Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “TERMS”

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 721 records · Page 40
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Sensitivity Study of Mini-Batch Size on a Long Short-Term Memory Network for In-situ Sensing of Core-to-shell Ratio of Microencapsulated Phase Change Materials

Microencapsulated phase change materials are being studied for applications for thermal energy storage in concentrated solar fields. During fabrication, the thickness of the encapsulation cannot be readily measured for real-time control. Therefore, a machine learning network, specifically a Long Short-Term Memory network, is being developed to estimate the ratio of the shell radius to core radius based on a one second temperature history. The mini-batch size determines how often the algorithm weights are updated during network training, and shuffle indicates whether the training data is shuffled during training. A general factorial design is used to analyze the effects of varying mini-batch size and shuffle, along with the core-to-shell ratio, on the RMSE of the response from the Long Short-Term Memory network. It was found that the network performed better for smaller core to shell ratios (less than 0.6) and had the lowest RMSE when the minibatch size was 128. The minimum RMSE found was 0.00501.

Shannon, Rebecca↗
DOE OSTI2024

Beginner's Guide to Understanding Power System Model Results for Long-Term Resource Plans

This presentation summarizes a recently published beginner's guide for understanding power system model results for long-term resource plans. This guide is designed to improve decision-making in the electricity planning process by strengthening dialogue between system planners and relevant stakeholders. It focuses on interpreting scenario results generated by power system planning models. These results are often used to inform integrated resource plans, which are long-term plans for a utility to meet future electricity requirements. Whether you are a state utility commissioner, commission staff, part of a state energy office, an intervenor in the planning process, or another stakeholder, this guide is designed to help you better weigh in on the modeling methods or interpret the model results to improve decision-making. This guide can enhance your understanding to ask the right questions, which can enable better understanding of results and key assumptions, data, and methods.

ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2025

Analysis of outgassing from natural fibers VIP core to improve their long-term performance

Buildings are the largest consumers of energy accounting for 40% of the total energy use in the US. The use of vacuum insulation panels (VIPs) with large R/in. values offers unique benefits in reducing building energy consumption such as providing excellent thermal insulation with significantly less thickness compared to convention insulation materials. Natural fibers are promising VIP core material because of their low cost and offer good thermal performance. However, the long-term thermal performance of the VIPs with natural fiber core can be compromised because of off-gassing from natural fibers. In this study, we utilize a mass spectrometer based analytical technique to investigate the composition of off-gassing from natural fibers. The result can be used to determine the optimal getters to help maintain the vacuum and improve the long-term thermal performance of the VIP for building insulation.

Tamraparni, Achutha [ORNL] (ORCID:0000000299658544↗
DOE OSTI2023

Regional Medium-Term Hourly Electricity Demand Forecasting Based on LSTM

This paper aims to forecast high-resolution (hourly) aggregated load for a certain region in the medium term (a few days to over a year). One region is defined as some places with similar climate characteristics because the climate influences people's daily lifestyles and hence the electric usage. We decom- pose the electric usage records into two parts: base load and seasonal load. Considering both temperature and time factors, different deep learning methods are adopted to characterize them. The first goal of our approach is to predict the peak load which is critical for power system planning. Furthermore, our proposed forecast method can provide the depiction of the hourly load profile to provide customized load curves for high- level real-time applications. The proposed method is tested on real-world historical data collected by CAISO, BPA, and PACW. The experimental results show that trained by three years of data, our method could reduce the prediction error for one-year lead hourly load below 5% MAPE, and predict the occurrence of the peak load for next year in CAISO with an error within three days. Furthermore, as a byproduct, an interesting observation on the impact of COVID-19 on human life was made and discussed based on these case studies.

deep learning↗
DOE OSTI2023

Regional Medium-Term Hourly Electricity Demand Forecasting Based on LSTM

This paper aims to forecast high-resolution (hourly) aggregated load for a certain region in the medium term (a few days to over a year). One region is defined as some places with similar climate characteristics because the climate influences people's daily lifestyles and hence the electric usage. We decompose the electric usage records into two parts: base load and seasonal load. Considering both temperature and time factors, different deep-learning methods are adopted to characterize them. The first goal of our approach is to predict the peak load which is critical for power system planning. Furthermore, our proposed forecast method can provide the depiction of the hourly load profile to provide customized load curves for high-level real-time applications. The proposed method is tested on real-world historical data collected by CAISO, BPA, and PACW. The experimental results show that trained by three years of data, our method could reduce the prediction error for a one-year lead hourly load below $5\%$ MAPE, and predict the occurrence of the peak load for next year in CAISO with an error within three days. Furthermore, as a byproduct, an interesting observation on the impact of COVID-19 on human life was made and discussed based on these case studies.

deep learning↗
DOE OSTI2023

Regional Medium-Term Hourly Electricity Demand Forecasting Based on LSTM: Preprint

This paper aims to forecast high-resolution (hourly) aggregated load for a certain region in the medium term (a few days to over a year). One region is defined as some places with similar climate characteristics because the climate influences people's daily lifestyles and hence the electric usage. We decompose the electric usage records into two parts: base load and seasonal load. Considering both temperature and time factors, different deep-learning methods are adopted to characterize them. The first goal of our approach is to predict the peak load which is critical for power system planning. Furthermore, our proposed forecast method can provide the depiction of the hourly load profile to provide customized load curves for high-level real-time applications. The proposed method is tested on real-world historical data collected by CAISO, BPA, and PACW. The experimental results show that trained by three years of data, our method could reduce the prediction error for a one-year lead hourly load below 5% MAPE, and predict the occurrence of the peak load for next year in CAISO with an error within three days. Furthermore, as a byproduct, an interesting observation on the impact of COVID-19 on human life was made and discussed based on these case studies.

deep learning↗
DOE OSTI2021

Short-term nodal load forecasting based on machine learning techniques

This paper introduces an advanced Short-term Nodal Load Forecasting (STNLF) method that forecasts nodal load profiles for the next day in power systems, based on the combined use of three machine learning techniques. Least Absolute Shrinkage and Selection Operator (LASSO) is employed to reduce the number of features for a single nodal load forecasting. Principal Component Analysis (PCA) is used to capture the features of historical loads in low-dimensional space compared to the original high-dimensional load space where features are barely possible to depict. Additionally, Bayesian Ridge Regression (BRR) is utilized to decide the parameters of the prediction model from a statistics perspective. Tests based on modified PJM load data demonstrate the effectiveness of the proposed STNLF method compared to the state-of-the-art General Regression Neural Network (GRNN) method. Moreover, the reliability of the day-ahead Unit Commitment (UC) solution is shown to have been improved, based on the forecasted load data using the proposed STNLF method.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2020

Carbon budget of the Harvard Forest Long-Term Ecological Research site: pattern, process, and response to global change

How, where, and why carbon (C) moves into and out of an ecosystem through time are long-standing questions in biogeochemistry. Here, we bring together hundreds of thousands of C-cycle observations at the Harvard Forest in central Massachusetts, USA, a mid-latitude landscape dominated by 80–120-yr-old closed-canopy forests. These data answered four questions: (1) where and how much C is presently stored in dominant forest types; (2) what are current rates of C accrual and loss; (3) what biotic and abiotic factors contribute to variability in these rates; and (4) how has climate change affected the forest’s C cycle? Harvard Forest is an active C sink resulting from forest regrowth following land abandonment. Soil and tree biomass comprise nearly equal portions of existing C stocks. Net primary production (NPP) averaged 680–750 g C·m –2 ·yr –1 ; belowground NPP contributed 38–47% of the total, but with large uncertainty. Mineral soil C measured in the same inventory plots in 1992 and 2013 was too heterogeneous to detect change in soil-C pools; however, radiocarbon data suggest a small but persistent sink of 10–30 g C·m –2 ·yr –1 . Net ecosystem production (NEP) in hardwood stands averaged ~300 g C·m –2 ·yr –1 . NEP in hemlock-dominated forests averaged ~450 g C·m –2 ·yr –1 until infestation by the hemlock woolly adelgid turned these stands into a net C source. Since 2000, NPP has increased by 26%. For the period 1992–2015, NEP increased 93%. The increase in mean annual temperature and growing season length alone accounted for ~30% of the increase in productivity. Interannual variations in GPP and NEP were also correlated with increases in red oak biomass, forest leaf area, and canopy-scale light-use efficiency. Compared to long-term global change experiments at the Harvard Forest, the C sink in regrowing biomass equaled or exceeded C cycle modifications imposed by soil warming, N saturation, and hemlock removal. Furthermore, the results of this synthesis and comparison to simulation models suggest that forests across the region are likely to accrue C for decades to come but may be disrupted if the frequency or severity of biotic and abiotic disturbances increases.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Long-term measurements in a mixed-grass prairie reveal a change in soil organic carbon recalcitrance and its environmental sensitivity under warming

Soil respiration, the major pathway for ecosystem carbon (C) loss, has the potential to enter a positive feedback loop with the atmospheric CO2 due to climate warming. For reliable projections of climate-carbon feedbacks, accurate quantification of soil respiration and identification of mechanisms that control its variability are essential. Process-based models simulate soil respiration as functions of belowground C input, organic matter quality, and sensitivity to environmental conditions. However, evaluation and calibration of process-based models against the long-term in situ measurements are rare. Here, we evaluate the performance of the Terrestrial ECOsystem (TECO) model in simulating total and heterotrophic soil respiration measured during a 16-year warming experiment in a mixed-grass prairie; calibrate model parameters against these and other measurements collected during the experiment; and explore whether the mechanisms of C dynamics have changed over the years. Calibrating model parameters against observations of individual years substantially improved model performance in comparison to pre-calibration simulations, explaining 79–86% of variability in observed soil respiration. Interannual variation of the calibrated model parameters indicated increasing recalcitrance of soil C and changing environmental sensitivity of microbes. Overall, we found that (1) soil organic C became more recalcitrant in intact soil compared to root-free soil; (2) warming offset the effects of increasing C recalcitrance in intact soil and changed microbial sensitivity to moisture conditions. Furthermore, these findings indicate that soil respiration may decrease in the future due to C quality, but this decrease may be offset by warming-induced changes in C cycling mechanisms and their responses to moisture conditions.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Short-term warming increased soil heterotrophic respiration due to enhanced active microbial community

Soil microbes, particularly active microbes, play a crucial role in conserving soil carbon under climate change, especially in forest ecosystems, constituting over 50% of global soil organic carbon. Nevertheless, the response mechanisms of active microbial community to climate warming and their subsequent impacts on soil heterotrophic respiration (Rh) remain insufficiently understood. To resolve this mechanistic uncertainty, we implemented a 3-year soil translocation experiment to investigate soil warming effects on active microbial community and Rh. We used 16S rRNA gene amplicon sequencing, Metatranscriptomics sequencing, coupled with correlation analyses to explore the linkages between warming-induced shifts in Rh and the active microbial community. Our results demonstrated that warming of soil induced a 55% enhancement in Rh. Concurrently, the Shannon index and Richness of active microbial community increased by 20% and 117%, respectively. Warming significantly altered active microbial community composition, inducing a compositional shift characterized by 0.5-to 2-fold increases in the abundances of Proteobacteria, Chloroflexi, and Actinobacteria. Our study also revealed 92–100% increases in the abundance of C-degradation functional genes encoding starch, cellulose, and lignin decomposition pathways within active microbial community under warming. Statistical analyses identified significant positive correlations between Rh and the abundance of Proteobacteria and Actinobacteria, as well as expression levels of functional genes associated with lignin and cellulose decomposition pathways. Furthermore, our results suggested that short-term warming increased Rh through altering diversity, species composition, and C degradation functional genes of active microbial community providing insights into the influence of microbial communities on soil C-climate feedbacks under climate warming.

Active microbial community↗
DOE OSTI2024

Long-term effects of cycle time and volume exchange ratio on poly(3-hydroxybutyrate-co-3-hydroxyvalerate) production from food waste digestate by Haloferax mediterranei cultivated in sequencing batch reactors for 450 days

Food waste digestate was fed into a sequencing batch reactor (SBR) for Haloferax mediterranei (HM) to produce poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV). This SBR was operated uninterruptedly for 450 days to test its stability, during which the cycle time and volume exchange ratio were varied to understand their impacts on the PHBV fermentation performance under ranged organic loading rates (OLR). Results showed that 1) PHBV productivity was proportional to OLR of food waste digestate; 2) substrate and product inhibitions were two limiting factors constraining substrate utilization and PHBV yields; 3) PHBV titer was dependent on the hydraulic retention time of the SBR while a volume exchange ratio lower than 0.5 is unfavorable due to the product inhibitor accumulation. Furthermore, this study for the first time demonstrated that the long-term stability of food waste-fed PHBV production by HM and revealed that inhibition effects could be barriers in SBR limiting the full-scale application of the technology.

09 BIOMASS FUELS↗
DOE OSTI2021

Bayesian inference of heterogeneous epidemic models: Application to COVID-19 spread accounting for long-term care facilities

Here we propose a high dimensional Bayesian inference framework for learning heterogeneous dynamics of a COVID-19 model, with a specific application to the dynamics and severity of COVID-19 inside and outside long-term care (LTC) facilities. We develop a heterogeneous compartmental model that accounts for the heterogeneity of the time-varying spread and severity of COVID-19 inside and outside LTC facilities, which is characterized by time-dependent stochastic processes and time-independent parameters in ~ 1500 dimensions after discretization. To infer these parameters, we use reported data on the number of confirmed, hospitalized, and deceased cases with suitable post-processing in both a deterministic inversion approach with appropriate regularization as a first step, followed by Bayesian inversion with proper prior distributions. To address the curse of dimensionality and the ill-posedness of the high-dimensional inference problem, we propose use of a dimension-independent projected Stein variational gradient descent method, and demonstrate the intrinsic low-dimensionality of the inverse problem. We present inference results with quantified uncertainties for both New Jersey and Texas, which experienced different epidemic phases and patterns. Moreover, we also present forecasting and validation results based on the empirical posterior samples of our inference for the future trajectory of COVID-19.

42 ENGINEERING↗
DOE OSTI2025

HydroEcoLSTM: A Python package with graphical user interface for hydro-ecological modeling with long short-term memory neural network

Machine learning (ML) is emerging as a promising tool for modeling hydro-ecological processes due to the increasing availability of large environmental data. However, the use of ML requires sufficient programming knowledge due to a lack of a graphical user interface (GUI). In this study, we introduced a GUI package, named HydroEcoLSTM, with the long short-term memory network (LSTM) as the core model, that allows non-ML experts to utilize their domain knowledge to construct complex ML models. We demonstrated the functionalities of HydroEcoLSTM with two practical examples, including (1) predictions of streamflow in both gauged and ungauged catchments and (2) predictions of multiple outputs (i.e., streamflow and isotope transport from two catchments). The simulation results obtained in both case experiments are satisfactory. In the first example, the average Nash–Sutcliffe Efficiency (NSE) for streamflow simulation during the testing period is 0.79 while the application of the trained model in two assumed ungauged catchments also achieves the average NSE of 0.68. In the second example, the average NSE for streamflow and instream isotope simulation during the testing period is 0.71. Ultimately, applications of HydroEcoLSTM with real-world examples demonstrate its potential use for practical applications and research without requiring extensive coding skills.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Optimal operation of solid-oxide electrolysis cells considering long-term chemical degradation

Optimizing the performance of solid oxide electrolysis cells (SOECs) for long-term hydrogen (H 2 ) production at high temperatures is crucial, as prolonged operation leads to efficiency losses and shorter cell lifespans due to chemical degradation. Here, in this work, we adopt a quasi-steady state approach for dynamic optimization over extended operational periods to address the disparity in timescales between cell operation and degradation. Integrating a 2-D non-isothermal SOEC model with balance-of-plant (BOP) equipment, we explore three optimization objectives: minimizing terminal degradation, maximizing integral efficiency, and minimizing the levelized cost of H 2 (LCOH). Our dynamic optimization algorithm reduces LCOH by 9.5% and 16% compared to strategies focusing solely on terminal degradation and integral efficiency, respectively. For electricity prices of 0.03 $\$$/mWh and 0.3 $\$$ mWh optimal replacement schedules range from 5 to 2 years, depending on the operational mode. Furthermore, a flexible operational mode yields additional improvements in LCOH over traditional galvanostatic and potentiostatic modes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Inductive predictions of hydrologic events using a Long Short-Term Memory network and the Soil and Water Assessment Tool

We present machine learning methods to predict hydrologic features such as streamflow and soil moisture from spatially and temporally varying hydrological and meteorological data. Here, we used a temporal reduction technique to reduce computation and memory requirements and trained a Long Short-Term Memory (LSTM) network to predict soil moisture and streamflow over multiple watersheds. We show LSTM networks can be trained in a fraction of the time required by complex process-based and attention-based models such as Soil and Water Assessment Tool (SWAT) and GeoMAN without sacrificing accuracy. We also demonstrate that outside data - sourced from a watershed other than the target - can be used to train LSTM to comparable or even superior prediction accuracy. The success of LSTM in such spatially-inductive settings shows hydrologic features can be predicted with minimal prior knowledge of the watershed in question. Finally, we make all methodologies of this work publicly available as an end-to-end software pipeline that facilitates rapid prototyping of hydrologic learners.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Experimental workflow to estimate model parameters for evaluating long term viscoelastic response of CO2 storage caprocks

Understanding the time-dependent behavior of reservoir and sealing formations is critical to assessing risks associated with geological carbon storage since time-dependent deformation strongly influences mechanical responses of some rock types. Many studies have evaluated the risk of CO2 leakage and induced seismicity by assuming poroelastic rheology in sealing formations. Few have considered viscoelastic or other time-dependent responses, where the existing literature adopts 1D models to represent long-term time-dependent responses. This is primarily because to date, the general form of a reasonable 3D time-dependent model for rocks remains unclear. In this paper, we address this unclear issue by proposing a new workflow to select constitutive modeling parameters to evaluate if a 3D viscoelastic model is reasonable using several-hour-long experimental data and a power-law response to extrapolate to the decades-long time frames of interest in geologic carbon storage. To provide experimental data, we conducted multi-level loading/unloading triaxial relaxation tests with four rock types. The experimental results showed that the maximum load relaxation observed is approximately 49%, with some rock types showing as little as 1.4%. Using a simple linear viscoelastic model, parameters were chosen such that a maximum deviation of 1.5 MPa in axial stress and 7 MPa in radial stress was attained with the extrapolated 30-year data. We found that a reasonable parameter range for the normalized elastic modulus is 0.1~2 for rocks with significant time-dependent responses and 0.01~0.06 for those with small time-dependent responses. No matter how significant time-dependent responses are for rocks considered, our results showed that the relaxation time has a general range of 1~10^10 s, whose time scale can be one or two orders higher than a time frame typically envisioned for CO2 injection projects.

Stress relaxation, 3D time-dependent model, viscoe↗
DOE OSTI2020

A long short-term memory embedding for hybrid uplifted reduced order models

In this paper, we introduce an uplifted reduced order modeling (UROM) approach through the integration of standard projection based methods with long short-term memory (LSTM) embedding. Our approach has three modeling layers or components. In the first layer, we utilize an intrusive projection approach to model dynamics represented by the largest modes. The second layer consists of an LSTM model to account for residuals beyond this truncation. This closure layer refers to the process of including the residual effect of the discarded modes into the dynamics of the largest scales. However, the feasibility of generating a low rank approximation tails off for higher Kolmogorov n -width systems due to the underlying nonlinear processes. The third uplifting layer, called super-resolution, addresses this limited representation issue by expanding the span into a larger number of modes utilizing the versatility of LSTM. Therefore, our model integrates a physics-based projection model with a memory embedded LSTM closure and an LSTM based super-resolution model. In several applications, we exploit the use of Grassmann manifold to construct UROM for unseen conditions. We performed numerical experiments by using the Burgers and Navier-Stokes equations with quadratic nonlinearity. Finally, our results show robustness of the proposed approach in building reduced order models for parameterized systems and confirm the improved trade-off between accuracy and efficiency.

42 ENGINEERING↗
DOE OSTI2020

Responses of gaseous sulfuric acid and particulate sulfate to reduced SO 2 concentration: A perspective from long-term measurements in Beijing

SO 2 concentration decreased rapidly in recent years in China due to the implementation of strict control policies by the government. Particulate sulfate ( p S O 4 2 - ) and gaseous H 2 SO 4 (SA) are two major products of SO 2 and they play important roles in the haze formation and new particle formation (NPF), respectively. We examined the change in p S O 4 2 - and SA concentrations in response to reduced SO 2 concentration using long-term measurement data in Beijing. Simulations from the Community Multiscale Air Quality model with a 2-D Volatility Basis Set (CMAQ/2D-VBS) were used for comparison. From 2013 to 2018, SO 2 concentration in Beijing decreased by ~81% (from 9.1 ppb to 1.7 ppb). p S O 4 2 - concentration in submicrometer particles decreased by ~60% from 2012–2013 (monthly average of ~10 μg·m - 3 ) to 2018–2019 (monthly average of ~4 μg·m - 3 ). Accordingly, the fraction of p S O 4 2 - in these particles decreased from 20–30% to <10%. Increased sulfur oxidation ratio was observed both in the measurements and the CMAQ/2D-VBS simulations. Despite the reduction in SO 2 concentration, there was no obvious decrease in SA concentration based on data from several measuring periods from 2008 to 2019. This was supported by the increased SA:SO 2 ratio with reduced SO 2 concentration and condensation sink. NPF frequency in Beijing between 2004 and 2019 remains relatively constant. Finally, this constant NPF frequency is consistent with the relatively stable SA concentration in Beijing, while different from some other cities where NPF frequency was reported to decrease with decreased SO 2 concentrations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗