Search NASA⌕ Search

SEARCH · Search NASA

Results for “Integrated mean squared error”

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 109 records · Page 6

Active Learning of Microgrid Frequency Dynamics Using Neural Ordinary Differential Equations

Accurate frequency modelling of inverter‐based resource (IBR)‐dominated power systems is crucial for ensuring stable, reliable and resilient operations, particularly given their inherent low‐inertia characteristics and fast dynamics that traditional swing equation‐based models inadequately capture. This paper explores neural ordinary differential equations (Neural ODEs) as a computationally efficient, data‐driven framework for modelling power system frequency dynamics, specifically within microgrids integrating high penetrations of distributed energy resources (DERs). The developed neural ODEs framework incorporates a neural network architecture designed to capture input dynamics. By actively perturbing the system with a known signal, the Python‐based neural ODEs framework was trained using measured system states and inputs, without the need for detailed system information. The framework, tested on a model of the Cordova, AK, microgrid, achieved a goodness of fit ranging from 60% to 99% across different state variables and maintained a mean square error in the 10 -6 p.u. range under square and step excitation signals. The proposed approach demonstrated robustness to measurement noise and initial condition variations while maintaining low computational complexity suitable for real‐time power system control applications. Furthermore, transfer learning enabled the neural ODEs model to adapt to the following changes in system topology or generator dispatch, highlighting its effectiveness for dynamic microgrids with frequently evolving configurations and diverse DERs.

Aryal, Tara [South Dakota State Univ., Brookings, ↗

Energy Performance Assessment of Radiant Cooling System through Modeling and Calibration at Component Level

The paper describes a case study of an information technology office building with a radiant cooling system and a conventional variable air volume (VAV) system installed side by side so that performancecan be compared. First, a 3D model of the building involving architecture, occupancy, and HVAC operation was developed in EnergyPlus, a simulation tool. Second, a different calibration methodology was applied to develop the base case for assessing the energy saving potential. This paper details the calibration of the whole building energy model to the component level, including lighting, equipment, and HVAC components such as chillers, pumps, cooling towers, fans, etc. Also a new methodology for the systematic selection of influence parameter has been developed for the calibration of a simulated model which requires large time for the execution. The error at the whole building level [measured in mean bias error (MBE)] is 0.2%, and the coefficient of variation of root mean square error (CvRMSE) is 3.2%. The total errors in HVAC at the hourly are MBE = 8.7% and CvRMSE = 23.9%, which meet the criteria of ASHRAE 14 (2002) for hourly calibration. Different suggestions have been pointed out to generalize the energy saving of radiant cooling system through the existing building system. So a base case model was developed by using the calibrated model for quantifying the energy saving potential of the radiant cooling system. It was found that a base case radiant cooling system integrated with DOAS can save 28% energy compared with the conventional VAV system.

Building Energy Simulation↗

Navigating the Noise: Bringing Clarity to ML Parameterization Design With O $\boldsymbol{\mathcal{O}}$(100) Ensembles

Abstract Machine‐learning (ML) parameterizations of subgrid processes (here of turbulence, convection, and radiation) may one day replace conventional parameterizations by emulating high‐resolution physics without the cost of explicit simulation. However, uncertainty about the relationship between offline and online performance (i.e., when integrated with a large‐scale general circulation model) hinders their development. Much of this uncertainty stems from limited sampling of the noisy, emergent effects of upstream ML design decisions on downstream online hybrid simulation. Our work rectifies the sampling issue via the construction of a semi‐automated, end‐to‐end pipeline for size ensembles of hybrid simulations, revealing important nuances in how systematic reductions in offline error manifest in changes to online error and online stability. For example, removing dropout and switching from a Mean Squared Error to a Mean Absolute Error loss both reduce offline error, but they have opposite effects on online error and online stability. Other design decisions, like incorporating memory, converting moisture input from specific humidity to relative humidity, using batch normalization, and training on multiple climates do not come with any such compromises. Finally, we show that ensemble sizes of may be necessary to reliably detect causally relevant differences online. By enabling rapid online experimentation at scale, we can empirically settle debates regarding subgrid ML parameterization design that would have otherwise remained unresolved in the noise.

Lin, Jerry [Department of Earth System Sciences Un↗

Digital Twin-Enabled Adaptive Control for Hydroelectric Systems: Turbine Governor and Voltage Regulation

This paper presents a comprehensive digital twin (DT) framework for hydroelectric systems that enables adaptive control of turbine governors and excitation systems without requiring detailed manufacturer specifications. The proposed framework integrates neural network-based system identification with stabilizing adaptive control laws for the installed turbine controller and middle-branch adaptive tuning for the installed voltage regulator. Using real operational data from Unit C-8 at Rocky Reach Dam (1,349 MW capacity), highfidelity neural network models are developed to capture turbine and generator dynamics without requiring detailed manufacturer specifications. The DT enables safe controller synthesis and validation in simulation before deployment. For turbine control, the proposed method achieves a 79.9% mean square error (MSE) reduction compared with that of an optimal controller. For voltage regulation, the adaptive excitation controller achieves approximately 42.6% MSE reduction while preserving installed protection logic. The results demonstrate that DT technology provides a practical pathway for modernizing hydropower control systems with minimal operational disruption.

Gui, Yonghao [ORNL] (ORCID:0000000250435534)↗

Prediction of stability constants of metal–ligand complexes by machine learning for the design of ligands with optimal metal ion selectivity

The new LOGKPREDICT program integrates HostDesigner molecular design software with the machine learning (ML) program Chemprop. By supplying HostDesigner with predicted log K values, LOGKPREDICT enhances the computer-aided molecular design process by ranking ligands directly by metal–ligand binding strength. Harnessing reliable experimental data from a historic National Institute of Standards and Technology (NIST) database and data from the International Union of Pure and Applied Chemistry (IUPAC), we train message passing neural net algorithms. The multi-metal NIST-based ML model has a root mean square error (RMSE) of 0.629 ± 0.044 (R 2 of 0.960 ± 0.006), while two versions of lanthanide-only IUPAC-based ML models have, respectively, RMSE of 0.764 ± 0.073 (R 2 of 0.976 ± 0.005) and 0.757 ± 0.071 (R 2 of 0.959 ± 0.007). For relative log K predictions on an out-of-sample set of six ligands, demonstrating metal ion selectivity, the RMSE value reaches a commendably low 0.25. Here we showcase the use of LOGKPREDICT in identifying ligands with high selectivity for lanthanides in aqueous solutions, a finding supported by recent experimental evidence. We also predict new ligands yet to be verified experimentally. Therefore, our ML models implemented through LOGKPREDICT and interfaced with the ligand design software HostDesigner pave the way for designing new ligands with predetermined selectivity for competing metal ions in an aqueous solution.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Patient-bounded extrapolation using low-dose priors for volume-of-interest imaging in C-arm CT

Purpose: Three-dimensional (3D) volume-of-interest (VOI) imaging with C-arm systems provides anatomical information in a predefined 3D target region at a considerably low x-ray dose. However, VOI imaging involves laterally truncated projections from which conventional reconstruction algorithms generally yield images with severe truncation artifacts. Heuristic based extrapolation methods, e.g., water cylinder extrapolation, typically rely on techniques that complete the truncated data by means of a continuity assumption and thus appear to be ad-hoc. It is our goal to improve the image quality of VOI imaging by exploiting existing patient-specific prior information in the workflow. Methods: A necessary initial step prior to a 3D acquisition is to isocenter the patient with respect to the target to be scanned. To this end, low-dose fluoroscopic x-ray acquisitions are usually applied from anterior–posterior (AP) and medio-lateral (ML) views. Based on this, the patient is isocentered by repositioning the table. In this work, we present a patient-bounded extrapolation method that makes use of these noncollimated fluoroscopic images to improve image quality in 3D VOI reconstruction. The algorithm first extracts the 2D patient contours from the noncollimated AP and ML fluoroscopic images. These 2D contours are then combined to estimate a volumetric model of the patient. Forward-projecting the shape of the model at the eventually acquired C-arm rotation views gives the patient boundary information in the projection domain. In this manner, we are in the position to substantially improve image quality by enforcing the extrapolated line profiles to end at the known patient boundaries, derived from the 3D shape model estimate. Results: The proposed method was evaluated on eight clinical datasets with different degrees of truncation. The proposed algorithm achieved a relative root mean square error (rRMSE) of about 1.0% with respect to the reference reconstruction on nontruncated data, even in the presence of severe truncation, compared to a rRMSE of 8.0% when applying a state-of-the-art heuristic extrapolation technique. Conclusions: The method we proposed in this paper leads to a major improvement in image quality for 3D C-arm based VOI imaging. It involves no additional radiation when using fluoroscopic images that are acquired during the patient isocentering process. The model estimation can be readily integrated into the existing interventional workflow without additional hardware.

60 APPLIED LIFE SCIENCES↗

Z-checker: A framework for assessing lossy compression of scientific data

Because of the vast volume of data being produced by today’s scientific simulations and experiments, lossy data compressor allowing user-controlled loss of accuracy during the compression is a relevant solution for significantly reducing the data size. However, lossy compressor developers and users are missing a tool to explore the features of scientific data sets and understand the data alteration after compression in a systematic and reliable way. To address this gap, we have designed and implemented a generic framework called Z-checker. On the one hand, Z-checker combines a battery of data analysis components for data compression. On the other hand, Z-checker is implemented as an open-source community tool to which users and developers can contribute and add new analysis components based on their additional analysis demands. In this article, we present a survey of existing lossy compressors. Then, we describe the design framework of Z-checker, in which we integrated evaluation metrics proposed in prior work as well as other analysis tools. Specifically, for lossy compressor developers, Z-checker can be used to characterize critical properties (such as entropy, distribution, power spectrum, principal component analysis, and autocorrelation) of any data set to improve compression strategies. For lossy compression users, Z-checker can detect the compression quality (compression ratio and bit rate) and provide various global distortion analysis comparing the original data with the decompressed data (peak signal-to-noise ratio, normalized mean squared error, rate–distortion, rate-compression error, spectral, distribution, and derivatives) and statistical analysis of the compression error (maximum, minimum, and average error; autocorrelation; and distribution of errors). Z-checker can perform the analysis with either coarse granularity (throughout the whole data set) or fine granularity (by user-defined blocks), such that the users and developers can select the best fit, adaptive compressors for different parts of the data set. Z-checker features a visualization interface displaying all analysis results in addition to some basic views of the data sets such as time series. To the best of our knowledge, Z-checker is the first tool designed to assess lossy compression comprehensively for scientific data sets.

97 MATHEMATICS AND COMPUTING↗

Quantification of Physical and Numerical Mixing in a Coastal Ocean Model Using Salinity Variance Budgets

Numerical mixing, the spurious mixing primarily generated by the discretization of advection, is often significant in estuarine and coastal models due to sharp, energetic fronts. In this study, we compare on- and offline estimates of numerical mixing in a submesoscale-resolving realistic simulation of the ocean state over the Texas-Louisiana continental shelf. While offline estimates of numerical mixing differ from online estimates, offline methods may be the only analysis available. This study offers insight into the differences between the on- and offline methods. We use two methods to estimate numerical mixing offline, based on salinity squared $s^2$ and volume-mean salinity variance $s^{\prime^2}$. Numerical mixing estimated from the $s^{\prime^2}$ budget is generally within 60\% of the magnitude for the online method but captures the temporal variability well. However, the $s^2$ budget compares poorly due to larger truncation errors associated with the tendency and advection terms, which can be reduced by increasing the model output frequency. We also investigate the effects of horizontal resolution on numerical mixing using a two-way nested grid. The volume-integrated numerical mixing constitutes 57\% of the bulk physical mixing -- the mixing prescribed by the turbulence closure scheme -- in the coarse model and may exceed the physical mixing by half an order of magnitude. We find that numerical mixing is reduced by 35\% on average in the nested model, less than expected based on scaling of the numerical mixing for an upwind advection scheme, likely due to new dynamical processes that emerge in the nested simulation.

54 ENVIRONMENTAL SCIENCES↗

A dynamic solvent chamber propagation estimation framework using RNN for warm solvent injection in heterogeneous reservoirs

Warm solvent injection (WSI), injecting low-temperature solvent into formations to reduce the viscosity of heavy oil, is a clean technology for heavy oil production through reducing greenhouse gas emissions and water usage. The success of WSI operation depends on the uniform development and propagation of solvent chambers in reservoirs. However, reservoir heterogeneity stemming from shale barriers plays a detrimental role in the conformance of solvent chamber development and oil production rate. In this work, we developed a novel recurrent neural network (RNN)-based framework with the capability of efficiently tracking and estimating the solvent chamber positions in heterogeneous reservoirs based on only production time-series data. The developed estimation model utilizes the “sequence-to-sequence" mapping methodology to correlate observed production time-series sequence and solvent chamber edge sequence via a long short-term memory (LSTM) algorithm. The trained RNN models exhibit high accuracy, evidenced by the predicted dynamic solvent chamber locations match the corresponding true locations from numerical simulation, with a high coefficient of determination (R 2 ) and a low mean squared error. Specifically, the achieved R 2 values exceed 0.98 on both the training and testing data. The developed RNN-based workflow was tested via several cases from both regularly- and irregularly-shaped shale barriers, and the results were promising. The predicted solvent chambers showed strong agreement with those obtained from numerical simulations. The major benefits of this workflow include reducing computational time and saving overall monitoring and tracking costs for conventional techniques. In conclusion, the present work would provide a good demonstration of the capability of practical integration of machine learning methods in solving engineering problems.

58 GEOSCIENCES↗

A machine learning approach for clinker quality prediction and nonlinear model predictive control design for a rotary cement kiln

Abstract Cement manufacturing is energy‐intensive (5Gj/t) and comprises a significant portion of the energy footprint of concrete systems. Incorporating modern monitoring, simulation and control systems will allow lower energy use, lower environmental impact, and lower costs of this widely used construction material. One of the goals of the CESMII roadmap project on the Smart Manufacturing of Cement included developing an analytical process model for clinker quality that includes the chemistry of the kiln feed and accounts for critical process variables. This predictive model will be used in nonlinear model predictive control system designed to significantly reduce process energy use while maintaining or improving product quality. In the cement manufacturing plant used in this study, the kiln feed (meal) is tested every 12 h and used to estimate the mineral composition of the cement kiln output (clinker) using the stoichiometry‐based Bogue's model and the expertise of the plant operators. During kiln operation, kiln output (clinker) is sampled and tested every 2 h to measure its chemical and mineral composition. The predicted and measured values of the clinker composition are used by the plant operators to adjust the kiln input stream and the production process characteristics to maintain stable operation and uniform product quality. However, the time delay between prediction and testing, along with inaccuracies inherent in the Bogue's model have made any process changes designed to minimize energy use problematic, especially in‐light of potential clinker quality issues that process changes often pose. A new analytical model that integrates quality information and process operation information has been developed from data collected from 2 years of production from an operating cement facility. To make the model fuel‐type‐independent, consumed heat energy was computed in the model instead of fuel type and amount. A Feedforward Network was trained and tailored from collected data. Many data‐based simulations were conducted to quantitatively evaluate the proposed model and the 5‐fold cross‐validation procedure was used to test the models. The resulting predictive model was shown to have a low root mean square error (MSE) with respect to the estimated clinker mineral composition compared to that using the industry standard “Bogue’ model”. The end goal of this work was to develop a single machine learning tool that allows the use of quality control data and process control variables to improve energy efficiency of the process in a continuous fashion. The proposed nonlinear model predictive control system (NMPC) can generate predicted kiln production characteristics based on manipulated variables in manner that accurately follows the target product quality values. Simulation results also show that the proposed model produced accurate predictions of kiln outputs that fell within the required constraints, while manipulating control variables within typical operational ranges.

Ali, Asem M.↗

QRF4P-NRT: Probabilistic Post-Processing of Near-Real-Time Satellite Precipitation Estimates Using Quantile Regression Forests

Accurate and reliable near-real-time satellite precipitation estimation is of great importance for operational large-scale flood forecasting and drought monitoring. The state-of-the-art precipitation post-processing model is based on a deterministic approach to construct relationships between satellites estimates and ground observations. We propose a probabilistic postprocessor, the Probabilistic Post-Processing of Near-Real-Time Satellite Precipitation Estimates using Quantile Regression Forests (QRF4P-NRT), based on quantile modeling, yielding both deterministic and probabilistic predictions. The experimental design incorporates different solutions of near-real-time predictors to further improve the model performance. Using the Integrated Multi-satellitE Retrievals Early Run for Global Precipitation Measurement Mission (IMERG-E) product as an example, we illustrate that the proposed method significantly improves the overall quality of the raw IMERG-E and is also superior to the bias-corrected product (IMERG Final Run, IMERG-F) at daily scale in a complex mountain basin. Evaluations of the corrected IMERG-E, raw IMERG-E, and IMERG-F using ground observation show that the corrected IMERG-E improves correlation coefficients (0.7), mean error (-0.14 mm/day) and root mean square error (3.3 mm/day) relative to the raw IMERG-E (0.31, -0.72 and 5.5 mm/day) and IMERG-F (0.34, -0.09 and 6.0 mm/day). The error decomposition further confirms that the QRF4P-NRT improves on the various deficiencies of the raw IMERG-E product. The ensemble assessment also demonstrates that the quantile outputs provide reliable prediction spread and sharp prediction intervals. The promising results indicate the great potential of the proposed method for probabilistic post-processing for near-real-time satellite precipitation estimates, and for further applications such as hydrological ensemble forecasting.

54 ENVIRONMENTAL SCIENCES↗

Temporal Forecasting of Distributed Temperature Sensing in a Thermal Hydraulic System With Machine Learning and Statistical Models

We benchmark performance of long-short term memory (LSTM) network machine learning model and autoregressive integrated moving average (ARIMA) statistical model in temporal forecasting of distributed temperature sensing (DTS). Data in this study consists of fluid temperature transient measured with two co-located Rayleigh scattering fiber optic sensors (FOS) in a forced convection mixing zone of a thermal tee. We treat each gauge of a FOS as an independent temperature sensor. We first study prediction of DTS time series using Vanilla LSTM and ARIMA models trained on prior history of the same FOS that is used for testing. The results yield maximum absolute percentage error (MaxAPE) and root mean squared percentage error (RMSPE) of 1.58% and 0.06% for ARIMA, and 3.14% and 0.44% for LSTM, respectively. Next, we investigate zero-shot forecasting (ZSF) with LSTM and ARIMA trained on history of the co-located FOS only, which is advantageous when limited training data is available. The ZSF MaxAPE and RMSPE values for ARIMA are comparable to those of the Vanilla use case, while the error values for LSTM increase. We show that in ZSF, performance of LSTM network can be improved by training on most correlated gauges between the two FOS, which are identified by calculating the Pearson correlation coefficient. The improved ZSF MaxAPE and RMSPE for LSTM are 4.4% and 0.33%, respectively. Performance of ZSF LSTM can be further enhanced through transfer learning (TL), where LSTM is re-trained on a subset of the FOS that is the target of forecasting. We show that LSTM pre-trained on correlated dataset and re-trained on 30% of testing target dataset achieves MaxAPE and RMSPE values of 2.32% and 0.28%, respectively.

ARIMA↗

WE-AB-303-08: Direct Lung Tumor Tracking Using Short Imaging Arcs

Purpose: Most current tumor tracking technologies rely on implanted markers, which suffer from potential toxicity of marker placement and mis-targeting due to marker migration. Several markerless tracking methods have been proposed: these are either indirect methods or have difficulties tracking lung tumors in most clinical cases due to overlapping anatomies in 2D projection images. We propose a direct lung tumor tracking algorithm robust to overlapping anatomies using short imaging arcs. Methods: The proposed algorithm tracks the tumor based on kV projections acquired within the latest six-degree imaging arc. To account for respiratory motion, an external motion surrogate is used to select projections of the same phase within the latest arc. For each arc, the pre-treatment 4D cone-beam CT (CBCT) with tumor contours are used to estimate and remove the contribution to the integral attenuation from surrounding anatomies. The position of the tumor model extracted from 4D CBCT of the same phase is then optimized to match the processed projections using the conjugate gradient method. The algorithm was retrospectively validated on two kV scans of a lung cancer patient with implanted fiducial markers. This patient was selected as the tumor is attached to the mediastinum, representing a challenging case for markerless tracking methods. The tracking results were converted to expected marker positions and compared with marker trajectories obtained via direct marker segmentation (ground truth). Results: The root-mean-squared-errors of tracking were 0.8 mm and 0.9 mm in the superior-inferior direction for the two scans. Tracking error was found to be below 2 and 3 mm for 90% and 98% of the time, respectively. Conclusions: A direct lung tumor tracking algorithm robust to overlapping anatomies was proposed and validated on two scans of a lung cancer patient. Sub-millimeter tracking accuracy was observed, indicating the potential of this algorithm for real-time guidance applications.

60 APPLIED LIFE SCIENCES↗

Virtual sensing-enabled digital twin framework for real-time monitoring of nuclear systems leveraging deep neural operators

Abstract Real-time monitoring is a foundation of nuclear digital twin technology, crucial for detecting material degradation and maintaining nuclear system integrity. Traditional physical sensor systems face limitations, particularly in measuring critical parameters in hard-to-reach or harsh environments, often resulting in incomplete data coverage. Machine learning-driven virtual sensors offer a transformative solution by complementing physical sensors in monitoring critical degradation indicators. This paper introduces the use of Deep Operator Networks (DeepONet) to predict key thermal-hydraulic parameters in the hot leg of pressurized water reactor. DeepONet acts as a virtual sensor, mapping operational inputs to spatially distributed system behaviors without requiring frequent retraining. Our results show that DeepONet achieves low mean squared and Relative L2 error, making predictions 1400 times faster than traditional CFD simulations . These characteristics enable DeepONet to function as a real-time virtual sensor, synchronizing with the physical system to track degradation conditions and provide insights within the digital twin framework for nuclear systems.

Hossain, Raisa↗

SU-F-J-10: Sliding Mode Control of a SMA Actuated Active Flexible Needle for Medical Procedures

Purpose: In medical interventional procedures such as brachytherapy, ablative therapies and biopsy precise steering and accurate placement of needles are very important for anatomical obstacle avoidance and accurate targeting. This study presents the efficacy of a sliding mode controller for Shape Memory Alloy (SMA) actuated flexible needle for medical procedures. Methods: Second order system dynamics of the SMA actuated active flexible needle was used for deriving the sliding mode control equations. Both proportional-integral-derivative (PID) and adaptive PID sliding mode control (APIDSMC) algorithms were developed and implemented. The flexible needle was attached at the end of a 6 DOF robotic system. Through LabView programming environment, the control commands were generated using the PID and APIDSMC algorithms. Experiments with artificial tissue mimicking phantom were performed to evaluate the performance of the controller. The actual needle tip position was obtained using an electromagnetic (EM) tracking sensor (Aurora, NDI, waterloo, Canada) at a sampling period of 1ms. During experiment, external disturbances were created applying force and thermal shock to investigate the robustness of the controllers. Results: The root mean square error (RMSE) values for APIDSMC and PID controllers were 0.75 mm and 0.92 mm, respectively, for sinusoidal reference input. In the presence of external disturbances, the APIDSMC controller showed much smoother and less overshooting response compared to that of the PID controller. Conclusion: Performance of the APIDSMC was superior to the PID controller. The APIDSMC was proved to be more effective controller in compensating the SMA uncertainties and external disturbances with clinically acceptable thresholds.

60 APPLIED LIFE SCIENCES↗

BESS-STAIR: a framework to estimate daily, 30m, and all-weather crop evapotranspiration using multi-source satellite data for the US Corn Belt

Abstract. With increasing crop water demands and drought threats, mapping andmonitoring of cropland evapotranspiration (ET) at high spatial and temporalresolutions become increasingly critical for water management andsustainability. However, estimating ET from satellites for precise waterresource management is still challenging due to the limitations in bothexisting ET models and satellite input data. Specifically, the process of ETis complex and difficult to model, and existing satellite remote-sensing datacould not fulfill high resolutions in both space and time. To address theabove two issues, this study presents a new high spatiotemporal resolution ETmapping framework, i.e., BESS-STAIR, which integrates a satellite-drivenwater–carbon–energy coupled biophysical model, BESS (Breathing Earth SystemSimulator), with a generic and fully automated fusion algorithm, STAIR(SaTallite dAta IntegRation). In this framework, STAIR provides daily 30'mmultispectral surface reflectance by fusing Landsat and MODIS satellite datato derive a fine-resolution leaf area index and visible/near-infrared albedo,all of which, along with coarse-resolution meteorological and CO 2 data, are used to drive BESS to estimate gap-free 30 m resolution daily ET.We applied BESS-STAIR from 2000 through 2017 in six areas across the US CornBelt and validated BESS-STAIR ET estimations using flux-tower measurementsover 12 sites (85 site years). Results showed that BESS-STAIR daily ETachieved an overall R2=0.75, with root mean square error RMSE=0.93 mm d -1 and relative error RE =27.9 % when benchmarkedwith the flux measurements. In addition, BESS-STAIR ET estimations capturedthe spatial patterns, seasonal cycles, and interannual dynamics well indifferent sub-regions. The high performance of the BESS-STAIR frameworkprimarily resulted from (1) the implementation of coupled constraints onwater, carbon, and energy in BESS, (2) high-quality daily 30 m data from theSTAIR fusion algorithm, and (3) BESS's applicability under all-skyconditions. BESS-STAIR is calibration-free and has great potentials to be areliable tool for water resource management and precision agricultureapplications for the US Corn Belt and even worldwide given the globalcoverage of its input data.

54 ENVIRONMENTAL SCIENCES↗

Machine Learning with Gradient-Based Optimization of Nuclear Waste Vitrification with Uncertainties and Constraints

Gekko is an optimization suite in Python that solves optimization problems involving mixed-integer, nonlinear, and differential equations. The purpose of this study is to integrate common Machine Learning (ML) algorithms such as Gaussian Process Regression (GPR), support vector regression (SVR), and artificial neural network (ANN) models into Gekko to solve data based optimization problems. Uncertainty quantification (UQ) is used alongside ML for better decision making. These methods include ensemble methods, model-specific methods, conformal predictions, and the delta method. An optimization problem involving nuclear waste vitrification is presented to demonstrate the benefit of ML in this field. ML models are compared against the current partial quadratic mixture (PQM) model in an optimization problem in Gekko. GPR with conformal uncertainty was chosen as the best substitute model as it had a lower mean squared error of 0.0025 compared to 0.018 and more confidently predicted a higher waste loading of 37.5 wt% compared to 34 wt%. The example problem shows that these tools can be used in similar industry settings where easier use and better performance is needed over classical approaches. Future works with these tools include expanding them with other regression models and UQ methods, and exploration into other optimization problems or dynamic control.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Can reanalysis products outperform mesoscale numerical weather prediction models in modeling the wind resource in simple terrain?

Mesoscale numerical weather prediction (NWP) models are generally considered more accurate than reanalysis products in characterizing the wind resource at heights of interest for wind energy, given their finer spatial resolution and more comprehensive physics. However, advancements in the latest ERA-5 reanalysis product motivate an assessment on whether ERA-5 can model wind speeds as well as a state-of-the-art NWP model – the Weather Research and Forecasting (WRF) Model. We consider this research question for both simple terrain and offshore applications. Specifically, we compare wind profiles from ERA-5 and the preliminary WRF runs of the Wind Integration National Dataset (WIND) Toolkit Long-term Ensemble Dataset (WTK-LED) to those observed by lidars at a site in Oklahoma, United States, and in a United States Atlantic offshore wind energy area. We find that ERA-5 shows a significant negative bias (~-1ms-1) at both locations, with a larger bias at the land-based site. WTK-LED-predicted wind speed profiles show a limited negative bias (~-0.5ms-1) offshore and a slight positive bias (~+0.5ms-1) at the land-based site. On the other hand, we find that ERA-5 outperforms WTK-LED in terms of the centered root-mean-square error (cRMSE) and correlation coefficient, for both the land-based and offshore cases, in all atmospheric stability conditions. We find that WTK-LED's higher cRMSE is caused by its tendency to overpredict the amplitude of the wind speed diurnal cycle. At the land-based site, this is partially caused by wind plant wake effects not being accurately captured by WTK-LED.

17 WIND ENERGY↗