Search NASA⌕ Search

SEARCH · Search NASA

Results for “Load prediction”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

A review of machine learning in building load prediction

The surge of machine learning in recent years has been empowering engineer modeling in various fields. The decreasing hardware cost, increasing data accessibility, and advances of building automation system (BAS) allow the collection and storage of a significant amount of building operation data. The two facts provide great opportunities of applying machine learning to building energy systems modeling and analysis. There are a great number of research papers on this topic but there lacks a comprehensive and general review to summarize the current development, limitations, gaps and future trend. In this review paper series, machine learning techniques in building energy system modeling and analysis are reviewed under the organization and logic of the machine learning definition by Tom M. Mitchell: a computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E. This paper is the first part of the review paper series, which focuses on building load prediction. First, the applications of building load prediction model (task T) are reviewed. Then, the modeling algorithms improving machine learning performance and accuracy (performance P) are reviewed. At the same time, the literature on the data perspective for modeling (experience E), including data engineering from sensors level to data level, pre-processing, feature extraction and selection, is reviewed. Finally, what is well-studied and what is lacking but with great potential are concluded; the gaps between present and future utilization of machine learning techniques are identified; the future trend and development are also predicted. The target readers of this paper are not only researchers from the building side who can get exposed to cutting edge machine learning tools, but also those from machine learning side who can understand the potential and challenge to apply machine learning in buildings.

Liang, Zhang↗
DOE OSTI2020

FAST.Farm Development and Validation of Structural Load Prediction Against Large Eddy Simulations

FAST.Farm is a mid-fidelity engineering tool developed by the National Renewable Energy Laboratory targeted at accurately and efficiently predicting wind turbine power production and structural loading in wind farm settings, including wake interactions between turbines. FAST.Farm is based on several principles of the dynamic wake meandering (DWM) model, but also addresses limitations of previous DWM implementations. Previous FAST.Farm studies have shown the similarities and differences between FAST.Farm and large eddy simulations for rigid turbine cases. The objective of this work is to quantify the ability of FAST.Farm to accurately predict turbine structural response in a small wind farm. This is done by comparing FAST.Farm structural response to SOWFA-OpenFAST results for three laterally-aligned turbines. The purpose of this study is to characterize the similarities and differences between FAST.Farm and a higher fidelity model for predicting turbine structural response in a wind farm for differing atmospheric inflows. Strong statistical agreement was found between FAST.Farm and SOWFA-OpenFAST structural response for the non-waked upstream turbine, and good agreement was found for the downstream turbines for most structural quantities. Higher differences were seen for downstream turbines with low ambient turbulence intensity or yawed turbines, suggesting areas for FAST.Farm wake dynamics modeling improvements. For all cases and turbines, small statistical differences were seen between blade deflections and bending moments, with larger differences for tower-top and tower-base bending moments. Overall, the results establish confidence for applying FAST.Farm to wind farm power and loads analyses and identify areas where further model validations and model improvements should be targeted.

49 EE - Wind and Water Power Program - Wind (EE-4W↗
DOE OSTI2020

Building thermal load prediction through shallow machine learning and deep learning

Building thermal load prediction informs the optimization of cooling plant and thermal energy storage. Physics-based prediction models of building thermal load are constrained by the model and input complexity. In this study, we developed 12 data-driven models (7 shallow learning, 2 deep learning, and 3 heuristic methods) to predict building thermal load and compared shallow machine learning and deep learning. The 12 prediction models were compared with the measured cooling demand. It was found XGBoost (Extreme Gradient Boost) and LSTM (Long Short Term Memory) provided the most accurate load prediction in the shallow and deep learning category, and both outperformed the best baseline model, which uses the previous day's data for prediction. Then, we discussed how the prediction horizon and input uncertainty would influence the load prediction accuracy. Major conclusions are twofold: first, LSTM performs well in short-term prediction (1 h ahead) but not in long term prediction (24 h ahead), because the sequential information becomes less relevant and accordingly not so useful when the prediction horizon is long. Second, the presence of weather forecast uncertainty deteriorates XGBoost's accuracy and favors LSTM, because the sequential information makes the model more robust to input uncertainty. Training the model with the uncertain rather than accurate weather data could enhance the model's robustness. Finally, our findings have two implications for practice. First, LSTM is recommended for short-term load prediction given that weather forecast uncertainty is unavoidable. Second, XGBoost is recommended for long term prediction, and the model should be trained with the presence of input uncertainty.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

3D ion gyro-orbit heat load predictions for NSTX-U

Abstract High power tokamaks operate with divertor heat loads capable of destroying the plasma facing components (PFCs). High fidelity heat load predictions are necessary to ascertain the PFC state for design and during operation. Typical heat flux calculations are 2D, time invariant, and assume that power flows directly along the magnetic field lines (the optical approximation). These assumptions neglect the complex 3D geometries employed to protect the PFCs, the time varying nature of the plasma and PFC thermal state, and the helical trajectories of ions with finite Larmor radii (the gyro-orbit approximation). An integrated software framework, the heat flux engineering analysis toolkit (HEAT), was developed to generate time varying optical heat loads applied to real engineering computer aided design (CAD) (Looby et al 2022 Fusion Sci. Technol. 78 10–27). Recently, an ion-gyro orbit module has been added to HEAT. This module calculates the helical trajectories of ions as they gyrate about the magnetic field lines using kinetic theory macro-particles to accelerate the calculation. First, the new gyro-orbit module will be presented. Next, a comparison to existing research is performed. Finally, an analysis of the gyro-orbit heat loads for NSTX-U is presented for diverted discharges using the engineering CAD models utilized for PFC fabrication. Including these gyro-orbit effects can enhance the PFC performance by ‘smearing’ out the magnetic shadows associated with the castellated fish-scaled geometry. Simultaneously, the helical trajectories can degrade performance when they load narrow regions on edges and corners with high heat fluxes. Analysis of the trade-offs between these competing effects is included, and regions for further investigation are identified.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Calibration of urban building energy model using smart meter data for district peak load prediction

Urban building energy modeling (UBEM) is a powerful approach to assessing baseline building energy performance and retrofits with new technologies across building stocks in cities. However, the accuracy of UBEM is often constrained by the limited availability of reliable data about building characteristics and operations, such as envelope efficiency levels, HVAC system performance, and end-use load patterns. Existing research has performed UBEM calibration using annual or monthly energy consumption data, which falls short when higher-resolution time series applications are needed, such as peak load prediction for utility operation planning. This study presents a new framework for calibrating building energy models at urban scale using smart meter data, targeting the accurate prediction of summer peak electricity loads to support robust grid planning. The framework first integrates various data sources to enhance baseline input assumptions for building models, and then calibrates the baseline models through a pattern-matching approach. A case study using CityBES and two years of AMI data from over 9000 residential customers in Portland, Oregon, demonstrated the workflow and its effectiveness. The calibrated models achieved a daily peak load mean absolute percentage error of 2.6 % during the heatwave in the calibration year, and 2.0 % in the validation year using another year of AMI data. Using the calibrated models, we analyzed the demand flexibility potential of the district building stock as an application of UBEM calibration. The findings affirm the appropriate use of UBEM for peak electric load forecasting and demand side management at the utility distribution system level.

AMI data↗
DOE OSTI2026

VRN3P: Variational Recurrent Neural Network Based Net-Load Prediction under High Solar Penetration

This is the final technical report for the SETO-funded VRN3P project (PNNL# 76914). The goal of this project, led by Pacific Northwest National Laboratory (PNNL), in collaboration with Lawrence Livermore National Laboratory (LLNL) and Portland General Electric (PGE), was to develop and validate a deep variational recurrent neural network-based net-load prediction (VRN3P) framework for probabilistic time-series forecasting of day-ahead net-load under high solar penetration scenarios. The project team reports successful design of a novel probabilistic net-load forecasting architecture, comprising of a variational autoencoder and a recurrent neural network, which demonstrates 30% improvement in forecast performance, 60% improvement in training time, and consumes 44% less memory, when compared with conventional baseline models. The team tested the VRN3P model performance on GridLAB-D test-cases representing varying BTM solar penetration levels of 20%, 30%, and 50%, with integrated time-series net-load profiles provided by the utility partner (PGE). The VRN3P model demonstrate <2% hourly MAPE (averaged over the year) for day- ahead net-load forecast on the test scenario with 20% BTM solar. Transfer learning extension of the VRN3P model has demonstrated 8.33× speed-up in training, while still achieving acceptable forecast performance of 2.24% hourly MAPE on the 30% BTM solar penetration test-scenario. A preliminary version of the VRN3P GridAPPS-D™has been developed, along with a web-based interactive user-interface (named ‘Forte’) which has made available on GitHub for public use.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Mechanical loading prediction through accelerometry data during walking and running

ABSTRACT Currently, there is no way to assess mechanical loading variables such as peak ground reaction forces (pGRF) and peak loading rate (pLR) in clinical settings. The purpose of this study was to develop accelerometry‐based equations to predict both pGRF and pLR during walking and running. One hundred and thirty one subjects (79 females; 76.9 ± 19.6 kg) walked and ran at different speeds (2–14 km·h −1 ) on a force plate–instrumented treadmill while wearing accelerometers at their ankle, lower back and hip. Regression equations were developed to predict pGRF and pLR from accelerometry data. Leave‐one‐out cross‐validation was used to calculate prediction accuracy and Bland–Altman plots. Our pGRF prediction equation was compared with a reference equation previously published. Body mass and peak acceleration were included for pGRF prediction and body mass and peak acceleration rate for pLR prediction. All pGRF equation coefficients of determination were above 0.96, and a good agreement between actual and predicted pGRF was observed, with a mean absolute percent error (MAPE) below 7.3%. Accuracy indices from our equations were better than previously developed equations. All pLR prediction equations presented a lower accuracy compared to those developed to predict pGRF. Walking and running pGRF can be predicted with high accuracy by accelerometry‐based equations, representing an easy way to determine mechanical loading in free‐living conditions. The pLR prediction equations yielded a somewhat lower prediction accuracy compared with the pGRF equations.

Veras, Lucas↗
DOE OSTI2020

Accuracy of HVAC Load Predictions: Validation of EnergyPlus and DOE-2 using FLEXLAB Measurements

The aim of the project reported here was to better understand the level of accuracy of three building energy simulation (BES) engines (‘engines’) — EnergyPlus™, DOE-2.1e, and DOE-2.2 — by identifying and investigating significant deviations between the performance predicted by these engines and actual performance as measured in the FLEXLAB® test facility at Lawrence Berkeley National Laboratory (LBNL). The specific test conditions included some of those prescribed in ANSI/ASHRAE Standard 140 - Standard Method of Test for the Evaluation of Building Energy Analysis Computer Programs. Detailed measurements of FLEXLAB performance, including indoor temperatures and heat fluxes and air-flow and water flow rates and temperatures in the Heating, Ventilating and Air Conditioning (HVAC) system, together with hourly weather data, were recorded and used in analyzing the simulation results from EnergyPlus v8.8, DOE-2.2 v3.65 and DOE-2.1e v127. These engines are commonly used in the United States for building energy code compliance, federal, state, and utility incentives programs, as well as energy efficient design of new buildings and energy retrofit of existing buildings. Seven conventional overhead mixing ventilation scenarios were tested and each engine was found to have a similar level of agreement with the measurements of space-level heating and sensible cooling loads. These results provide useful information regarding the accuracy of these engines in predicting the cooling and heating load elements of whole building energy performance. This information is intended for practitioners who are concerned about transitioning between simulation tools with different engines and for managers of utility programs leveraging these tools for evaluating and/or projecting measure savings to be incentivized under their programs. The results of the comparisons of simulated and measured performance indicate that the predictions from all three engines are not significantly different. The 24-hour average value of the absolute mean bias indicates the likely magnitude of the error in any particular case. The average mean bias is reduced by cancelation of overprediction in one case by underprediction in another. The daytime absolute mean biases, which may be more important for both energy performance and occupant comfort, are ~6%, presumably because of the greater complexity involved in simulating in the presence of solar radiation. EnergyPlus typically overpredicts the cooling load and/or underpredicts the heating load by ~1.5% and the DOE-2 engines typically underpredict the cooling load by approximately the same amount. The Root Mean Square Error is relatively more sensitive to shorter term variations in the difference between predicted and measured loads; the three engines have similar values, ~10%, suggesting that the uncertainties in their predictions of peak loads may also be similar in magnitude. The implication of these results is that users, both designers and program analysts, can use EnergyPlus, DOE-2.1e, or DOE-2.2 to model conventional commercial buildings equipped with overhead mixing ventilation with a similar level of confidence. Further work is required to better understand the variability in the level of agreement between the engine predictions and FLEXLAB measurements, where a particular engine will agree well with FLEXLAB in some cases and not so well in others and another engine will agree or disagree in different cases. As the sources of this variability are identified and eliminated or reduced significantly, it is recommended that the experimental capabilities and methods developed in the study reported here should be applied to validating heating and cooling load calculations for spaces with different types of furniture and miscellaneous loads. These methods should then be applied to low energy space conditioning systems in EnergyPlus including, in particular, radiant slab and radiant ceiling panel cooling and heating systems and ‘mixed mode’ systems that combine mechanical cooling and natural ventilation systems, focusing on controls, including control of thermal mass. The work reported here addresses the conventional method of heating and cooling occupied spaces; other methods, such as the use of radiant heating and cooling systems have the potential to provide equivalent occupant comfort, or better, with lower energy consumption. These systems are addressed more explicitly in EnergyPlus but there is a need for empirical validation to give users the same level of confidence in modeling these systems that they have, or should have, in modeling conventional systems, based on the results presented here.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Difference in load predictions obtained with effective turbulence vs. a dynamic wake meandering modeling approach

Abstract. According to the international standard for wind turbine design, the effects of wind turbine wakes on structural loads can be considered in two ways: (1) by augmenting the ambient turbulence levels with the effective turbulence model (EFF) and then calculating the resulting loads and (2) by performing dynamic wake meandering (DWM) simulations, which compute wake effects and loads for all turbines on a farm at once. There is no definitive answer in scientific literature as to the consequences of choosing one model over the other, but the two approaches are unarguably very different. The work presented here expounds on these differences and investigates to what extent they affect the simulated structural loads. We consider an idealized 4×4 rectangular array of National Renewable Energy Laboratory 5 MW wind turbines with a spacing of 5 by 8 rotor diameters and three wind speed scenarios at high ambient turbulence. Load simulations are performed in OpenFAST with EFF and in FAST.Farm with the DWM implementation. We compare ambient turbulence, wind farm turbulence, and loads between both approaches. When omnidirectional results are compared, EFF wind farm turbulence intensity is consistently higher by 0.2 % (above-rated wind speed) to 2.7 % (below-rated wind speed). However, for certain wind directions, the EFF turbulence can be lower than FAST.Farm by almost 9 %. Wind speeds within the farm were found to differ by up to 3 m s−1 due to the lack of wake deficits in the EFF approach, leading to longer tails toward low values in the FAST.Farm mean load distributions. Consistent with the turbulence results, the median EFF load standard deviations are also consistently higher, by a maximum of 20 % and 17 % for blade-root out-of-plane and tower-base fore-aft moments, respectively.

17 WIND ENERGY↗
DOE OSTI2020

New vortex‐lift and tangential‐force models for HAWT aerodynamic load prediction

Abstract Horizontal axis wind turbines (HAWTs) experience three‐dimensional rotational and unsteady aerodynamic phenomena at the rotor blades sections. These highly unsteady three‐dimensional effects have a dramatic impact on the aerodynamic load distributions on the blades, in particular, when they occur at high angles of attack due to stall delay and dynamic stall. Unfortunately, there is no complete understanding of the flow physics yet at these unsteady 3D flow conditions, and hence, the existing published theoretical models are often incapable of modelling the impact on the turbine response realistically. The purpose of this paper is to provide an insight on the combined influence of the stall delay and dynamic stall on the blade load history of wind turbines in controlled and uncontrolled conditions. New dynamic stall vortex and nonlinear tangential force coefficient modules, which integrally take into account the three dimensional rotational effect, are also proposed in this paper. This module along with the unsteady influence of turbulent wind speed and tower shadow is implemented in a blade element momentum (BEM) model to estimate the aerodynamic loads on a rotating blade more accurately. This work presents an important step to help modelling the combined influence of the stall delay and dynamic stall on the load history of the rotating wind turbine blades which is vital to have lighter turbine blades and improved wind turbine design systems.

Elgammi, Moutaz↗
DOE OSTI2023

EV charging site day-ahead load prediction in a synthetic environment for RL based grid-informed charging

Ensuring grid health in the face of increasing demand for power is an emerging challenge especially due to transportation electrification. A free market approach to influencing electric vehicle (EV) load through grid-informed hourly dynamic pricing is introduced in this work. The setting of charging price is done by a reinforcement learning (RL) agent that learns the complicated dynamics by interacting with a synthetic environment. This synthetic environment is a combination of distribution feeder simulation, EV charger user behavior dynamics, and EV charging simulation. A key module in this synthetic environment involves obtaining the day-ahead charging profile of EV charging stations based on real-world past data. The day-ahead prediction is also useful in other traditional optimizations related to EV charge scheduling. The proposed approach involves using EV charging data from two different past time horizons – one to determine the shape of the daily profile and the other to determine a scaling value to capture actual energy consumption. Real-world charging data over many years from the ACN charging network has been used to demonstrate the ability to predict the day-ahead profile with only charge session data. Both Python and MATLAB have been used for data cleaning, processing, analysis, and prediction.

Suryanarayana, Harish↗
DOE OSTI2024

Short-Term Load Forecasting Considering EV Charging Loads with Prediction Interval Evaluation

Short-term load forecasting plays a critical role in power system planning and operation. Along with the electrification of various loads, electricity demands are becoming increasingly hard to predict. Notably, the recent rise in electric vehicles (EVs) has further contributed to this unpredictability. To address this issue, this paper proposes a probabilistic load forecasting strategy utilizing Gaussian process regression, structured in a day-ahead manner. While many works focus on deterministic prediction, probabilistic forecasting offers additional insights into variability and uncertainty, enabling more flexible and reliable operation for power systems. To enhance the accuracy of the load forecasting model, the inputs include features related to EV charging habits as well as commonly used weather information. The load forecasting results are evaluated using various metrics, including conventional ones that assess the accuracy of point forecasts, as well as additional metrics that test the reliability of prediction intervals. The proposed load forecasting method is finally tested on real residential power consumption data and EV charging data sampled from real-world sources. The results prove that the new features can greatly improve the performance of the load forecasting method.

electrical vehicle↗
DOE OSTI2024

End-Use Savings Shapes Measure Documentation: Dispatch Schedule Generation for Demand Flexibility Measures

This supplemental document describes the methodology used for determining the dispatch timing of various EUSS demand flexibility measures. Demand flexibility measures are designed to reduce/dispatch electricity demand in buildings during especially beneficial/critical times. The method used in this work utilizes predictions of building loads to generate a schedule that reflects the periods when the building's daily peak load occurs to support decision making in demand flexibility measures. The dispatch schedule generation method described in this document creates an hourly schedule that includes a load dispatch (peak) window for each day for a whole year based on load prediction, with options using different prediction methods: perfect prediction, bin-sampling method, fixed schedule, and outdoor air temperature (OAT)-based prediction method. The perfect prediction method performs a simulation to obtain the annual load profile as predicted load, representing the scenario of perfect load prediction. The bin-sampling method (1) categorizes days into representative bins by temperature characteristics, (2) performs simulations on sample days from each of those bins to create representative (or predicted) load, and (3) assigns representative loads for all days in a year based on the bin categorization. The fixed schedule method defines uniform start and end time of peak window with assumed fixed daily peak time, for all days in a season or a year. The OAT-based prediction method uses the statistics of OAT (minimum and maximum) as the indicators of peak load, with specified delay response time from building loads to temperature. Given the load prediction, daily peak periods are determined as a time window with specified length in each day that include the predicted daily peak load and with a secondary rule such as maximizing energy saving potential. The dispatch schedule generation method is not a standalone measure and is intended to be combined with other demand flexibility measures that could leverage the peak schedule and apply demand controls on specific systems or devices for demand response, such as measures described in "Measure Documentation - Thermostat Control for Load Shedding" and "Measure Documentation - Thermostat Control for Load Shifting".

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Machine learning surrogate of physics-based building-stock simulator for end-use load forecasting

Building energy models are used to simulate heat and mass transfer and estimate end-use load in buildings. With the proliferation of solar photovoltaics on residential and commercial buildings, increasingly, buildings are expected to provide grid services, for which accurate and computationally efficient building energy simulations and end-use load prediction are imperative. Existing building energy simulation tools, however, have significant computational overhead that make them less practical in real-time deployment for optimization, design, uncertainty quantification and control in building energy management systems. Here this article presents a data-driven machine learning model based on light gradient boosting method (LightGBM) as a surrogate for a physics-based simulator for residential buildings to predict end-use load. The machine learning based surrogate model accounts for time-series related variables, seasonality and trend component of end-use load, and history of end-use load. The accuracy of the surrogate model is assessed on the prediction of the load profiles of 100 different houses in Cook County, Illinois, USA. The LightGBM surrogate model is shown to reduce the root-mean-squared error by 53% relative to a reference decision tree (DT) based model reported previously in the literature. Moreover, the model predicts the load spikes and high-ramp rate events throughout the year which are often the Achilles heel of other models in the literature. The machine learning based surrogate model is demonstrated to be computationally efficient, with a ten-fold reduction in the computational time compared to a physics-based building energy simulation, and suitable for uncertainty analysis and real-time control of building characteristics in response to uncertainty.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2020

State-of-the-art on research and applications of machine learning in the building life cycle

Fueled by big data, powerful and affordable computing resources, and advanced algorithms, machine learning has been explored and applied to buildings research for the past decades and has demonstrated its potential to enhance building performance. This study systematically surveyed how machine learning has been applied at different stages of building life cycle. By conducting a literature search on the Web of Knowledge platform, we found 9579 papers in this field and selected 153 papers for an in-depth review. The number of published papers is increasing year by year, with a focus on building design, operation, and control. However, no study was found using machine learning in building commissioning. There are successful pilot studies on fault detection and diagnosis of HVAC equipment and systems, load prediction, energy baseline estimate, load shape clustering, occupancy prediction, and learning occupant behaviors and energy use patterns. None of the existing studies were adopted broadly by the building industry, due to common challenges including (1) lack of large scale labeled data to train and validate the model, (2) lack of model transferability, which limits a model trained with one data-rich building to be used in another building with limited data, (3) lack of strong justification of costs and benefits of deploying machine learning, and (4) the performance might not be reliable and robust for the stated goals, as the method might work for some buildings but could not be generalized to others. Finally, findings from the study can inform future machine learning research to improve occupant comfort, energy efficiency, demand flexibility, and resilience of buildings, as well as to inspire young researchers in the field to explore multidisciplinary approaches that integrate building science, computing science, data science, and social science.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Predicting Wind Loading and Instability in Solar Tracking PV Arrays

Wind loading and the fluctuating pressure loads it creates on PV panel surfaces are associated with multiple degradation mechanisms and failures. Modest wind speeds create reversing loads that can initiate cell cracks and weather cracked cells. Stronger wind speeds and extreme weather events can lead to larger scale forces and the aerodynamic instability known as torsional galloping. All these effects are dependent on the complex coupling between wind speed, panel orientation, and a myriad of other hardware and site-specific factors. In this work, we present the latest developments from our work to build an open-source, high-performance computing (HPC) fluid dynamics solver to predict and mitigate these effects. This simulation package allows users to easily specify different array layouts, solar-tracking angles, panel geometries, and weather conditions before automatically generating a refined computational mesh and solving for the unsteady loading on each panel surface. Small domains (e.g., a single panel row in isolation) can be solved on a modern laptop, while larger domains or very high-fidelity studies can be solved on distributed or HPC resources with minimal modifications to the underlying problem specification. We present preliminary case studies obtained using this simulation package and highlight how increased wind speeds combined with sub-optimal tracking angles can exacerbate degradation drivers.

aerodynamics↗
DOE OSTI2025

Data efficiency assessment of generative adversarial networks in energy applications

This study investigates the data requirements of generative artificial intelligence (AI), particularly generative adversarial networks (GANs), for reliable data augmentation in energy applications. Generative AI, though seen as a solution to data limitations, requires substantial data to learn meaningful distributions—a challenge often overlooked. This study addresses the challenge through synthetic data generation for critical heat flux (CHF) and power grid demand, focusing on renewable and nuclear energy. Two variants of GAN employed are conditional GAN (cGAN) and Wasserstein GAN (wGAN). Our findings include the strong dependency of GAN on data size, with performance declining on smaller datasets and varying performance when generalizing to unseen experiments. Mass flux and heated length significantly influence CHF predictions. wGAN is more robust to feature exclusion, making it suitable for constrained synthetic data generation. In energy demand forecasting, wGAN performed well for solar, wind, and load predictions. Longer lookback hours and larger datasets improved predictions, especially for load power. Seasonal variations posed challenges, with wGAN achieving a relatively high error of Root Mean Squared Error (RMSE) of 0.32 for load power prediction, compared to RMSE of 0.07 under same-season conditions. Feature exclusions impacted cGAN the most, while wGAN showed greater robustness. This study concludes that, while generative AI is effective for data augmentation, it requires substantial data and careful training to generate realistic synthetic data and generalize to new experiments in engineering applications.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗