Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “reduced order systems”

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 595 records · Page 33
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Evaluation of Minimum NOx Emission From Ammonia Combustion

Abstract Ammonia (NH3) is being explored as a hydrogen carrier with no carbon emissions. However, if burned directly as NH3, rather than being completely decomposed back to N2/H2, the fuel-bound nitrogen comes with a potentially significant NOx emissions penalty. Indeed, several existing studies are showing ammonia combustion NOx emissions that exceed current natural gas fueled, DLN technologies by one to two orders of magnitude. Therefore, it is important to establish the theoretical minimum NOx emissions for an ammonia combustor, to determine how much NOx levels can be reduced via further technology development. In other words, the purpose of this work is not to analyze the performance of a specific combustor but, rather, the fundamental limits of what is achievable. This study quantifies this minimum NOx level for a two-stage combustor system for a given combustor exit temperature and residence time, with a constraint on unburned fuel levels. As expected, the optimum configuration is a rich front end combustor to burn and crack ammonia with significant H2 production, followed by an NO relaxation reactor, followed by a lean stage that consumes the remaining H2. The optimum residence time and stoichiometry of each zone are determined in the fast mixing limit, which essentially balances between NOx production in the primary and secondary zones. These results show minimum NOx levels are in 200–400 ppm range at 1 bar, but drop to levels of ∼25 ppm at 20 bar. These NOx emissions are dominated by NOx production in the primary stage which relaxes to equilibrium levels quite slowly. As processes controlling NOx relaxation to equilibrium in the primary stage dominate overall NO emission levels, combustor NOx sensitivities are essentially opposite that of natural gas fired, DLN systems. Specifically, NOx values drop with increased combustor residence time, increased pressure, and increased combustor exit temperature. These results also suggest that the most important strategy for NOx minimization is to provide sufficient relaxation time after the primary zone for NOx to approach equilibrium—this can be done via kinetic means to accelerate this relaxation rate, such as enhancing pressure or temperature, or increasing residence times. Indeed, this work shows that low pressure combustors specifically optimized for ammonia will have residence times that are one to two orders of magnitude larger than current natural gas systems. By doing so, NOx levels below 10 ppm may be achievable. Finally, we discuss the sensitivity of these values to uncertainties in ammonia kinetics.

Engineering↗
DOE OSTI2024

SPARC-X: Quantum simulations at extreme scale - reactive dynamics from first principles

We have developed the massively parallel electronic structure code SPARC-X: a computational framework for performing Kohn-Sham Density Functional Theory (DFT) calculations that can scale linearly with the number of atoms in the system, while being able to leverage petascale and emerging exascale parallel computers to study chemical phenomena at unprecedented length and time scales. SPARC-X exploits a recent breakthrough in electronic structure methodologies: systematically improvable, strictly local, orthonormal, discontinuous real-space bases that efficiently and systematically capture the local chemistry of the system. With further adaptation using new machine-learning techniques and the use of the massively parallel Spectral Quadrature (SQ) electronic structure method, the algorithmic complexity and prefactor associated with DFT calculations involving semilocal as well as hybrid functionals are dramatically reduced. Using petascale computational resources, SPARC-X enables quantum mechanical simulations at length and time scales previously accessible only by empirical approaches, e.g., 1,000,000 atoms for a few picoseconds using semilocal functionals or 1,000 atoms for a few picoseconds using hybrid functionals. Using exascale resources, the sizes and times targeted are two orders of magnitude larger. Such a capability has applications in a wide variety of chemical sciences, including reactive interfaces where large length- and/or long time-scales are needed and traditional force fields fail. This is particularly important in dynamic catalysis, where bond breaking and formation must be understood in detail. We developed, tested, and employed the SPARC-X framework to understand the photocatalytic properties of TiO 2 nanoparticles, revealing finite size effects that cannot be captured with standard model systems or functionals. This integrated development and application strategy ensures that SPARC-X remains a robust, efficient, and scalable software package for quantum simulations on current petascale and emerging exascale computing resources.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Climate Base State Influences on South Asian Monsoon Processes Derived From Analyses of E3SMv2 and CESM2

Abstract The effects of differences in climate base state are related to processes associated with the present‐day South Asian monsoon simulations in the Energy Exascale Earth System Model version 2 (E3SMv2) and the Community Earth System Model version 2 (CESM2). Though tropical Pacific and Indian Ocean base state sea surface temperatures (SSTs) are over 1°C cooler in E3SMv2 compared to CESM2, and there is an overall reduction of Indian sector precipitation, the pattern of South Asian monsoon precipitation is similar in the two models. Monsoon‐ENSO teleconnections, dynamically linked by the large‐scale east‐west atmospheric circulation, are reduced in E3SMv2 compared to CESM2. In E3SMv2, this is related to cooler tropical SSTs and ENSO amplitude that is less than half that in CESM2. Comparison to a tropical Pacific pacemaker experiment shows, to a first order, that the base state SSTs and ENSO amplitude contribute roughly equally to lower amplitude monsoon‐ENSO teleconnections in E3SMv2.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Confronting Domain Shift in Trained Neural Networks

Neural networks (NNs) are known as universal function approximators and can interpolate nonlinear functions between observed data points. However, when the target domain for deployment shifts from the training domain and NNs must extrapolate, the results are notoriously poor. Prior work Martinez et al. (2019) has shown that NN uncertainty estimates can be used to correct binary predictions in shifted domains without retraining the model. We hypothesize that this approach can be extended to correct real-valued time series predictions. As an exemplar, we consider two mechanical systems with nonlinear dynamics. The first system consists of a spring-mass system where the stiffness changes abruptly, and the second is a real experimental system with a frictional joint that is an open challenge for structural dynamicists to model efficiently. Our experiments will test whether 1) NN uncertainty estimates can identify when the input domain has shifted from the training domain and 2) whether the information used to calculate uncertainty estimates can be used to correct the NN’s time series predictions. While the method as proposed did not significantly improve predictions, our results did show potential for modifications that could improve models’ predictions and play a role in structural health monitoring systems that directly impact public safety.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Role of Cation Ordering on Device Performance in (Ag,Cu)InSe 2 Solar Cells with KF Post-Deposition Treatment

CuInSe 2 (CIS) has been proposed as an attractive bottom cell candidate in tandem solar cells. However, to justify the coupling with high-performance top cells (e.g., perovskites, GaAs), significant work on improving the efficiency is required. To this extent, several authors have demonstrated the benefits of alkali post-deposition treatments (PDT) to increase device open-circuit voltage (V oc ) in CIS and how Ag alloying - (Ag,Cu)InSe 2 (ACIS) - reduces defect density and enhances current collection in devices. Herein, we present a detailed study of the role that KF-PDT plays on CIS and ACIS absorber composition and structure, and propose an explanation for the decreased V oc observed when silver and potassium coexist in the system (ACIS + KF). Through a suite of synchrotron-based techniques, we investigate the nanoscale chemical distribution of the films and the formation of secondary phases. Through photoluminescence imaging, we observed a high degree of passivation with the addition of KF, and synchrotron-based X-ray diffraction confirmed the absence of a KInSe 2 surface layer usually considered to be a passivating agent. Raman spectroscopy and synchrotron X-ray fluorescence show the increased presence of Cu- and Se-poor clusters in ACIS + KF, which are correlated to significantly reduced X-ray beam-induced current (XBIC). An increase in the intensity of the E/B 2 stretching mode of CIS is attributed to cation ordering near the junction and is found to track inversely to bulk V oc measurements. The cation ordering is hypothesized to arise from the formation and redistribution of defects that normally occur near the surfaces of CIS as a consequence of its polar character. Here, these defects compensate each other, and the overall inhomogeneity of the charge distribution generates electrostatic potential fluctuations that greatly increase the saturation current and hence reduce the open-circuit voltage of the device.

36 MATERIALS SCIENCE↗
DOE OSTI2020

A versatile and low-cost chip-to-world interface: enabling ICP-MS characterization of isotachophoretically separated lanthanides on a microfluidic device

Microfluidics offer novel and state-of-the-art pathways to process materials. Microfluidic systems drastically reduce timeframes and costs associated with traditional lab-scale efforts in the area of analytical sample preparations. The challenge arises in effectively connecting microfluidics to off-chip analysis tools to accurately characterize samples after treatment on-chip. Here, a fused silica microfluidic device (MFD) is connected to a fused-silica capillary in order to create a world-to-chip connection between the MFD and an inductively coupled plasma mass spectrometer (ICP-MS). Specifically, this is demonstrated by utilizing isotachophoresis (ITP) on a microfluidic chip to separate samples containing up to 14 lanthanides into elementally pure bands. The separated sample is successfully transferred across a 7 nL dead space at the microchip-capillary junction, such that separation resolution is maintained and even increased through the interface and into the ICP-MS, where the elemental composition of the separated sample is analysed. Lanthanide samples of varying composition are detected using ICP-MS, demonstrating this versatile and cost-effective approach that does not degrade the separation quality achieved on the MFD. This simple and inexpensive connection enables fast, low-cost sample preparation immediately prior to injection into an ICP-MS or other analytical instruments.

Lackey, Hope E.↗
DOE OSTI2025

Efficient shallow Ritz method for 1D diffusion problems

This paper studies the shallow Ritz method for solving the one-dimensional diffusion problem. It is shown that the shallow Ritz method improves the order of approximation dramatically for non-smooth problems. To realize this optimal or nearly optimal order of the shallow Ritz approximation, we develop a damped block Newton (dBN) method that alternates between updates of the linear and non-linear parameters. Per each iteration, the linear and the non-linear parameters are updated by exact inversion and one step of a modified, damped Newton method applied to a reduced non-linear system, respectively. The computational cost of each dBN iteration is $\mathcal{O}$(n). Starting with the non-linear parameters as a uniform partition of the interval, numerical experiments show that the dBN is capable of efficiently moving mesh points to nearly optimal locations. In conclusion, to improve the efficiency of the dBN further, we propose an adaptive damped block Newton (AdBN) method by combining the dBN with the adaptive neuron enhancement (ANE) method [28].

Diffusion problems↗
DOE OSTI2021

Magnetic and structural properties of Mn X NiSn ( X = Mn, Fe, Co)

Crystal structure and magnetic properties of Heusler alloys MnXNiSn (X = Mn, Fe, Co) are investigated using density functional theory and compared with experimental results. The parent alloy Mn 2 NiSn, which crystallizes in the inverse Heusler structure, is found to be ferrimagnetic, in agreement with previous experimental and theoretical work. The Fe and Co substitutions cause the alloys to assume a fairly well-ordered Y structure and enhance the magnetization substantially. We find that the strong nearest neighbour Mn-X exchange changes from antiferromagnetic (X = Mn) to ferromagnetic (X = Fe, Co), which explains and actually overestimates the experimental changes. A striking feature of the system is that Fe and Co have opposite effects on the Curie temperature T c : they reduce and enhance T c , respectively. We qualitatively explain this behaviour in terms of two-sublattice model based on the nearest-neighbour exchange.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Medium and Heavy-Duty Electric Vehicle Infrastructure - Load Operations and Deployment (HEVI-LOAD) v1.0

In order to support the California state goals to reduce the carbon and pollution emissions (AB2127), The goal of this project (HEVI-Pro) is to project the charging infrastructure needs by future medium- and heavy- duty electric vehicles, and assess optimal deployment the infrastructure in California and provide grid impacts analysis. The MHDV projections will also consider transportation system and electric system interaction with light duty vehicles. The project consists of 2 approaches in general: top-down approach that takes aggregated MHDV adoptions as the inputs and provide the county-level projections of charging load profile and infrastructure need, and the bottom-up approach that that incorporates more granular (temporal, spatial and duty-cycle-specific) behaviors of a variety of MHDVs into activity simulations/optimizations for further analysis.

Wang, Bin↗
DOE OSTI2022

Design and test of a special magnet for the Mu2e experiment

The Mu2e experiment (Muon-to-Electron Conversion Experiment) is looking for evidence that a muon can change into an electron without the emission of neutrinos. The experiment will search for the charged-lepton flavor violating (CLFV) neutrino-less conversion of a negative muon into an electron in the field of a nucleus. The conversion process results in a monochromatic electron with an energy of 104.97 MeV, slightly below the muon rest mass. The goal of the experiment is to improve the previous upper limit by four orders of magnitude and reach a SES (single event sensitivity) of 3 × 10 -17 on the conversion rate, a 90% CL of 8 × 10 -17 , and a 5σ discovery reach at 2 × 10 -16 . The experiment will use an intense pulsed negative muon beam, which is essential to reducing backgrounds. The other essential element is a sophisticated magnetic system composed of three consecutive solenoids (figure 1) that form the muon beam.

43 PARTICLE ACCELERATORS↗
DOE OSTI2022

On the Verification of Deep Reinforcement Learning Solution for Intelligent Operation of Distribution Grids

Capabilities of deep reinforcement learning (DRL) in obtaining fast decision policies in high dimensional and stochastic environments have led to its extensive use in operational research, including the operation of distribution grids with high penetration of distributed energy resources (DER). However, the feasibility and robustness of DRL solutions are not guaranteed for the system operator, and hence, those solutions may be of limited practical value. This paper proposes an analytical method to find feasibility ellipsoids that represent the range of multi-dimensional system states in which the DRL solution is guaranteed to be feasible. Empirical studies and stochastic sampling determine the ratio of the discovered to the actual feasible space as a function of the sample size. In addition, the performance of logarithmic, linear, and exponential penalization of infeasibility during the DRL training are studied and compared in order to reduce the number of infeasible solutions

Hosseini, Mohammad Mehdi↗
DOE OSTI2021

Preparing Lamellae from Vitreous Biological Samples using a Dual-Beam Scanning Electron Microscope for Cryo-Electron Tomography

Presented here is a protocol for preparing cryo-lamellae from plunge-frozen grids of Plasmodium falciparum-infected human erythrocytes, which could easily be adapted for other biological samples. The basic principles for preparing samples, milling, and viewing lamellae are common to all instruments and the protocol can be followed as a general guide to on-grid cryo-lamella preparation for cryo-electron microscopy (cryoEM) and cryo-electron tomography (cryoET). Electron microscopy grids supporting the cells are plunge-frozen into liquid nitrogen-cooled liquid ethane using a manual or automated plunge freezer, then screened on a light microscope equipped with a cryo-stage. Frozen grids are transferred into a cryo-scanning electron microscope equipped with a focused ion beam (cryoFIB-SEM). Grids are routinely sputter coated prior to milling, which aids dispersal of charge build-up during milling. Alternatively, an e-beam rotary coater can be used to apply a layer of carbon-platinum to the grids, the exact thickness of which can be more precisely controlled. Once inside the cryoFIB-SEM an additional coating of an organoplatinum compound is applied to the surface of the grid via a gas injection system (GIS). This layer protects the front edge of the lamella as it is milled, the integrity of which is critical for achieving uniformly thin lamellae. Regions of interest are identified via SEM and milling is carried out in a step-wise fashion, reducing the current of the ion beam as the lamella reaches electron transparency, in order to avoid excessive heat generation. A grid with multiple lamellae is then transferred to a transmission electron microscope (TEM) under cryogenic conditions for tilt-series acquisition. A robust and contamination-free workflow for lamella preparation is an essential step for downstream techniques, including cellular cryoEM, cryoET, and sub-tomogram averaging. Development of these techniques, especially for lift-out and milling of high-pressure frozen samples, is of high-priority in the field.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Herbaceous Feedstock 2019 State of Technology Report

The U.S. Department of Energy (DOE) promotes the production of advanced liquid transportation fuels from lignocellulosic biomass by funding fundamental and applied research that advances the State of Technology (SOT). As part of its involvement with this mission, Idaho National Laboratory (INL) completes an annual SOT report for biomass feedstock logistics. This report summarizes supply system impacts of Bioenergy Technologies Office (BETO)-funded research and development efforts at INL and elsewhere (such as the High-Tonnage Feedstock Logistics projects (Webb et al. 2013a, Webb et al. 2013b, Webb et al. 2013c, Webb and Sokhansanj 2014, Sokhansanj et al. 2014)) that lead to improvements in feedstock supply systems. These include improvements to and observed performance of innovative harvest and collection methods, storage technologies, transportation and handling approaches, and advanced preprocessing technologies. Biomass quality and variability, and the interface between feedstock quality and conversion performance are key drivers in addition to delivered feedstock cost. In this report, we estimate the benefits of R&D improvements to individual supply system unit operations and present the status of feedstock logistics technology development for converting biomass into biofuels. These analyses are supported by experimental data where possible and help to align the SOT relative to the cost goals defined in the Multi-Year Program Plan. The 2019 Herbaceous SOT incorporates several technology changes in feedstock preprocessing and introduces opportunities from the integrated landscape management (ILM) strategy and increased grower participation to reduce biomass access costs, while maintaining or improving grower profitability. During FY18 uneven flow from the horizontal bale grinder was identified as a significant issue limiting preprocessing system throughput. Based on FSL-funded research at INL, the 2019 Herbaceous SOT replaces the horizontal bale grinder used in the first stage size reduction with a bale processor. The improved uniformity of biomass flow entering the PDU eliminated slugging flow from the first stage size reduction and improved the throughput of downstream operations. In order to achieve moisture reduction through frictional heating during grinding (which allowed elimination of the costly rotary drum dryer in previous SOTs), the second stage grinder was changed from a rotary shear, which does not remove moisture, back to a hammer mill. Finally, the 2019 Herbaceous SOT introduces modified three-pass and two-pass corn stover supply curves derived from the BT16 resource assessment, based on FY19 modeling results (WBS 4.2.1.20) quantifying economic benefits of ILM in the supply area, together with modeling results (WBS 1.2.1.5) identifying ILM strategies to increase grower participation. The 2019 Herbaceous SOT report documents the current modeled cost of an herbaceous feedstock supply system from harvest to the pretreatment reactor throat for hydrocarbon fuel production via biochemical conversion, based on equipment and processes now available or potentially available in the near term. The modeled cost also considers both the required quality and the availability of the biomass resources. The 2019 Herbaceous SOT predicts a modeled delivered feedstock cost of $81.37 /dry ton (2016$); this is a $2.30/dry ton (2016$) decrease from the 2018 Herbaceous SOT. Technology improvements that contributed to this modeled cost reduction include reduced cost for the new preprocessing design and quantification of the opportunities of the integrated landscape management (ILM) strategy and an increased grower participation rate to reduce the grower payment portion of biomass access costs, while maintaining or improving grower profitability. A greenhouse gas emissions (GHG) assessment was completed by Argonne National Laboratory using the 2019 Greenhouse Gases, Regulated Emissions, and Energy use in Transportation model, estimating an increase of 14.89 kg CO2e/ton from the 2018 SOT (69.27 kg CO2e/ton in 2018 to 84.16 kg CO2e/ton in 2019). The increase of energy consumption during preprocessing along with higher transportation distance to access low cost biomass from further distance contributed to the increase of GHG emissions in the 2019 Herbaceous SOT. The reason for the increased transportation distances was the cost tradeoff of going farther from the biorefinery to access the cheaper ILM-derived counties (the cheaper price outweighed the cost of increased supply radius).

09 BIOMASS FUELS↗
DOE OSTI2025

Physics-Informed Graph Neural Networks for Collaborative Dynamic Reconfiguration and Voltage Regulation in Unbalanced Distribution Systems

Network reconfiguration has long been employed as a strategic approach to minimize power distribution system losses and effectively regulate voltage levels. Tap-changing voltage regulators are also critical for controlling bus voltages, especially in accommodating the increasing integration of distributed energy resources (DERs) with intermittent outputs. This paper introduces novel methodologies to address the challenges of dynamic reconfiguration and optimal tap setting in unbalanced three-phase distribution systems. We propose an approximated mixed-integer quadratically constrained program (MIQCP) to model dynamic reconfiguration, along with a pioneering formulation for voltage regulator (VR) tap-setting based on Special Ordered Set type 1 (SOS1). To mitigate computational complexity, we propose a physics-informed spatial-temporal graph convolutional network (STGCN) with an integrated link classifier. The proposed approach enables efficient solution generation by fixing specific variables in the MIQCP instance and solving the simplified sub-MIP using an MIP solver. Numerical studies demonstrate the superior prediction accuracy of our STGCN model compared to baseline neural network models, resulting in reduced DER curtailment and voltage deviation with shorter computation time.

dynamic reconfiguration↗
DOE OSTI2020

An accelerated approach to inline non-LTE modeling

A goal across many physics and engineering disciplines is to use radiation-hydrodynamics codes to accurately simulate systems that are expected to deviate from local thermodynamic equilibrium (LTE), such as laser-driven hohlraums, stellar coronae, and supernova ejecta. The computational expense of traditional non-LTE (NLTE) opacity calculations can dominate the cost of simulating these systems and render NLTE physics untenable in most cases. In this paper, we present a new parallel implementation for NLTE opacity calculations that employs advanced load balancing, vectorization, and memory management to significantly reduce the computational costs of these calculations. Using the radiation-hydrodynamics code xRAGE , we demonstrate this new implementation on three test problems and present order-of-magnitude speed improvements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Reduced order models for thermal radiative transfer problems based on moment equations and data-driven approximations of the Eddington tensor

Here a new group of structure and asymptotic preserving reduced-order models (ROMs) for multidimensional nonlinear thermal radiative transfer (TRT) problems is presented. They are formulated by means of the nonlinear projective approach and data compression techniques. The nonlinear projection is applied to the Boltzmann transport equation (BTE) to derive a hierarchy of low-order moment equations. Approximation of the Eddington tensor that provides exact closure for the system of moment equations is found with projection-based data-driven methodologies. These include the (i) proper orthogonal decomposition (POD), (ii) dynamic mode decomposition (DMD) and (iii) a variant of the DMD. A parameterization is derived for this ROM for the temperature of radiation incoming to the problem domain (the radiation drive temperature). This parameterization is informed from results of a dimensionless study of the TRT problem. Analysis of the ROMs is performed on the classical Fleck-Cummings TRT multigroup test problem in 2D geometry with a radiation-driven Marshak wave. Numerical results are presented to demonstrate the performance of these ROMs for the simulation of evolving radiation and heat waves. Results show these models to be sufficiently accurate for practical computations with rather low-rank representations of the Eddington tensor. As the rank of the approximation is increased, the errors of solutions generated by the ROMs gradually decreases.

42 ENGINEERING↗
DOE OSTI2022

The Mu2e experiment — Searching for charged lepton flavor violation

The Mu2e experiment will search for a Standard Model violating rate of neutrinoless conversion of a muon into an electron in the presence of an aluminum nucleus. Observation of this charged lepton flavor violating process would be an unambiguous sign of new physics. Mu2e will improve upon previous searches for this process by four orders of magnitude. This requires the world’s highest-intensity muon beam, a detector system capable of efficiently reconstructing the 105 MeV/c conversion electron signal, and minimizing sensitivity to background events. A pulsed 8 GeV proton beam strikes a target, producing pions that decay into muons. Beam outside the pulse must be suppressed to < 1 0 – 10 to reduce beam-related backgrounds. The muon beam is guided from the production target along the transport system and onto the aluminum stopping target. Conversion electrons leave the stopping target and propagate inside a solenoidal magnetic field to the tracker and electromagnetic calorimeter. The tracker is a system of straw tube panels filled with Ar/CO 2 at 1 atm that tracks particles inside of a solenoidal B-field and measures their momenta with ~ 100 keV/c resolution to resolve signal events from decay-in-orbit backgrounds. The CsI calorimeter provides E / p and is used to seed the track reconstruction algorithm with σ E / E ~ 10 % and σ t < 500 ps . Additionally, a novel cosmic ray veto with greater than 99.99% efficiency brings the expected number of background events to fewer than one over three years of running. Finally, to normalize the experiment, the stopping target monitor measures the rate of capture photons from muons incident on the stopping target by using a system of high-purity germanium and lanthanum bromide scintillators.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

A Non‐Intrusive Machine Learning Framework for Debiasing Long‐Time Coarse Resolution Climate Simulations and Quantifying Rare Events Statistics

Abstract Due to the rapidly changing climate, the frequency and severity of extreme weather is expected to increase over the coming decades. As fully‐resolved climate simulations remain computationally intractable, policy makers must rely on coarse‐models to quantify risk for extremes. However, coarse models suffer from inherent bias due to the ignored “sub‐grid” scales. We propose a framework to non‐intrusively debias coarse‐resolution climate predictions using neural‐network (NN) correction operators. Previous efforts have attempted to train such operators using loss functions that match statistics. However, this approach falls short with events that have longer return period than that of the training data, since the reference statistics have not converged. Here, the scope is to formulate a learning method that allows for correction of dynamics and quantification of extreme events with longer return period than the training data. The key obstacle is the chaotic nature of the underlying dynamics. To overcome this challenge, we introduce a dynamical systems approach where the correction operator is trained using reference data and a coarse model simulation nudged toward that reference. The method is demonstrated on debiasing an under‐resolved quasi‐geostrophic model and the Energy Exascale Earth System Model (E3SM). For the former, our method enables the quantification of events that have return period two orders longer than the training data. For the latter, when trained on 8 years of ERA5 data, our approach is able to correct the coarse E3SM output to closely reflect the 36‐year ERA5 statistics for all prognostic variables and significantly reduce their spatial biases.

Barthel Sorensen, B.↗