simulations.

27 ARPA - Advanced Research Projects Agency-Energy↗
DOE OSTI2021

Real-World Driving Features for Identifying Intelligent Driver Model Parameters: Preprint

Driver behavior models play a significant role in representing different driving styles and the associated relationships with traffic patterns and vehicle energy consumption in simulation studies. The models often serve as a proxy for baseline human driving when assessing energy-saving strategies that alter vehicle velocity. Such models are especially important in connectivity-enabled energy-saving strategy research because they can easily adapt to changing driving conditions like posted speed limits or change in traffic light state. While numerous driver models exist, parametric driver models provide the flexibility required to represent variability in real-world driving through different combinations of model parameters. These model parameters must be informed by a representative set of parameter values for the driver model to adequately represent a real-world driver. It stands to reason that determining the parameter values from real-world driving data would serve the purpose of representing a real-world driver. While the published literature is replete with techniques and consequences of using real-world driving data to determine parameter values for parametric driver models, none have explored them in the context of using shorter driving features where the parameter values may change over the course of a single trip for the same driver. In this study we consider the “Intelligent Driver Model” (IDM) as the driver behavior model to explore and real-world driving data from the Transportation Secure Data Center (TSDC) maintained at the National Renewable Energy Laboratory (NREL). The TSDC includes real-world travel data from across the United States, from which NREL has created a wide range of driving routes consisting of road features such as speed limits, stop locations and turn locations. The real-world driving data are categorized into different driving regimes and extracted into driving features. The driving features are then used to calibrate the parameter values for the IDM. The distribution of the parameters and the relationships among them are reported. The insights obtained from this study enable judicious usage of IDM or similar parametric driver models to represent baseline human driver behavior in simulations.

27 ARPA - Advanced Research Projects Agency-Energy↗
DOE OSTI2025

High‐Resolution National‐Scale Water Modeling Is Enhanced by Multiscale Differentiable Physics‐Informed Machine Learning

Abstract The National Water Model (NWM) is a key tool for flood forecasting, planning, and water management. Key challenges facing the NWM include calibration and parameter regionalization when confronted with big data. We present two novel versions of high‐resolution (∼37 km 2 ) differentiable models (a type of hybrid model): one with implicit, unit‐hydrograph‐style routing and another with explicit Muskingum‐Cunge routing in the river network. The former predicts streamflow at basin outlets whereas the latter presents a discretized product that seamlessly covers rivers in the conterminous United States (CONUS). Both versions use neural networks to provide a multiscale parameterization and process‐based equations to provide a structural backbone, which were trained simultaneously (“end‐to‐end”) on 2,807 basins across the CONUS and evaluated on 4,997 basins. Both versions show great potential to elevate future NWM performance for extensively calibrated as well as ungauged sites: the median daily Nash‐Sutcliffe efficiency of all 4,997 basins is improved to around 0.68 from 0.48 of NWM3.0. As they resolve spatial heterogeneity, both versions greatly improved simulations in the western CONUS and also in the Prairie Pothole Region, a long‐standing modeling challenge. The Muskingum‐Cunge version further improved performance for basins >10,000 km 2 . Overall, our results show how neural‐network‐based parameterizations can improve NWM performance for providing operational flood predictions while maintaining interpretability and multivariate outputs. The modeling system supports the Basic Model Interface (BMI), which allows seamless integration with the next‐generation NWM. We also provide a CONUS‐scale hydrologic data set for further evaluation and use.

Song, Yalan [Civil and Environmental Engineering T↗
DOE OSTI2020

Active- and transfer-learning applied to microscale-macroscale coupling to simulate viscoelastic flows

Active- and transfer-learning are applied to microscale dynamics of polymer flows for the multiscale discovery of effective constitutive approximations required in viscoelastic flow simulation. The result is macroscopic rheology directly connected to a microstructural model. Micro and macroscale simulations are adaptively coupled by means of Gaussian process regression (GPR) to run the expensive microscale computations only as necessary. This multiscale method is demonstrated with flows of a polymer solution as a model system. At the microscale level dissipative particle dynamics (DPD) is employed to model the fluid as a suspension of bead-spring micro-structures subjected to steady shear flow. The results yield the non-Newtonian viscosity and the first normal stress difference at strain rates as training data used in a GPR model. DPD parameters are calibrated with respect to experimental data for a real polymer solution. Compliance with these data requires adjustment of the DPD model's cutoff radius, which then becomes a function of the second invariant of the strain rate tensor. The FENE-P model is chosen for the macroscale description using the spectral element method (SEM) to simulate channel flow and flow past a circular cylinder. The DPD results at the lowest possible shear strain rate yield an estimate of the zero-shear rate viscosity, which allows the initiation of the macroscale flow by SEM as a Newtonian fluid. The resulting strain-rate field is surveyed to determine additional shear strain rate sampling points for the DPD system. This new information allows an initial fitting of parameters of the constitutive equation followed by new SEM simulations at the macroscale. Additionally, guided by active-learning GPR to select new sampling points, this process continues until convergence is achieved. The effectiveness of this new simulation paradigm for viscoelastic flows is tested with different macroscale operating conditions. The effective closure learned in the channel simulation is then transferred directly to the flow past a circular cylinder at low Reynolds number, where the results show that only two additional DPD simulations are required to achieve a satisfactory constitutive model. With an increase of the Reynolds number, the active-learning scheme automatically detects the inaccuracy of the learned constitutive model, and initiates additional DPD simulations for the extra data needed to once again close the microscale-macroscale coupled system. This new paradigm of active- and transfer-learning for multiscale modeling is readily applicable to other microscale-macroscale coupled simulations of complex fluids and other materials. Furthermore, the coupling between microscale and macroscale solvers can be seamlessly implemented with our open source multiscale universal interface (MUI) library.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

SWAT model application for evaluating agricultural conservation practice effectiveness in reducing phosphorous loss from the Western Lake Erie Basin

Lake Erie is threatened by eutrophication and harmful algal blooms due to excess nutrient loading from agricultural sources. Agricultural conservation practices (ACPs) have been developed and implemented to reduce nutrient losses but estimating ACP effectiveness is challenging. The Soil and Water Assessment Tool (SWAT) has been used to investigate ACP effectiveness for water quality improvement. Many SWAT applications have been developed by different investigators to evaluate ACP effectiveness for reducing nutrient, particularly phosphorus (P), loading in the agriculturally-dominated Western Lake Erie Basin (WLEB). Our objective is to establish what has been achieved by past modeling research and make suggestions for future applications and improvements. We synthesized the findings of 28 SWAT modeling studies within the WLEB. Models generally performed satisfactorily against accepted criteria for streamflow and sediment, but performance for P loads, like soluble reactive P, was mostly “unsatisfactory”. The “unsatisfactory” performance maybe due to imperfections and idealizations in model formulations and/or parameterization. Thus, simulations of P transport and transformation processes need improvement. In addition, model parameter selection is the key part of model set-up. Most SWAT modeling studies used default values during initial set-up, then performed calibration and validation. It was found that the calibrated P related parameter values varied widely across different studies, even within the same watershed with some values unrealistic for the study areas. The phenomena of different combinations of model parameters producing similar outputs indicates equifinality. Equifinality in the baseline model may impact results when ACPs are incorporated. Furthermore, the unrealistic values used in ACP assessment undermine the credibility of ACP effectiveness. Future model applications should try to re-examine the calibrated P parameters and make sure they are realistic for the study area as well as reduce equifinality by constraining the model with characterization of watershed conditions, better understanding of hydrologic processes, and parameter values based on real-world observations. As a result, future model applications should focus on improving P transport and transformation processes, using measured watershed characteristics for parameterization, and improving reflections of climate change, which could result in more accurate assessments of ACP effectiveness to meet targeted goals.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Manifold learning for coarse-graining atomistic simulations: Application to amorphous solids

In this work, we introduce a generalized machine learning framework to probabilistically parameterize upper-scale models in the form of nonlinear PDEs consistent with a continuum theory, based on coarse-grained atomistic simulation data of mechanical deformation and flow processes. The proposed framework utilizes a hypothesized coarse-graining methodology with manifold learning and surrogate-based optimization techniques. Coarse-grained high-dimensional data describing quantities of interest of the multiscale models are projected onto a nonlinear manifold whose geometric and topological structure is exploited for measuring behavioral discrepancies in the form of manifold distances. A surrogate model is constructed using Gaussian process regression to identify a mapping between stochastic parameters and distances. Derivative-free optimization is employed to adaptively identify a unique set of parameters of the upper-scale model capable of rapidly reproducing the system's behavior while maintaining consistency with coarse-grained atomic-level simulations. The proposed method is applied to learn the parameters of the shear transformation zone (STZ) theory of plasticity that describes plastic deformation in amorphous solids as well as coarse-graining parameters needed to translate between atomistic and continuum representations. We show that the methodology is able to successfully link coarse-grained microscale simulations to macroscale observables and achieve a high-level of parity between the models across scales.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Alkali-silica reaction expansion model for confined concrete with stress-dependency and casting direction anisotropy

Alkali-silica reaction (ASR) is a deleterious chemical reaction between alkali hydroxyl ions and types of silica found in some aggregates of concrete. Owners and regulators of nuclear power plants aim to ensure the safety of the concrete structures with optimal maintenance strategies. A new model was developed for predicting the expansion of concrete structures affected by alkali-silica reaction. The model includes a novel combination of existing models as an alkali-silica reaction advancement model, a casting direction anisotropic expansion model, a stress-dependent anisotropic expansion model, and a material property evolution model dependent on the degree of ASR expansion. The model parameters were calibrated based on existing literature data and data generated by previous efforts of this study. The calibrated model was then validated with the experiments carried out in previous efforts of this study. The model was shown to accurately predict the ASR-expansion of large-scale reinforced concrete specimens with confinement.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Krypton-81 dating of the deep Continental Intercalaire aquifer with implications for chlorine-36 dating

Deep groundwater samples from the Continental Intercalaire (CI) aquifer in the Northern Tunisian Sahara have been analyzed for noble gases 3 He, 4 He, Ne and 81 Kr, and for 14 C to better constrain the groundwater residence time of this large transboundary aquifer. Its significant radiogenic 4 He content and background-level 14 C both indicate water older than a few tens of thousands of years. Distinct helium concentrations and Ne/He ratios suggest different groundwater flow paths through the Tozeur and Kebili regions. 81 Kr is applied for the first time on the CI aquifer, providing more direct evidence for the presence of old groundwater with 81 Kr/Kr ratios at 63% – 15% of the atmospheric value, corresponding to residence times of 150 – 630 ka. These newly obtained 81 Kr ages enable a re-evaluation of the previously reported data from the Tozeur region in order to better constrain the input values of 36 Cl/Cl and Cl. The initial 36 Cl/Cl ratio is found to be 4 – 6 times larger than previously assumed, and the initial Cl concentration 6 – 15 times smaller. The results of 81 Kr dating and the recalibrated 36 Cl dating define an eastward age progression across > 500 km of flow path from the Algerian to Tunisian sections. Here, this study exemplifies the utility of 81 Kr not only as an independent and conservative chronometer, but also as a supplemental measurement to better calibrate the parameters required for other age tracers, including cosmogenic 36 Cl and radiogenic 4 He.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Third-body stabilization of supercritical CO 2 in CO oxidation: development and application of a ReaxFF force field for the CO/O/CO 2 system

Supercritical CO 2 (scCO 2 ) plays a crucial role as a solvent in separation processes, advanced power cycles, and materials processing. Nonetheless, the atomistic comprehension of how the dense scCO 2 matrix influences the fundamental reaction of carbon monoxide (CO) is still insufficiently explored. Experimental studies and molecular dynamics (MD) simulations frequently fail to detect the highly reactive, transient intermediates, such as atomic oxygen (O), that drive these reactions. Here, to address this issue, we have developed a novel ReaxFF reactive force field for the CO 2 /CO/O system. The force field parameters were calibrated using density functional theory and second-order Møller-Plesset calculations to model CO 2 crystal properties, intermolecular interactions, bond dissociation curves, and reaction energy barriers. The force field reproduces the cohesive energy of the CO 2 crystal, the pressure characteristics of bulk scCO 2 , the equation-of-state behavior over a wide pressure–density range, the pressure dependence of the C–O bond length under compression, and the structural properties of liquid and scCO 2 , as documented by experiments, ab-initio MD, and prominent non-reactive models. The force field was subsequently applied to study the CO + O → CO 2 reaction. In a dilute environment, the reaction is inefficient as the newly formed CO 2 rapidly dissociates due to excess kinetic and potential energy acquired from the exothermic reaction. Conversely, in a dense scCO 2 environment, the surrounding matrix acts as an efficient third body, stabilizing the emerging CO2 product via molecular collisions. Statistical analysis confirms an average excess energy dissipation of 133.9 ± 3.6 kcal/mol over 112.4 ± 17.9 ps. Kinetic energy decomposition reveals that ∼ 92% of the excess kinetic energy is stored in internal (rotational and vibrational) degrees of freedom. This ReaxFF force field establishes a mechanistic foundation for third-body stabilization in dense reactive environments.

Chowdhury, Emdadul Haque [Pennsylvania State Univ.↗
DOE OSTI2020

Nonlinear sparse Bayesian learning for physics-based models

This paper addresses the issue of overfitting while calibrating unknown parameters of over-parameterized physics-based models with noisy and incomplete observations. Here, a semi-analytical Bayesian framework of nonlinear sparse Bayesian learning (NSBL) is proposed to identify sparsity among model parameters during Bayesian inversion. NSBL offers significant advantages over machine learning algorithm of sparse Bayesian learning (SBL) for physics-based models, such as 1) the likelihood function or the posterior parameter distribution is not required to be Gaussian, and 2) prior parameter knowledge is incorporated into sparse learning (i.e. not all parameters are treated as questionable). NSBL employs the concept of automatic relevance determination (ARD) to facilitate sparsity among questionable parameters through parameterized prior distributions. The analytical tractability of NSBL is enabled by employing Gaussian ARD priors and by building a Gaussian mixture-model approximation of the posterior parameter distribution that excludes the contribution of ARD priors. Subsequently, type-II maximum likelihood is executed using Newton's method whereby the evidence and its gradient and Hessian information are computed in a semi-analytical fashion. We show numerically and analytically that SBL is a special case of NSBL for linear regression models. Subsequently, a linear regression example involving multimodality in both parameter posterior pdf and model evidence is considered to demonstrate the performance of NSBL in cases where SBL is inapplicable. Next, NSBL is applied to identify sparsity among the damping coefficients of a mass-spring-damper model of a shear building frame. These numerical studies demonstrate the robustness and efficiency of NSBL in alleviating overfitting during Bayesian inversion of nonlinear physics-based models.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Impact of lead on an axisymmetric, single bead blown powder DED overhung geometry

Metal additive manufacturing can be utilized for the near net-shape manufacture of components in a layer wise technique. Traditionally, the buildup process is conducted in the direction of gravity or bottom to top in the vertical orientation; however, with the availability of commercial systems with additive heads or fixturing that can index from vertical, and the need to manufacture larger parts without modifying available systems, the limitation of manufacturing from bottom to top is removed enabling the buildup of larger components printed at an angle from vertical. The effect of gravity on the melt pool and as-deposited component quality when printing off vertical is unknown in literature, especially for axisymmetric components. This understanding is critical for the advancement of manufacturing components of increased size in 4 and 5-axis additive systems. This investigation utilizes blown powder directed energy deposition to evaluate the change in as-printed geometry when the start point is altered in relation to gravity while the part rotates to manufacture an axisymmetric component. The objectives of this work are to determine the impact of the deposition location on the geometric variability on axisymmetric components. This investigation tests the hypothesis that differences in layer height exist due to a change in catchment efficiency when the deposition location is moved to a tangent of the round geometry due to a change in the melt pool dynamics. It was found the ideal deposition location when printing at 26.6-degrees from vertical was at −90-degrees from the top center point of the component, along the tangent, where gravity was pushing the melt pool down, but the rotation of the part was pulling the deposited material towards the top center of the component. This work provides an understanding of layer height stability and catchment efficiency to guide print orientation strategy for high-aspect ratio components. It was found the −90-degree lead deposition had the best layer height stability at 0.6 % as compared to the top-center at 6.3 % and +90-degree deposition location at 9.2 % for the nominal programmed layer height. The change in layer height also effected the final diameter the greatest for the top center deposition location where the diameter diverged by 2.5 % during the overbuild condition and converged by 0.6 %. In conclusion, this finding will increase the manufacturing efficiency of axisymmetric components by increasing the passive stability of the printing process for the successful manufacture of parts without the need for perfectly calibrated manufacturing parameters.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Stress localization investigation of additively manufactured GRCop-42 thin-wall structure

A full-field crystal plasticity (CP) framework is presented for the GRCop-42 alloy to study microscopic mechanical behavior and local stress heterogeneities. The microstructures of additively manufactured (AM) materials are often unique relative to conventionally processed materials, and the local thermal histories drive these differences during the build process. These thermal histories depend on the process parameters (laser power, scan speed, and scan strategy) and the part geometry. Prior research has shown that the mechanical properties of thin-walled structures can vary significantly with wall thickness due to changes in the thermal boundary conditions during manufacturing. It is, therefore, desirable to perform CP simulations based on the phenomenological constitutive model to predict the local mechanical responses induced by microstructural heterogeneities. This work generates representative microstructures based on experimentally collected grain information (i.e., texture) for grain scale stress analysis, and the material constitutive parameters are calibrated using the experimental mechanical testing data. Here, we specifically investigated the effect of crystallographic texture and grain morphologies on the size-dependent mechanical properties of AM GRCop-42. The selection of appropriate material properties for implementing an effective free surface boundary condition and the influence of adjacent buffer layers are also discussed. Analysis of local field results reveals a strong correlation between stress localization and the initial grain orientation. However, no significant relationship between the misorientation of the individual adjacent grains and the average misorientation is observed.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Improving a Multilevel Turbulence Closure Model for a Shallow Lake in Comparison With Other 1-D Models

Lakes differ from lands in water availability, heat capacity, albedo, and roughness, which affect local surface-atmospheric interactions. This study modified a multilevel upper ocean model (UOM) for lake applications and evaluated its performance in Lake Taihu (China) with comprehensive measurements against three popular one-dimensional (1-D) lake models. These models were based on different concepts, including the self-similarity (FLake), the wind-driven eddy diffusion (LISSS), the k-ε turbulence closure (SIMSTRAT), and a simplified turbulence closure (UOM). The surface flux scheme in these models was unified to exclude the discrepancies in representing air-lake exchanges. All models in their default formulations presented obvious cold water temperature biases and largely underestimated the lake surface temperature (LST) diurnal range. For each model, these deficiencies were significantly reduced by incorporating new physics schemes or calibrated tunable parameters based on systematic sensitivity tests. The primary modifications for UOM included (1) a new scheme of decreased surface roughness lengths to better characterize the shallow lake, (2) a solar radiation penetration scheme with increased light extinction coefficient and surface absorption fraction to account for the high water turbidity, and (3) turbulent Prandtl number increased by a factor of 20 to reduce the turbulent vertical mixing. All other models were improved in these three aspects (roughness, extinction, and mixing) within their original formulations. Given these improvements, UOM showed superior performance to other models in capturing LST diurnal cycle and daily to seasonal variations, as well as summer-autumn vertical stratification changes. The new UOM is well suited for application in shallow lakes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Modeling root system growth around obstacles

State-of-the-Art models of Root System Architecture (RSA) do not allow simulating root growth around rigid obstacles. Yet, the presence of obstacles can be highly disruptive to the root system. We grew wheat seedlings in sealed petri dishes without obstacle and in custom 3D-printed rhizoboxes containing obstacles. Time-lapse photography was used to reconstruct the wheat root morphology network. We used the reconstructed wheat root network without obstacle to calibrate an RSA model implemented in the R-SWMS software. The root network with obstacles allowed calibrating the parameters of a new function that models the influence of rigid obstacles on wheat root growth. Experimental results show that the presence of a rigid obstacle does not affect the growth rate of the wheat root axes, but that it does influence the root trajectory after the main axis has passed the obstacle. The growth recovery time, i.e. the time for the main root axis to recover its geotropism-driven growth, is proportional to the time during which the main axis grows along the obstacle. Qualitative and quantitative comparisons between experimental and numerical results show that the proposed model successfully simulates wheat RSA growth around obstacles. Our results suggest that wheat roots follow patterns that could inspire the design of adaptive engineering flow networks.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Single Crystal Plasticity Model with Deformation Twinning for the High Rate Deformation of β-HMX

Deformation twinning is an active deformation mechanism during loading along certain crystallographic directions of β-cyclotetramethylene tetranitramine (β-HMX). In this work, a finite strain thermomechanical model is extended to include twinning as a deformation mechanism in addition to plastic slip. The stress is derived from the free energy expression including a term representing the equation of state. The crystal plasticity framework is used to divide the total strain into inelastic and elastic component, where the elastic part is used in the expression for the free energy. The twin systems are treated as pseudo slip systems and the shear rate on the twin systems is evaluated in terms of the twin resistance and the projection of the stress tensor. The model parameters were calibrated against a set of plate impact experiments performed on β-HMX by Dick et al. The remaining plate impact experiments are used to evaluate the predictive capability of the model. The quality of the model fits and predictions is discussed from a physical and modeling perspective. Particularly, the role of twin modeling on results is highlighted. Finally, the model is used to explore the relationship between the propensity for twinning and crystal orientation.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗