SEARCH · Engineering Papers

Results for “cost model”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 721 records · Page 40

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Accelerating multilevel Markov Chain Monte Carlo using machine learning models

Here, this work presents an efficient approach for accelerating multilevel Markov Chain Monte Carlo (MCMC) sampling for large-scale problems using low-fidelity machine learning models. While conventional techniques for large-scale Bayesian inference often substitute computationally expensive high-fidelity models with machine learning models, thereby introducing approximation errors, our approach offers a computationally efficient alternative by augmenting high-fidelity models with low-fidelity ones within a hierarchical framework. The multilevel approach utilizes the low-fidelity machine learning model (MLM) for inexpensive evaluation of proposed samples thereby improving the acceptance of samples by the high-fidelity model. The hierarchy in our multilevel algorithm is derived from geometric multigrid hierarchy. We utilize an MLM to accelerate the coarse level sampling. Training machine learning model for the coarsest level significantly reduces the computational cost associated with generating training data and training the model. We present an MCMC algorithm to accelerate the coarsest level sampling using MLM and account for the approximation error introduced. We provide theoretical proofs of detailed balance and demonstrate that our multilevel approach constitutes a consistent MCMC algorithm. Additionally, we derive the expression for cost reduction due to machine learning model to facilitate cost analysis of the hierarchical sampling algorithm. Our technique is demonstrated on a standard benchmark inference problem in groundwater flow, where we estimate the probability density of a quantity of interest using a four-level MCMC algorithm. Our proposed algorithm accelerates multilevel sampling by a factor of two while achieving similar accuracy compared to sampling using the standard multilevel algorithm.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Cost optimization of low-salt-rejection reverse osmosis

Low-salt-rejection reverse osmosis (LSRRO) is an emerging membrane-based desalination technology for concentrating brines with potentially lower energy consumption and cost than thermally driven processes. We assess the technoeconomic performance of LSRRO by using an optimization model that minimizes the levelized cost of water (LCOW). We develop detailed models for the RO module and other equipment using WaterTAP (Water treatment Technoeconomic Assessment Platform), an open-source simulation and optimization software for water treatment. We improve upon previous LSRRO cost optimization studies by optimizing the salt permeability for each stage, which has significant implications for technoeconomic performance. We present the cost-optimal design, operation, and performance for three cases: 35 g/L TDS feed at 70% recovery, 70 g/L at 55%, and 125 g/L at 35%. The cost-optimal LCOW and specific energy consumption (SEC) for these cases are 0.70, 1.89, and 7.41 $/m 3 and 3.9, 8.4, and 30.4 kWh/m 3 , respectively. For each case, we use sensitivity analyses to provide guidance on key factors that impact economic viability. We also present the cost-optimal LCOW and SEC for feed concentrations between 5 and 220 g/L TDS and recoveries between 30 and 90%. So, we compare LSRRO to other high-salinity desalination technologies and find that it may be cost-competitive below feed concentrations of 125 g/L.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

Quantifying spatiotemporal variability in occupant exposure to an indoor airborne contaminant with an uncertain source location

Well-mixed zone models are often employed to compute indoor air quality and occupant exposures. While effective, a potential downside to assuming instantaneous, perfect mixing is underpredicting exposures to high intermittent concentrations within a room. When such cases are of concern, more spatially resolved models, like computational-fluid dynamics methods, are used for some or all of the zones. But, these models have higher computational costs and require more input information. A preferred compromise would be to continue with a multi-zone modeling approach for all rooms, but with a better assessment of the spatial variability within a room. Here to do so, we present a quantitative method for estimating a room's spatiotemporal variability, based on influential room parameters. Our proposed method disaggregates variability into the variability in a room's average concentration, and the spatial variability within the room relative to that average. This enables a detailed assessment of how variability in particular room parameters impacts the uncertain occupant exposures. To demonstrate the utility of this method, we simulate contaminant dispersion for a variety of possible source locations. We compute breathing-zone exposure during the releasing (source is active) and decaying (source is removed) periods. Using CFD methods, we found after a 30 minutes release the average standard deviation in the spatial distribution of exposure was approximately 28% of the source average exposure, whereas variability in the different average exposures was lower, only 10% of the total average. We also find that although uncertainty in the source location leads to variability in the average magnitude of transient exposure, it does not have a particularly large influence on the spatial distribution during the decaying period, or on the average contaminant removal rate. By systematically characterizing a room's average concentration, its variability, and the spatial variability within the room important insights can be gained as to how much uncertainty is introduced into occupant exposure predictions by assuming a uniform in-room contaminant concentration. We discuss how the results of these characterizations can improve our understanding of the uncertainty in occupant exposures relative to well-mixed models.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2025

Open-source microreactor design models for technoeconomic assessments

Technoeconomic analyses for advanced nuclear technologies are essential for identifying cost drivers which allow for optimizing the technology to achieve better economic performance. Microreactors are emerging as a promising and reliable-energy solution, offering inherent safety, low capital investment, and rapid deployment capabilities. Past studies have conducted technoeconomic analyses for microreactors, but there remains a need for open-source technoeconomic models that can enhance collaboration, transparency, and consistency when performing this type of analysis. The present article introduces an open-source technoeconomic model for microreactors based on bottom-up cost estimation. Since no real cost data for microreactors exists, this model leverages cost data and insights gleaned from the Microreactor Applications, Research, Validation and Evaluation (MARVEL) project. It estimates the first-of-a-kind cost of two microreactor technologies and can also calculate the N th of a kind (NOAK) cost via accounting for learning and mass factory production. The two technologies considered in this paper are the liquid–metal thermal reactor (LTMR) and the gas-cooled microreactor (GCMR). In conclusion, the goal of this study is to demonstrate bottom-up cost estimation of these microreactor technologies and provide the cost estimation models that other users can leverage for various applications such as design optimization and financial planning.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2026

Value of Geothermal Energy Storage for Supply-Side and Demand-Side Applications

This report presents the results of a study examining the value potential for geothermal energy storage (GES), a long-duration energy storage resource that stores thermal and/or geomechanical energy in the subsurface. GES could benefit the overall U.S. power system by temporally shifting electricity generation (supply-side) or meeting building heating and cooling load (demand-side). This report analyzes supply-side and demand-side opportunities independently because of differences in applications and models. Currently there is significant uncertainty about the development costs for GES, with only a limited number of demonstration plants for electric energy storage and building heating and cooling storage developments. In this report, we estimate the value of supply-side and demand-side GES to the bulk power system in the contiguous United States. Because of the significant uncertainty about GES development costs, this analysis does not consider GES deployment costs but instead focuses on the value of GES to the U.S. electricity system. The estimated values of GES provide reference points for economically competitive commercial cost targets. Supply-side GES is modeled as part of an enhanced geothermal system (EGS) generation plant in NREL's Regional Energy Deployment System (ReEDS) capacity expansion model (Ho et al. 2021). In contrast to conventional geothermal plants, which generate constant power, EGS plants have unique features that may allow for in-reservoir energy storage for flexible generation. Demand-side GES for heating and cooling, including seasonal hot and cold storage and short-duration heat pump storage, is incorporated into a price-taker model using Cambium electricity marginal cost projections. To establish an upper bound for the value of GES, analysis focused on favorable scenarios for storage with high generation from zero marginal cost, variable renewable energy resources. High penetrations of variable renewable energy generation can increase hourly electricity price variability, which increases the value of temporal energy arbitrage for storage technologies like GES.

15 GEOTHERMAL ENERGY↗

DOE OSTI2024

Cost effectiveness of preemptive school closures to mitigate pandemic influenza outbreaks of differing severity in the United States

Background: Nonpharmaceutical interventions (NPIs) may be considered as part of national pandemic preparedness as a first line defense against influenza pandemics. Preemptive school closures (PSCs) are an NPI reserved for severe pandemics and are highly effective in slowing influenza spread but have unintended consequences. Methods: We used results of simulated PSC impacts for a 1957-like pandemic (i.e., an influenza pandemic with a high case fatality rate) to estimate population health impacts and quantify PSC costs at the national level using three geographical scales, four closure durations, and three dismissal decision criteria (i.e., the number of cases detected to trigger closures). At the Chicago regional level, we also used results from simulated 1957-like, 1968-like, and 2009-like pandemics. Our net estimated economic impacts resulted from educational productivity costs plus loss of income associated with providing childcare during closures after netting out productivity gains from averted influenza illness based on the number of cases and deaths for each mitigation strategy. Results: For the 1957-like, national-level model, estimated net PSC costs and averted cases ranged from $\$7.5$ billion (2016 USD) averting 14.5 million cases for two-week, community-level closures to $\$97$ billion averting 47 million cases for 12-week, county-level closures. We found that 2-week school-by-school PSCs had the lowest cost per discounted life-year gained compared to county-wide or school district–wide closures for both the national and Chicago regional-level analyses of all pandemics. The feasibility of spatiotemporally precise triggering is questionable for most locales. Theoretically, this would be an attractive early option to allow more time to assess transmissibility and severity of a novel influenza virus. However, we also found that county-wide PSCs of longer durations (8 to 12 weeks) could avert the most cases (31–47 million) and deaths (105,000–156,000); however, the net cost would be considerably greater ($\$88$-$\$103$ billion net of averted illness costs) for the national-level, 1957-like analysis. Conclusions: We found that the net costs per death averted ($\$180,000$-$\$4.2$ million) for the national-level, 1957-like scenarios were generally less than the range of values recommended for regulatory impact analyses ($\$4.6$ to 15.0 million). This suggests that the economic benefits of national-level PSC strategies could exceed the costs of these interventions during future pandemics with highly transmissible strains with high case fatality rates. In contrast, the PSC outcomes for regional models of the 1968-like and 2009-like pandemics were less likely to be cost effective; more targeted and shorter duration closures would be recommended for these pandemics.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2024

A phase-field diffraction model for thermo-hydro-mechanical propagating fractures

This paper introduces a novel diffraction based thermo-hydraulic–mechanical (THM) model for fracture propagation using a phase-field fracture (PFF) approach. The key innovation of the THM-PFF model lies in its integrated treatment of four solution variables—displacements, phase-field, pressure, and temperature—each governed by a combination of conservation of momentum (mechanics problem), a variational inequality (constrained minimization problem), mass conservation (pressure problem), and energy conservation (temperature problem). This leads to a new formulation of a coupled variational inequality system. A major advancement is the development of an extended fixed-stress algorithm, where displacements, phase-field, pressures, and temperatures are solved in a staggered sequence. An important aspect of this work is the global coupling of pressures and temperatures across the domain using diffraction systems, with diffraction coefficients defined by material parameters weighted by the diffusive phase-field variable. To ensure robust local mass conservation, we employ enriched Galerkin finite elements (EG) for both pressure and temperature diffraction equations. By enriching the continuous Galerkin basis functions with discontinuous piecewise constants, EG accurately represents solution and parameter discontinuities while preserving local mass and energy conservation—crucial aspects for THM problems and realistic behavior. Moreover, the use of a predictor–corrector local mesh adaptivity scheme is employed, allowing the model to handle small phase-field length-scale parameters while maintaining high numerical accuracy and reasonable computational cost. Furthermore, these new model and algorithmic developments represent significant advances in the field and have been substantiated through rigorous numerical tests.

Diffraction systems↗

DOE OSTI2023

Gumby: Quantifying multi-modal model resiliency

With the rise of cheap data and sensors, more use cases are emerging for multi-input models. Research has shown that including multiple data modalities can improve performance, suggesting that deep learning models can successfully learn to leverage complementary information from different modalities. However, this improved predictive power comes with unanticipated costs: additional inputs change model resiliency and expand the threat space for adversarial attacks. We first provide theoretical underpinnings for how adversarial success scales with input dimension. We then characterize the performance of a suite of multispectral deep learning models with different fusion approaches, quantify their relative reliance on different input bands, and evaluate their robustness to naturalistic and adversarial image corruptions.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

Decomposing Loosely Coupled Mixed-Integer Programs for Optimal Microgrid Design

Microgrids are frequently employed in remote regions, in part because access to a larger electric grid is impossible, difficult, or compromises reliability and independence. Although small microgrids often employ spot generation, in which a diesel generator is attached directly to a load, microgrids that combine these individual loads and augment generators with photovoltaic cells and batteries as a distributed energy system are emerging as a safer, less costly alternative. In this work, we present a model that seeks the minimum-cost microgrid design and ideal dispatched power to support a small remote site for one year with hourly fidelity under a detailed battery model; this mixed-integer nonlinear program (MINLP) is intractable with commercial solvers but loosely coupled with respect to time. A mixed-integer linear program (MIP) approximates the model, and a partitioning scheme linearizes the bilinear terms. We introduce a novel policy for loosely coupled MIPs in which the system reverts to equivalent conditions at regular time intervals; this separates the problem into subproblems that we solve in parallel. We obtain solutions within 5% of optimality in at most six minutes across 14 MIP instances from the literature and solutions within 5% of optimality to the MINLP instances within 20 minutes.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

Markov Process to Evaluate the Value Proposition of a Risk-Informed Predictive Maintenance Strategy

To achieve high-capacity factors, the nuclear fleet has relied on labor-intensive and time-consuming operation and preventive maintenance programs for plant systems. Manually performed inspection, calibration, testing, and maintenance of plant assets at periodic frequencies, along with time-based replacement of assets irrespective of condition, have resulted in a costly, labor-centric business model. Fortunately, there are technologies that can eliminate unnecessary preventive maintenance activities by deploying risk-informed predictive maintenance, enabling the transition to a technology-centric business model. The technology-centric business model will enable plants to optimize and automate maintenance activities, leading to cost reductions since labor is a rising cost and technology is a declining cost. The implementation of scalable technologies and methodologies across plant systems and across the nuclear fleet is critical for successful deployment of a risk-informed predictive maintenance strategy at commercial nuclear power plants. The work presented in the report is being developed as part of a collaborative research effort between Idaho National Laboratory and Public Service Enterprise Group Nuclear, LLC. This report describes the technical basis using the Markov Process to evaluate the value proposition for the risk-informed predictive maintenance strategy for the circulating water system. The plant process data from the Salem nuclear plant’s circulating water system is utilized to develop a Markov chain risk models and formulation to estimate the loss and gain in revenue based on plant availability. The outcomes presented in this report provides the technical basis for extensive quantitative evaluation of a scalable risk-informed predictive maintenance strategy as part of future research.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2022

Home charging for all: Techno-economic and life cycle assessment of multi-unit dwelling electric vehicle charging hubs

Ubiquitous electric vehicle adoption can drastically reduce greenhouse gas emissions (GHG) but will require equitable home charging infrastructure for all residences. Unlike single-family homes, Multi-Unit Dwellings (MUD) currently lack market share and access to nearby charging infrastructure partly due to expensive capital costs for reluctant residential property-owners. Further, this work evaluates the levelized cost of charging (LCOC) for Battery Electric Vehicles (BEVs) at MUD community charging hubs through a techno-economic analysis (TEA) that leverages real-world charging data and costs. Three different MUD charging types are investigated for a baseline and optimistic case: 1.9-kW Level 1 (L1), 6.6-kW Level 2 (L2), and 50-kW Direct Current Fast Charging (DCFC) stations under three different ownership models: resident, utility, and private company. Results demonstrate L1 and L2 chargers to be less expensive than the gasoline equivalent across the United States. Further, utility and private company ownership models, which avoid initial costs for residential property owners, result in a large LCOC premium for L1 and moderate LCOC premium for L2 relative to the resident ownership model. In contrast, DCFC is shown to be expensive for baseline scenarios but economical for optimistic scenarios especially under private company ownership. This work also performs a cradle to grave (C2G) life cycle assessment (LCA) of an average passenger BEV and gasoline conventional vehicle (CV) using yearly (grid mix & vehicle parameters), hourly (grid mix), and state-level (grid mix) resolution. Results show BEVs to have lower GHG emissions (-86% to-10%) than gasoline CVs in the contiguous U.S. Next, the system boundary of the TEA is extended to the total cost of ownership (TCO) of BEVs. The TCO is then coupled with the C2G GHG emissions to calculate the cost of GHG emissions reduction. Ultimately, the cost of GHG emissions reduction from MUD BEVs relative to gasoline CVs is shown to be negative for every scenario except baseline DCFC, meaning MUD BEV charging infrastructure can be a cost-effective endeavor to reduce GHG emissions.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2024

Scale effects on core design, fuel costs, and spent fuel volume of pressurized water reactors

The desire to improve the economic competitiveness and deployment pace of nuclear energy through modularization, manufacturing, and series production had led to the development of smaller size reactors. As the standard 17x17 fuel technology is mainly maintained in the pressurized water reactors (PWRs) category, this translates into a lower number of fuel assemblies in the core and sometimes a reduced fuel height. To assess the impact of such scale change in core design on fuel cycle cost and spent fuel volume, a scoping analysis tool is developed based on infinite lattice calculations, leakage, fuel management reduced models, and levelized unit cost of electricity (LCOE) estimate. As such, cost dynamics driven by fuel specific power, burnup, core leakage, feed, cycle length, fuel assembly height as well as uranium market data are captured with consistent set of assumptions and analysis methods. A selection of 5 reactor designs representative of leading PWR developers is assessed and compared. Pursuing higher specific powers and optimal burnups are highlighted as the main fuel cost reduction drivers, nevertheless, practical limitations and opportunities must be evaluated to establish the feasibility of such enhanced fuel operation. In consequence, a detailed core design is performed using SIMULATE3 code for 5 PWR variations including natural and forced coolant circulation modes, two reactor scales, power densities of 73, 112, and 123 kW/l and higher discharge burnups. Design and optimization are performed at the lattice level, for the reflector, and at the core loading level. Satisfactory steady-state operation including power distribution, coolant operating limits, and reactivity requirements are analyzed and reported in this paper. The fuel economics of the detailed core designs confirm the scoping analysis findings. Despite the unlocked power uprates in small PWRs, the achievable burnup for a given fuel specific power requires more enrichment and shorter fuel height results in higher fabrication costs per mass of fuel, which makes scaling down core size a more expensive endeavor on the fuel cycle front. Spent fuel volumes are reported for the PWRs designed in this paper. Furthermore, these volumes are driven by the core average discharge burnup regardless of the scale in consideration. Additional cost and core performance aspects related to heavy reflector gains, fuel-reflector substitution, and disposal cost policy in the U.S. are examined.

42 ENGINEERING↗

DOE OSTI2022

Advanced Ceramic Membranes/Modules for Ultra Efficient Hydrogen (H2) Production/Carbon Dioxide (CO2) Capture for Coal-Based Polygeneration Plants: Fabrication, Testing, and CFD Modeling

Inorganic membrane-based systems are a promising technology for precombustion CO2 capture with simultaneous H2 production. State-of-the-art packages for high temperature and pressure service consist of multiple tube membrane bundles prepared in a "candle filter" configuration, in which the membrane tubes are open at one end and sealed at the other. This configuration is used for practical reasons, specifically the need to minimize problems due to thermal expansion mismatch between the ceramic tube bundle and the steel housing. However, the primary technical problem with the candle filter format for commercial-scale installations is the inability to purge the tube side (typically the permeate side), a feature that is crucial for high H2 recovery. In this study, the focus is to design and fabricate the first dual-end open full ceramic multiple tube membrane bundle that enables tube side (permeate) purge for gas separation applications. An additional key feature of this design is the simplified module layout, as the membrane bundles can be installed end-to-end with tube side (permeate) flow directly from one bundle to the next. This layout simplifies the membrane to housing seals and yields significant improvement in membrane packing density. Detailed focus areas in our studies include: (i) Materials development and preparation of the tube-to-tube sheet potting for the dual-ended bundle; (ii) the sealing and optimal module configuration design to minimize membrane stress upon module mounting; (iii) the demonstration, via the fabrication of CMS and Pd-alloy membranes supported on full-size, dual-ended ceramic support bundles, of the first example of a purgeable ceramic membrane and module; and (iv) development of a CFD model of the membrane module for calculation of feed flow distribution, and for use in scale-up, and capital cost estimating. The CFD model was validated using experimental data with the multi-tubular membrane system, employing He/N2 as a model gas mixture (surrogate for H2/CO2), and has been shown to be quite accurate. Employing the model, we are able to study the effects of operating pressure and temperature, feed and sweep gas flow rates, and the choice of membrane tube configuration on system performance.

20 FOSSIL-FUELED POWER PLANTS↗

DOE OSTI2022

Advanced Ceramic Membranes/Modules for Ultra Efficient Hydrogen (H2) Production/Carbon Dioxide (CO2) Capture for Coal-Based Polygeneration Plants: Fabrication, Testing and CFD Modeling

20 FOSSIL-FUELED POWER PLANTS↗

DOE OSTI2022

Life Cycle Analysis and Life Cycle Costing for Municipal Solid Waste (MSW) Management and Materials Redeployment to Support ARPA-E’s Potential Program on Waste to Energy and Materials

Municipal solid waste infrastructure in the United States contains complex supply chains to transport, sort, and sequester waste. This project developed models to estimate the environmental and economic costs of waste infrastructure in the U.S. using data from several municipalities. Life cycle costing (LCC) models were developed for waste management relevant supply chain steps, including waste collection, sorting, landfilling, waste-to-energy, ash processing, and recycling. A waste-to-materials and energy (WTM&E) life cycle analysis (LCA) and LCC dashboard was developed in collaboration with Argonne National Laboratory (ANL) and an LCC tool was developed for additional analysis.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2020

Machine learning for nonintrusive model order reduction of the parametric inviscid transonic flow past an airfoil

Fluid flow in the transonic regime finds relevance in aerospace engineering, particularly in the design of commercial air transportation vehicles. Computational fluid dynamics models of transonic flow for aerospace applications are computationally expensive to solve because of the high degrees of freedom as well as the coupled nature of the conservation laws. While these issues pose a bottleneck for the use of such models in aerospace design, computational costs can be significantly minimized by constructing special, structure-preserving surrogate models called reduced-order models. In this work, we propose a machine learning method to construct reduced-order models via deep neural networks and we demonstrate its ability to preserve accuracy with a significantly lower computational cost. In addition, our machine learning methodology is physics-informed and constrained through the utilization of an interpretable encoding by way of proper orthogonal decomposition. Application to the inviscid transonic flow past the RAE2822 airfoil under varying freestream Mach numbers and angles of attack, as well as airfoil shape parameters with a deforming mesh, shows that the proposed approach adapts to high-dimensional parameter variation well. Notably, the proposed framework precludes the knowledge of numerical operators utilized in the data generation phase, thereby demonstrating its potential utility in the fast exploration of design space for diverse engineering applications. Comparison against a projection-based nonintrusive model order reduction method demonstrates that the proposed approach produces comparable accuracy and yet is orders of magnitude computationally cheap to evaluate, despite being agnostic to the physics of the problem.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

Developing Decommissioning Cost Estimating Relationships and the Decommissioning Estimate Analysis Library Estimating System - 20423

DOE-EM's Consolidated Business Center (EM-CBC) Office of Cost Estimating has developed a system of Decommissioning Cost Estimating Relationships (CERs) and Models to improve the reliability and efficiency of its estimates. The development builds on the multi-year process of collecting and organizing historical cost data from DOEEMs decommissioning projects and estimate model development to support project and program-level decommissioning estimates. The latest work has extended these models down to project scope element and component building area levels, allowing the actual costs of work for these estimate components to be assembled into estimates that reflect specific projects. The CERs have been incorporated into two estimator-friendly products - as cost libraries in the MII estimating program and as an MS Excel-based model, and are available for DOE federal staff use. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2021

Simple, low-cost and accurate data-driven geophysical forecasting with learned kernels

Modelling geophysical processes as low-dimensional dynamical systems and regressing their vector field from data is a promising approach for learning emulators of such systems. We show that when the kernel of these emulators is also learned from data (using kernel flows, a variant of cross-validation), then the resulting data-driven models are not only faster than equation-based models but are easier to train than neural networks such as the long short-term memory neural network. In addition, they are also more accurate and predictive than the latter. When trained on geophysical observational data, for example the weekly averaged global sea-surface temperature, considerable gains are also observed by the proposed technique in comparison with classical partial differential equation-based models in terms of forecast computational cost and accuracy. When trained on publicly available re-analysis data for the daily temperature of the North American continent, we observe significant improvements over classical baselines such as climatology and persistence-based forecast techniques. Although our experiments concern specific examples, the proposed approach is general, and our results support the viability of kernel methods (with learned kernels) for interpretable and computationally efficient geophysical forecasting for a large diversity of processes.

58 GEOSCIENCES↗