Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “model inference”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Novel symmetry-preserving neural network model for phylogenetic inference

Abstract Motivation Scientists world-wide are putting together massive efforts to understand how the biodiversity that we see on Earth evolved from single-cell organisms at the origin of life and this diversification process is represented through the Tree of Life. Low sampling rates and high heterogeneity in the rate of evolution across sites and lineages produce a phenomenon denoted “long branch attraction” (LBA) in which long nonsister lineages are estimated to be sisters regardless of their true evolutionary relationship. LBA has been a pervasive problem in phylogenetic inference affecting different types of methodologies from distance-based to likelihood-based. Results Here, we present a novel neural network model that outperforms standard phylogenetic methods and other neural network implementations under LBA settings. Furthermore, unlike existing neural network models in phylogenetics, our model naturally accounts for the tree isomorphisms via permutation invariant functions which ultimately result in lower memory and allows the seamless extension to larger trees. Availability and implementation We implement our novel theory on an open-source publicly available GitHub repository: https://github.com/crsl4/nn-phylogenetics.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2015

Mixed Initiative Visual Analytics Using Task-Driven Recommendations

Visual data analysis is composed of a collection of cognitive actions and tasks to decompose, internalize, and recombine data to produce knowledge and insight. Visual analytic tools provide interactive visual interfaces to data to support tasks involved in discovery and sensemaking, including forming hypotheses, asking questions, and evaluating and organizing evidence. Myriad analytic models can be incorporated into visual analytic systems, at the cost of increasing complexity in the analytic discourse between user and system. Techniques exist to increase the usability of interacting with such analytic models, such as inferring data models from user interactions to steer the underlying models of the system via semantic interaction, shielding users from having to do so explicitly. Such approaches are often also referred to as mixed-initiative systems. Researchers studying the sensemaking process have called for development of tools that facilitate analytic sensemaking through a combination of human and automated activities. However, design guidelines do not exist for mixed-initiative visual analytic systems to support iterative sensemaking. In this paper, we present a candidate set of design guidelines and introduce the Active Data Environment (ADE) prototype, a spatial workspace supporting the analytic process via task recommendations invoked by inferences on user interactions within the workspace. ADE recommends data and relationships based on a task model, enabling users to co-reason with the system about their data in a single, spatial workspace. This paper provides an illustrative use case, a technical description of ADE, and a discussion of the strengths and limitations of the approach.

active data↗
DOE OSTI2020

A Novel Modeling Framework for Computationally Efficient and Accurate Real-Time Ensemble Flood Forecasting With Uncertainty Quantification

A novel modeling framework that simultaneously improves accuracy, predictability, and computational efficiency is presented. It embraces the benefits of three modeling techniques integrated together for the first time: surrogate modeling, parameter inference, and data assimilation. The use of polynomial chaos expansion (PCE) surrogates significantly decreases computational time. Parameter inference allows for model faster convergence, reduced uncertainty, and superior accuracy of simulated results. Ensemble Kalman filters assimilate errors that occur during forecasting. To examine the applicability and effectiveness of the integrated framework, we developed 18 approaches according to how surrogate models are constructed, what type of parameter distributions are used as model inputs, and whether model parameters are updated during the data assimilation procedure. We conclude that (1) PCE must be built over various forcing and flow conditions, and in contrast to previous studies, it does not need to be rebuilt at each time step; (2) model parameter specification that relies on constrained, posterior information of parameters (so-called Selected specification) can significantly improve forecasting performance and reduce uncertainty bounds compared to Random specification using prior information of parameters; and (3) no substantial differences in results exist between single and dual ensemble Kalman filters, but the latter better simulates flood peaks. The use of PCE effectively compensates for the computational load added by the parameter inference and data assimilation (up to ~80 times faster). Therefore, the presented approach contributes to a shift in modeling paradigm arguing that complex, high-fidelity hydrologic and hydraulic models should be increasingly adopted for real-time and ensemble flood forecasting.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Estimation of process steady state with autoregressive models and Bayesian inference

To improve efficiency, separations engineers will typically design process circuits containing recirculating streams, which mix one or more of the process outputs with the feed material. Doing so can improve efficiency, but will cause a delay in the system reaching steady state conditions until the recirculating load mass flows stabilize. In testing separation circuits, engineers will often test a variety of factors and complete an analysis from sample results. Knowledge of if a process is at steady state, as well as the steady state conditions of a process, is essential for a valid techno-economic analysis. However, the definition of process steady state is often poorly defined, or does not include uncertainty quantification. If the performance of a process operating under two different sets of conditions are compared, an engineer who does not test for steady state or quantify steady state conditions risks producing a faulty analysis. In this work, a Bayesian statistical method for testing if all streams are at steady state is further motivated and then derived. Then after testing for steady state, the same model is used with a prior distribution that enforces a steady state assumption to estimate steady state conditions. Further, these methods were validated in a solvent extraction pilot plant where steady state conditions for all outflows were inferred with uncertainty quantification. Analysis is completed with functions available to the reader as part of the BayesMassBal (V 1.1.0) software package written in R.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2024

Analyzing inference workloads for spatiotemporal modeling

Ensuring power grid resiliency, forecasting climate conditions, and optimization of transportation infrastructure are some of the many application areas where data is collected in both space and time. Spatiotemporal modeling is about modeling those patterns for forecasting future trends and carrying out critical decision-making by leveraging machine learning/deep learning. Once trained offline, field deployment of trained models for near real-time inference could be challenging because performance can vary significantly depending on the environment, available compute resources and tolerance to ambiguity in results. Users deploying spatiotemporal models for solving complex problems can benefit from analytical studies considering a plethora of system adaptations to understand the associated performance-quality trade-offs. To facilitate the co-design of next-generation hardware architectures for field deployment of trained models, it is critical to characterize the workloads of these deep learning (DL) applications during inference and assess their computational patterns at different levels of the execution stack. In this paper, we develop several variants of deep learning applications that use spatiotemporal data from dynamical systems. We study the associated computational patterns for inference workloads at different levels, considering relevant models (Long short-term Memory, Convolutional Neural Network and Spatio-Temporal Graph Convolution Network), DL frameworks (Tensorflow and PyTorch), precision (FP16, FP32, AMP, INT16 and INT8), inference runtime (ONNX and AI Template), post-training quantization (TensorRT) and platforms (Nvidia DGX A100 and Sambanova SN10 RDU). Overall, our findings indicate that although there is potential in mixed-precision models and post-training quantization for spatiotemporal modeling, extracting efficiency from contemporary GPU systems might be challenging. Instead, co-designing custom accelerators by leveraging optimized High Level Synthesis frameworks (such as SODA High-Level Synthesizer for customized FPGA/ASIC targets) can make workload-specific adjustments to enhance the efficiency.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Sensitivity-Informed Bayesian Inference for Home PLC Network Models with Unknown Parameters

Bayesian inference is used to calibrate a bottom-up home PLC network model with unknown loads and wires at frequencies up to 30 MHz. A network topology with over 50 parameters is calibrated using global sensitivity analysis and transitional Markov Chain Monte Carlo (TMCMC). The sensitivity-informed Bayesian inference computes Sobol indices for each network parameter and applies TMCMC to calibrate the most sensitive parameters for a given network topology. A greedy random search with TMCMC is used to refine the discrete random variables of the network. This results in a model that can accurately compute the transfer function despite noisy training data and a high dimensional parameter space. The model is able to infer some parameters of the network used to produce the training data, and accurately computes the transfer function under extrapolative scenarios.

42 ENGINEERING↗
DOE OSTI2017

A reduced order model to analytically infer atmospheric CO 2 concentration from stomatal and climate data

To address questions related to the acceleration or deceleration of the global hydrological cycle or links between the carbon and water cycles over land, reliable data for past climatic conditions based on proxies are required. In particular, the reconstruction of palaeoatmospheric CO 2 content (C a ) is needed to assist the separation of natural from anthropogenic C a variability and to explore phase relations between C a and air temperature T a time series. Both T a and C a are needed to fingerprint anthropogenic signatures in vapor pressure deficit, a major driver used to explain acceleration or deceleration phases in the global hydrological cycle. Current approaches to C a reconstruction rely on a robust inverse correlation between measured stomatal density in leaves (ν) of many plant taxa and C a . There are two methods that exploit this correlation: The first uses calibration curves obtained from extant species assumed to represent the fossil taxa, thereby restricting the suitable taxa to those existing today. The second is a hybrid eco-hydrological/physiological approach that determines C a with the aid of systems of equations based on quasi-instantaneous leaf-gas exchange theories and fossil stomatal data collected along with other measured leaf anatomical traits and parameters. In this contribution, a reduced order model (ROM) is proposed that derives C a from a single equation incorporating the aforementioned stomatal data, basic climate (e.g. temperature), estimated biochemical parameters of assimilation and isotope data. The usage of the ROM is then illustrated by applying it to isotopic and anatomical measurements from three extant species. The ROM derivation is based on a balance between the biochemical demand and atmospheric supply of CO 2 that leads to an explicit expression linking stomatal conductance to internal CO 2 concentration (C i ) and C a . The resulting expression of stomatal conductance from the carbon economy of the leaf is then equated to another expression derived from water vapor gas diffusion that includes anatomical traits. When combined with isotopic measurements for long-term C i /C a , C a can be analytically determined and is interpreted as the time-averaged C a that existed over the life-span of the leaf. Key advantages of the proposed ROM are: 1) the usage of isotopic data provides constraints on the reconstructed atmospheric CO 2 concentration from ν, 2) the analytical form of this approach permits direct links between parameter uncertainties and reconstructed C a , and 3) the time-scale mismatch between the application of instantaneous leaf-gas exchange expressions constrained with longer-term isotopic data is reconciled through averaging rules and sensitivity analysis. The latter point was rarely considered in prior reconstruction studies that combined models of leaf-gas exchange and isotopic data to reconstruct C a from ν. The proposed ROM is not without its limitations given the need to a priori assume a parameter related to the control on photosynthetic rate. Our work here further explores immanent constraints for the aforementioned photosynthetic parameter.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

A tale of two planets: Disparate evolutionary models for Mars inferred from radiogenic isotope compositions of Martian meteorites

The radiogenic isotopic compositions of basaltic Martian meteorites (shergottites) and clinopyroxene/olivine cumulate meteorites (nakhlite/chassignites) are used to define the global evolution of Mars. However, the two main groups of meteorites demonstrate that their sources underwent divergent styles of magmatic evolution. The shergottites portray a planet that differentiated ~4.52 billion years ago via solidification of a magma ocean, producing incompatible element-depleted and -enriched reservoirs that remained isolated until melt production. In contrast, the reservoir from which the nakhlite/chassignites derive may have formed earlier, produced melts that fractionated Sm/Nd and Hf/W differently, was compositionally less variable, and experienced a significantly more complex history following primordial differentiation than the shergottite sources. The disparate histories recorded by these two groups of meteorites elucidate important questions that could be addressed by acquiring additional samples. Obtaining samples that shared the isotopic systematics of the shergottites would provide confidence that extrapolating the primordial differentiation history of Mars from shergottite radiogenic isotope systematics is reasonable. Returned samples from Mars will also constrain the physical locations of the meteorite source regions, providing insights into the general structure of the Martian mantle. In addition, they will help constrain the phases present in the martian mantle during melting and the conditions under which they are stable. Finally, identifying an evolved lithology that satisfies the geochemical and isotopic constraints placed on the incompatible element-enriched endmember observed in the shergottites would define the nature of magmatic evolution on Mars and whether it is more akin to processes on the Earth or the Moon.

Mars↗
DOE OSTI2019

Non-Intrusive Inference Reduced Order Model for Fluids Using Deep Multistep Neural Network

In this effort we propose a data-driven learning framework for reduced order modeling of fluid dynamics. Designing accurate and efficient reduced order models for nonlinear fluid dynamic problems is challenging for many practical engineering applications. Classical projection-based model reduction methods generate reduced systems by projecting full-order differential operators into low-dimensional subspaces. However, these techniques usually lead to severe instabilities in the presence of highly nonlinear dynamics, which dramatically deteriorates the accuracy of the reduced-order models. In contrast, our new framework exploits linear multistep networks, based on implicit Adams–Moulton schemes, to construct the reduced system. The advantage is that the method optimally approximates the full order model in the low-dimensional space with a given supervised learning task. Moreover, our approach is non-intrusive, such that it can be applied to other complex nonlinear dynamical systems with sophisticated legacy codes. We demonstrate the performance of our method through the numerical simulation of a two-dimensional flow past a circular cylinder with Reynolds number Re = 100. The results reveal that the new data-driven model is significantly more accurate than standard projection-based approaches.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2012

Inferring Land Surface Model Parameters for the Assimilation of Satellite-Based L-Band Brightness Temperature Observations into a Soil Moisture Analysis System

The Soil Moisture and Ocean Salinity (SMOS) satellite mission provides global measurements of L-band brightness temperatures at horizontal and vertical polarization and a variety of incidence angles that are sensitive to moisture and temperature conditions in the top few centimeters of the soil. These L-band observations can therefore be assimilated into a land surface model to obtain surface and root zone soil moisture estimates. As part of the observation operator, such an assimilation system requires a radiative transfer model (RTM) that converts geophysical fields (including soil moisture and soil temperature) into modeled L-band brightness temperatures. At the global scale, the RTM parameters and the climatological soil moisture conditions are still poorly known. Using look-up tables from the literature to estimate the RTM parameters usually results in modeled L-band brightness temperatures that are strongly biased against the SMOS observations, with biases varying regionally and seasonally. Such biases must be addressed within the land data assimilation system. In this presentation, the estimation of the RTM parameters is discussed for the NASA GEOS-5 land data assimilation system, which is based on the ensemble Kalman filter (EnKF) and the Catchment land surface model. In the GEOS-5 land data assimilation system, soil moisture and brightness temperature biases are addressed in three stages. First, the global soil properties and soil hydraulic parameters that are used in the Catchment model were revised to minimize the bias in the modeled soil moisture, as verified against available in situ soil moisture measurements. Second, key parameters of the "tau-omega" RTM were calibrated prior to data assimilation using an objective function that minimizes the climatological differences between the modeled L-band brightness temperatures and the corresponding SMOS observations. Calibrated parameters include soil roughness parameters, vegetation structure parameters, and the single scattering albedo. After this climatological calibration, the modeling system can provide L-band brightness temperatures with a global mean absolute bias of less than 10K against SMOS observations, across multiple incidence angles and for horizontal and vertical polarization. Third, seasonal and regional variations in the residual biases are addressed by estimating the vegetation optical depth through state augmentation during the assimilation of the L-band brightness temperatures. This strategy, tested here with SMOS data, is part of the baseline approach for the Level 4 Surface and Root Zone Soil Moisture data product from the planned Soil Moisture Active Passive (SMAP) satellite mission.

Reichle, Rolf H.↗