Search NASA⌕ Search

SEARCH · Search NASA

Results for “Network performance forecasting”

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 199 records · Page 11

Physics informed neural network can retrieve rate and state friction parameters from acoustic monitoring of laboratory stick-slip experiments

Various machine learning (ML) and deep learning (DL) techniques have been recently applied to the forecasting of laboratory earthquakes from friction experiments. The magnitude and timing of shear failures in stick-slip cycles are predicted using features extracted from the recorded ultrasonic or acoustic emission (AE) signals. In addition, the Rate and State Friction (RSF) constitutive laws are extensively used to model the frictional behavior of faults. In this work, we use data from shear experiments coupled with passive acoustic (variance, kurtosis, and AE rate) interleaved with active source ultrasonic monitoring (transmitted wave amplitude) to develop physics-informed neural network (PINN) models incorporating the RSF law and AE rate generation equation with wave amplitude serving as a proxy for friction state variable. This PINN framework allows learning RSF parameters from stick-slip experiments rather than measuring them through a series of velocity step experiments. We observe that when the stick-slip cycles are irregular, the PINN models outperform the data-driven DL models. Transfer learning (TL) PINN models are also developed by pre-training on data collected at one normal stress level followed by forecasting shear failures and retrieving RSF parameters at other stress levels (i.e., with different recurrence intervals) after retraining on a limited amount of new data. Our findings suggest that TL models perform better compared to standalone models. Both standalone and TL PINN-estimated RSF parameters and their ground truth values show excellent agreements thus demonstrating that RSF parameters can be retrieved from laboratory stick-slip experiments using the corresponding acoustic data and that the transmitted wave amplitude provides a good representation of the evolving frictional state during stick-slips.

58 GEOSCIENCES↗

Sean Michaletz Directors Post Doc Fellow Report

Predicting climate change effects on plant function is a central challenge of global change biology and a primary mission of DOE. Although increasing temperatures and drought have been associated with reduced growth and increased mortality of plants, accurate prediction of such responses is limited by a lack of process-based theory linking climate and whole-plant physiology. This inability to predict forest mortality can cause significant biases in climate forecasts. One way forward is metabolic scaling theory (MST), which proposes that physiologic rates – from cells to the globe – are governed by the rates of resource distribution through vascular networks and the kinetics of resource utilization by metabolic reactions. MST has traditionally not considered rates of resource acquisition from organism-environment interactions, but it has an ideal mechanistic basis for doing so. As a first step towards integrating these processes, Sean has extended MST to characterize effects of temperature and precipitation on plant growth and ecosystem production. Sean’s post doc fellowship aimed to address a remaining shortcoming in that the new theory does not yet consider the physical processes of resource acquisition, and thus cannot mechanistically predict plant performance in a changing climate.

54 ENVIRONMENTAL SCIENCES↗

Development of lean, efficient, and fast physics-framed deep-learning-based proxy models for subsurface carbon storage

In this work, we present deep-learning-based surrogate models for CCUS developed with four different algorithms and a physics-framed two-phase flow problem involving displacement of water by CO 2 . The deep-learning models were trained using 3D datasets describing the pressure plume, CO 2 saturation plume, and water extraction rate generated by numerical simulation. The hyperparameters defining the architecture of the neural networks were optimized to determine the slimmest network size and training parameters that give the most efficient performance at the least training cost. To develop a robust model that closely mimics the governing physical laws, the discretized form of the two-phase fluid transport equation was used to formulate the supervised deep-learning task. The algorithms investigated in this study predicted the data to above 95% accuracy, with the multi-layer perceptron model demonstrating the best performance by balancing training speed, prediction time, and prediction accuracy with lean network capacity. Furthermore, the surrogate models simultaneously predict reservoir pressure and CO 2 saturation in every grid block, including the surface well extraction rate and bottomhole pressure, at all simulation times for a given static model realization in just a few seconds on a standard desktop computer. A key outcome of this study is that limits can be placed on network design parameters to avoid over designing neural networks, with associated efficiencies in training and prediction times. This is very useful because large volumes of data may be generated in CCUS projects and over-design of neural network architectures imposes penalties that are antithetical to the goal of near-real time forecasting.

58 GEOSCIENCES↗

A search for sterile neutrinos at the MINOS experiment

MINOS is a long baseline neutrino oscillation experiment based at the Fermi National Accelerator Laboratory in Illinois, USA. The experiment was designed to study neutrino oscillation phenomena. The v μ beam produced by the NuMI beam facility at FNAL is used along with two functionally identical detectors. The Near Detector at FNAL and a Far Detector 735 km away in the Soudan Underground Laboratory in northern Minnesota. Comparison of the observed spectra of neutrinos at the two detectors provides the evidence for neutrino oscillations. This thesis presents work on the postulated phenomena of sterile neutrinos. Oscillations between active and sterile neutrinos will lead to a deficit in the expected rate of measured Neutral Current interactions at the Far Detector. A technique for selecting Neutral Current events utilizing an Artificial Neural Network is presented with resulting overall efficiency of 91.1% and purity of 66.0%. A method of predicting the expected Charged and Neutral Current energy spectra at the Far Detector given the data recorded at the Near Detector is presented. A model to search for oscillations between sterile and active neutrinos is developed. Sources of systematic uncertainty that can effect the results of the analysis are discussed. The analysis developed is applied to a Standard Model 3 flavour oscillation model as a cross check under the scenarios with and without v e appearance. The oscillation parameters measured by this model are Δm 32 2 = (2.39 -0.15 +0.23 ) x 10 -3 eV 2 and θ 23 = 0.727 -0.11 +0.22 for the no v e appearance result. An analysis of the resulting prediction reveals no evidence for active neutrino disappearance. The analysis is then performed using the 4 flavour neutrino oscillation model developed. Again this is done under the 2 scenarios of v e appearance and no v e appearance. The results of this analysis are Δm 31 2 = 2.44 -0.14 +0.23 x 10 -3 eV 2 , θ 23 = 0.755 -0.12 +0.19 and θ 34 = 0.00 +0.35 for no v e appearance and Δm 31 2 = (2.46 -0.14 +0.21 ) x 10 -3 eV 2 , θ 23 = 0.849 -0.19 +0.12 and θ 34 = 0.00 +0.60 for v e appearance. This is consistent with no oscillations between active and sterile neutrinos.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Automated Framework for Groundwater Monitoring Using DWT with LSTM and Transformers

Environmental monitoring is critical for safeguarding public health and ecological well-being. Traditional data structuring and workflow monitoring methods consume significant time and effort, hindering timely insights and effective decision-making. Our study addresses this challenge by presenting an AI framework that automates data cleaning, structuring, and modeling processes, specifically targeting applications in groundwater monitoring. By leveraging automation for data processing and model training, our framework establishes a novel and efficient paradigm for environmental monitoring, with its potential application to the vast network of over a hundred Department of Energy Environmental Management (DoE-EM) cleanup sites across the country. It analyzes data streams from a network of groundwater Internet-of-Things (IoT) sensors deployed at the Savannah River Site (SRS) for prediction modeling. This allows human experts to focus on analysis and decision-making, ultimately leading to better environmental outcomes.The framework employs multivariate time-series forecasting methods to study and model the behavior of varying chemical analytes. The continuous learning process is enabled by utilizing deep learning techniques. It allows the framework to become more nuanced in its analysis over time, adapting to the specific characteristics of the environmental site and the evolving nature of contaminant behavior. Deep learning models known for sequence modeling, LSTM, and Transformers are employed for time series forecasting. Data processing and structuring are essential components significantly impacting the final model's performance. This hypothesis was proven by presenting a comparative analysis of model performance with processed and unprocessed data. The feature engineering approach utilized was the Discrete Wavelet Transform, which works well with time series data.

Discrete Wavelet Transform (DWT)↗

Forecasting district-scale energy dynamics through integrating building network and long short-term memory learning algorithm

With the development of data-driven techniques, building energy prediction on a district level has attracted increasing attention in recent years for revealing energy use patterns and reduction potentials. However, data acquisition in large building groups is difficult and adjacent buildings also interact with each other. To reduce data cost and incorporate the inter-building impact with the data-driven building energy model, this study proposes a deep learning predictive approach that fuses the building network model with a long short-term memory learning model for district-scale building energy modeling. The building network was constructed based on correlations between the energy use intensity of buildings, which can significantly reduce the computational complexity of the deep learning models for energy dynamic prediction. Five typical building groups with energy use data from 2015 to 2018 on two institutional campuses were selected to perform the validation experiment with TensorFlow. Based on the prediction error assessments, the results suggest that for total building energy use intensity prediction, the proposed model can achieve a mean absolute percentage error of 6.66% and a root mean square error of 0.36 kWh/m 2 , compared to 12.05% and 0.63 kWh/m 2 of the conventional artificial neural network model and to 11.06% and 0.89 kWh/m 2 for the support vector regression model.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Probabilistic Prediction of Geomagnetic Storms and the K p index

Geomagnetic activity is often described using summary indices to summarize the likelihood of space weather impacts, as well as when parameterizing space weather models. The geomagnetic index K p in particular, is widely used for these purposes. Current state-of-the-art forecast models provide deterministic K p predictions using a variety of methods – including empirically-derived functions, physics-based models, and neural networks – but do not provide uncertainty estimates associated with the forecast. This paper provides a sample methodology to generate a 3-hour-ahead K p prediction with uncertainty bounds and from this provide a probabilistic geomagnetic storm forecast. Specifically, we have used a two-layered architecture to separately predict storm (K p ≥ 5 – ) and non-storm cases. As solar wind-driven models are limited in their ability to predict the onset of transient-driven activity we also introduce a model variant using solar X-ray flux to assess whether simple models including proxies for solar activity can improve the predictions of geomagnetic storm activity with lead times longer than the L1-to-Earth propagation time. By comparing the performance of these models we show that including operationally-available information about solar irradiance enhances the ability of predictive models to capture the onset of geomagnetic storms and that this can be achieved while also enabling probabilistic forecasts.

79 ASTRONOMY AND ASTROPHYSICS↗

Temporally-consistent koopman autoencoders for forecasting dynamical systems

Absence of sufficiently high-quality data often poses a key challenge in data-driven modeling of high-dimensional spatio-temporal dynamical systems. Koopman Autoencoders (KAEs) harness the expressivity of deep neural networks (DNNs), the dimension reduction capabilities of autoencoders, and the spectral properties of the Koopman operator to learn a reduced-order feature space with simpler, linear dynamics. However, the effectiveness of KAEs is hindered by limited and noisy training datasets, leading to poor generalizability. To address this, we introduce the Temporally-Consistent Koopman Autoencoder (tcKAE), designed to generate accurate long-term predictions even with limited and noisy training data. This is achieved through a consistency regularization term that enforces prediction coherence across different time steps, thus enhancing the robustness and generalizability of tcKAE over existing models. We provide analytical justification for this approach based on Koopman spectral theory and empirically demonstrate tcKAE’s superior performance over state-of-the-art KAE models across a variety of test cases, including simple pendulum oscillations, kinetic plasma, and fluid flow data.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Monitoring of Liquid Metal Reactor Heater Zones with Recurrent Neural Network Learning of Temperature Time Series

Advanced high-temperature fluid reactors (ARs), such as sodium fast reactors (SFRs) and molten salt cooled reactors (MSCRs) utilize high-temperature fluids at ambient pressure. To melt the fluid during reactor startup and prevent fluid freezing during cooldown, the thermal–hydraulic systems of such ARs include heater zones consisting of specific heaters with controllers, temperature sensors, and thermal insulation. The failure of heater zones due to insulation material degradation or improper installation, resulting in parasitic heat losses, can lead to fluid freezing. The detection of faults using a heat-transfer model is difficult because of a lack of knowledge of the experimental details. Data-driven machine learning of heater zone temperature time series offers a viable alternative. In this study, we benchmarked the performance of recurrent neural networks (RNNs) in an analysis of heat-up transient temperature time series of heater zones installed on a liquid sodium vessel. The RNN models include long short-term memory (LSTM) and gated recurrent unit (GRU) networks, as well as their bi-directional variants, BiLSTM and BiGRU. Anomalous temperature points were designated using a percentile-based threshold applied to residual fluctuations in the detrended temperature time series. Additionally, the impact of the exponentially weighted moving average (EWMA) method on detection accuracy was examined. The RNN models’ performance was assessed using precision, recall, and F 1 score metrics. Results demonstrated that RNN models effectively detect anomalies in temperature time series with the best models for each heater zone achieving F 1 scores of over 93%. To explain the variations in RNN model performance across different heater zones, we used Kullback–Leibler (KL) divergence to quantify the relative entropy between training and testing data, and the Detrended Fluctuation Analysis (DFA) to assess long-range temporal correlations. For datasets with strong long-range correlations and minimal relative entropy between training and testing data, GRU is the best-performing model. When the data exhibits weaker long-term correlations and a significant relative entropy between training and testing distributions, BiGRU shows the best performance. For the data sets with intermediate values of both KL divergence and DFA, the best performance is obtained with LSTM and BiLSTM, respectively.

gated recurrent unit↗

Improving in vitro to in vivo extrapolation by incorporating toxicokinetic measurements: A case study of lindane-induced neurotoxicity

Approaches for extrapolating in vitro toxicity testing results for prediction of human in vivo outcomes are needed. The purpose of this case study was to employ in vitro toxicokinetics and PBPK modeling to perform in vitro to in vivo extrapolation (IVIVE) of lindane neurotoxicity. Lindane cell and media concentrations in vitro, together with in vitro concentration-response data for lindane effects on neuronal network firing rates, were compared to in vivo data and model simulations as an exercise in extrapolation for chemical-induced neurotoxicity in rodents and humans. Time- and concentration-dependent lindane dosimetry was determined in primary cultures of rat cortical neurons in vitro using “faux” (without electrodes) microelectrode arrays (MEAs). In vivo data were derived from literature values, and physiologically based pharmacokinetic (PBPK) modeling was used to extrapolate from rat to human. The previously determined EC{sub 50} for increased firing rates in primary cultures of cortical neurons was 0.6 μg/ml. Media and cell lindane concentrations at the EC{sub 50} were 0.4 μg/ml and 7.1 μg/ml, respectively, and cellular lindane accumulation was time- and concentration-dependent. Rat blood and brain lindane levels during seizures were 1.7–1.9 μg/ml and 5–11 μg/ml, respectively. Brain lindane levels associated with seizures in rats and those predicted for humans (average = 7 μg/ml) by PBPK modeling were very similar to in vitro concentrations detected in cortical cells at the EC{sub 50} dose. PBPK model predictions matched literature data and timing. These findings indicate that in vitro MEA results are predictive of in vivo responses to lindane and demonstrate a successful modeling approach for IVIVE of rat and human neurotoxicity. - Highlights: • In vitro to in vivo extrapolation for lindane neurotoxicity was performed. • Dosimetry of lindane in a micro-electrode array (MEA) test system was assessed. • Cell concentrations at the MEA EC{sub 50} equaled rat brain levels associated with seizure. • PBPK-predicted human brain levels at seizure also equaled EC{sub 50} cell concentrations. • In vitro MEA results are predictive of lindane in vivo dose–response in rats/humans.

60 APPLIED LIFE SCIENCES↗

High‐Resolution National‐Scale Water Modeling Is Enhanced by Multiscale Differentiable Physics‐Informed Machine Learning

Abstract The National Water Model (NWM) is a key tool for flood forecasting, planning, and water management. Key challenges facing the NWM include calibration and parameter regionalization when confronted with big data. We present two novel versions of high‐resolution (∼37 km 2 ) differentiable models (a type of hybrid model): one with implicit, unit‐hydrograph‐style routing and another with explicit Muskingum‐Cunge routing in the river network. The former predicts streamflow at basin outlets whereas the latter presents a discretized product that seamlessly covers rivers in the conterminous United States (CONUS). Both versions use neural networks to provide a multiscale parameterization and process‐based equations to provide a structural backbone, which were trained simultaneously (“end‐to‐end”) on 2,807 basins across the CONUS and evaluated on 4,997 basins. Both versions show great potential to elevate future NWM performance for extensively calibrated as well as ungauged sites: the median daily Nash‐Sutcliffe efficiency of all 4,997 basins is improved to around 0.68 from 0.48 of NWM3.0. As they resolve spatial heterogeneity, both versions greatly improved simulations in the western CONUS and also in the Prairie Pothole Region, a long‐standing modeling challenge. The Muskingum‐Cunge version further improved performance for basins >10,000 km 2 . Overall, our results show how neural‐network‐based parameterizations can improve NWM performance for providing operational flood predictions while maintaining interpretability and multivariate outputs. The modeling system supports the Basic Model Interface (BMI), which allows seamless integration with the next‐generation NWM. We also provide a CONUS‐scale hydrologic data set for further evaluation and use.

Song, Yalan [Civil and Environmental Engineering T↗

Long short-term memory embedded nudging schemes for nonlinear data assimilation of geophysical flows

Reduced rank nonlinear filters are increasingly utilized in data assimilation of geophysical flows, but often require a set of ensemble forward simulations to estimate forecast covariance. On the other hand, predictor-corrector type nudging approaches are still attractive due to their simplicity of implementation when more complex methods need to be avoided. However, optimal estimate of nudging gain matrix might be cumbersome. In this paper, we put forth a fully nonintrusive recurrent neural network approach based on a long short-term memory (LSTM) embedding architecture to estimate the nudging term, which plays a role not only to force the state trajectories to the observations but also acts as a stabilizer. Furthermore, our approach relies on the power of archival data and the trained model can be retrained effectively due to power of transfer learning in any neural network applications. In order to verify the feasibility of the proposed approach, we perform twin experiments using Lorenz 96 system. Our results demonstrate that the proposed LSTM nudging approach yields more accurate estimates than both extended Kalman filter (EKF) and ensemble Kalman filter (EnKF) when only sparse observations are available. With the availability of emerging AI-friendly and modular hardware technologies and heterogeneous computing platforms, we articulate that our simplistic nudging framework turns out to be computationally more efficient than either the EKF or EnKF approaches.

42 ENGINEERING↗

A Review of Behind-the-Meter Solar Generation Modeling and Forecasting

Solar photovoltaic systems largely integrated within the distribution grid are operated 'behind-the-meter' and power generation cannot be directly monitored by most utilities. The increasing penetration of behind-the-meter solar photovoltaic systems can deter efficient network and market operations due to variability and uncertainty in net load, which is exacerbated by limited visibility and the difficulty in analyzing the hosting capacity. Risk introduced by behind-the-meter solar contributions may hinder reliable and secure grid operations due to biased system monitoring and forecasts. Accurate behind-the-meter estimations, together with capacity and specification forecasts, thus play a key role in balancing supply and demand and this article reviews the pertinent literature, identifying key characteristics and predictive methods for efficient behind-the-meter solar photovoltaic generation. Forecasting is central to methods herein. The fundamental characteristics of behind-the-meter solar forecasting, including which methods are applicable for scenario-driven use cases, are driven by the metrics most useful for system-wide performance evaluation. To this aim, the literature is reviewed with a focus on forecasting applications for aggregate, regional behind-the-meter generation useful to bulk system and utility operations. As distinguished from net load forecasting, subtleties in these coincident tasks are explored before concluding with recommendations for current practice and future implementations.

behind-the-meter↗

Estimating soybean yields from high-temporal-resolution multi-source data using deep learning

Accurate and timely crop yield prediction is crucial for ensuring food security and maintaining stable agricultural markets. In recent years, there has been a surge in interest in leveraging high-temporal-resolution, multi-source data for effective crop growth monitoring and yield estimation. A notable challenge arises from the difficulty in capturing the intricate interactions between variables across different time steps within these high-temporal-resolution time series datasets. This complexity hinders the reliable extraction of yield information from voluminous and often noisy datasets, especially during periods of extreme weather events. Here, in this study, we propose an Attention and Graph Isomorphism Network-enhanced Bi-directional Long Short-Term Memory network (AGB-LSTM) for estimating county-level soybean yield in the United States. This model integrates a diverse set of remote sensing data, including Near-Infrared Reflectance of Vegetation (NIRv), Sun-Induced chlorophyll Fluorescence (SIF), and Gross Primary Productivity (GPP), along with environmental covariates. The AGB-LSTM effectively leverages information related to crop yield from high-temporal-resolution time series data (5-days), achieving an accuracy of R²= 0.67 and rRMSE = 14.46%. This approach significantly outperforms traditional machine learning methods such as Random Forest (RF) (R²= 0.52, rRMSE = 17.36%) and Bi-LSTM (R²= 0.58, rRMSE = 16.17%). Sensitivity experiments with different time steps and ranges demonstrated that our model could accurately and stably predict yields 1 to 2 months before harvest. Moreover, data with a finer temporal resolution consistently improved prediction performance, resulting in an approximately 20% increase in and an approximately 20% decrease in rRMSE compared to using monthly composites. We also evaluated the robustness of the model under extreme climate events and observed strong performance (R²= 0.50, rRMSE = 21.32%). Finally, yield mapping for major soybean-producing regions in North America in 2023 revealed spatial patterns that closely matched USDA yield reports. Our findings suggest that the AGB-LSTM model is a promising and effective method for estimating yield and has notable potential for global crop yield forecasting.

Deep learning↗

A shared-mobility-based framework for evacuation planning and operations under forecast uncertainty

To meet evacuation needs from carless populations who need personalized assistance to evacuate safely, in this article we propose a ridesharing-based evacuation program that recruits volunteer drivers before a disaster strikes, and then matches volunteer drivers with evacuees once demand is realized. Here we optimize resource planning and evacuation operations under uncertain spatiotemporal demand, and construct a two-stage stochastic mixed-integer program to ensure high demand fulfillment rates. We consider three formulations to improve the number of evacuees served, by minimizing an expected penalty cost, imposing a probabilistic constraint, and enforcing a constraint on the conditional value at risk of the total number of unserved evacuees, respectively. We discuss the benefits and disadvantages of the different risk measures used in the three formulations, given certain carless population sizes and the variety of evacuation modes available. We also develop a heuristic approach to provide quick, dynamic and conservative solutions. We demonstrate the performance of our approaches using five different networks of varying sizes based on regions of Charleston County, South Carolina, an area that experienced a mandatory evacuation order during Hurricane Florence, and utilize real demographic data and hourly traffic count data to estimate the demand distribution.

97 MATHEMATICS AND COMPUTING↗

Forecasting influenza activity using machine-learned mobility map

Human mobility is a primary driver of infectious disease spread. However, existing data is limited in availability, coverage, granularity, and timeliness. Data-driven forecasts of disease dynamics are crucial for decision-making by health officials and private citizens alike. In this work, we focus on a machine-learned anonymized mobility map (hereon referred to as AMM) aggregated over hundreds of millions of smartphones and evaluate its utility in forecasting epidemics. We factor AMM into a metapopulation model to retrospectively forecast influenza in the USA and Australia. We show that the AMM model performs on-par with those based on commuter surveys, which are sparsely available and expensive. We also compare it with gravity and radiation based models of mobility, and find that the radiation model’s performance is quite similar to AMM and commuter flows. Additionally, we demonstrate our model’s ability to predict disease spread even across state boundaries. Our work contributes towards developing timely infectious disease forecasting at a global scale using human mobility datasets expanding their applications in the area of infectious disease epidemiology.

60 APPLIED LIFE SCIENCES↗

A Privacy Preserving Model-Free Optimization and Control Framework for Demand Response from Residential Thermal Loads

We consider the problem of optimizing the cost of procuring electricity for a large collection of homes managed by a load serving entity, by pre-cooling or pre-heating the thermal inertial loads in the homes to avoid procuring power during periods of peak electricity pricing. We would like to accomplish this objective in a completely privacy-preserving and model-free manner, that is, without direct access to the state variables (temperatures or power consumption) or the dynamical models (thermal characteristics) of individual homes, while guaranteeing personal comfort constraints of the consumers. We propose a two-stage optimization and control framework to address this problem. In the first stage, we use a long short-term memory (LSTM) network to predict hourly electricity prices, based on historical pricing data and weather forecasts. Given the hourly price forecast and thermal models of the homes, the problem of designing an optimal power consumption trajectory that minimizes the total electricity procurement cost for the collection of thermal loads can be formulated as a large-scale integer program (with millions of variables) due to the on-off cyclical dynamics of such loads. We provide a simple heuristic relaxation to make this large-scale optimization problem model-free and computationally tractable. In the second stage, we translate the results of this optimization problem into distributed open-loop control laws that can be implemented at individual homes without measuring or estimating their state variables, while simultaneously ensuring consumer comfort constraints. We demonstrate the performance of this approach on a large-scale test case comprising of 500 homes in the Houston area and benchmark its performance against a direct model-based optimization and control solution.

Sivaranjani, S.↗

Streamlining Ocean Dynamics Modeling with Fourier Neural Operators: A Multiobjective Hyperparameter and Architecture Optimization Approach

Training an effective deep learning model to learn ocean processes involves careful choices of various hyperparameters. We leverage DeepHyper’s advanced search algorithms for multiobjective optimization, streamlining the development of neural networks tailored for ocean modeling. The focus is on optimizing Fourier neural operators (FNOs), a data-driven model capable of simulating complex ocean behaviors. Selecting the correct model and tuning the hyperparameters are challenging tasks, requiring much effort to ensure model accuracy. DeepHyper allows efficient exploration of hyperparameters associated with data preprocessing, FNO architecture-related hyperparameters, and various model training strategies. We aim to obtain an optimal set of hyperparameters leading to the most performant model. Moreover, on top of the commonly used mean squared error for model training, we propose adopting the negative anomaly correlation coefficient as the additional loss term to improve model performance and investigate the potential trade-off between the two terms. The numerical experiments show that the optimal set of hyperparameters enhanced model performance in single timestepping forecasting and greatly exceeded the baseline configuration in the autoregressive rollout for long-horizon forecasting up to 30 days. Utilizing DeepHyper, we demonstrate an approach to enhance the use of FNO in ocean dynamics forecasting, offering a scalable solution with improved precision.

97 MATHEMATICS AND COMPUTING↗