Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Real--Time Dynamic Modeling”

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 685 records · Page 38
All sourcesNASA NTRSDOE OSTI
NASA NTRS2009

Dynamic Structural Fault Detection and Identification

Aircraft structures are designed to guarantee safety of flight in some required operational envelope. When the aircraft becomes structurally impaired, safety of flight may not be guaranteed within that previously safe operational envelope. In this case the safe operational envelope must be redefined in-flight and a means to prevent excursion from this new envelope must be implemented. A specific structural failure mode that may result in a reduced safe operating envelope, the exceedance of which could lead to catastrophic structural failure of the aircraft, will be addressed. The goal of the DFEAP program is the detection of this failure mode coupled with flight controls adaptation to limit critical loads in the damaged aircraft structure. The DFEAP program is working with an F/A-18 aircraft model. The composite wing skins are bonded to metallic spars in the wing substructure. Over time, it is possible that this bonding can deteriorate due to fatigue. In this case, the ability of the wing spar to transfer loading between the wing skins is reduced. This failure mode can translate to a reduced allowable compressive strain on the wing skin and could lead to catastrophic wing buckling if load limiting of the wing structure is not applied. The DFEAP program will make use of a simplified wing strain model for the healthy aircraft. The outputs of this model will be compared in real-time to onboard strain measurements at several locations on the aircraft wing. A damage condition is declared at a given location when the strain measurements differ sufficiently from the strain model. Parameter identification of the damaged structure wing strain parameters will be employed to provide load limiting control adaptation for the aircraft. This paper will discuss the simplified strain models used in the implementation and their interaction with the strain sensor measurements. Also discussed will be the damage detection and identification schemes employed and the means by which the damaged aircraft parameters will be used to provide load limiting that keeps the aircraft within the safe operational envelope.

Smith, Timothy↗
NASA NTRS2010

Template Matching Approach to Signal Prediction

A new approach to signal prediction and prognostic assessment of spacecraft health resolves an inherent difficulty in fusing sensor data with simulated data. This technique builds upon previous work that demonstrated the importance of physics-based transient models to accurate prediction of signal dynamics and system performance. While models can greatly improve predictive accuracy, they are difficult to apply in general because of variations in model type, accuracy, or intended purpose. However, virtually any flight project will have at least some modeling capability at its disposal, whether a full-blown simulation, partial physics models, dynamic look-up tables, a brassboard analogue system, or simple hand-driven calculation by a team of experts. Many models can be used to develop a predict, or an estimate of the next day s or next cycle s behavior, which is typically used for planning purposes. The fidelity of a predict varies from one project to another, depending on the complexity of the simulation (i.e. linearized or full differential equations) and the level of detail in anticipated system operation, but typically any predict cannot be adapted to changing conditions or adjusted spacecraft command execution. Applying a predict blindly, without adapting the predict to current conditions, produces mixed results at best, primarily due to mismatches between assumed execution of spacecraft activities and actual times of execution. This results in the predict becoming useless during periods of complicated behavior, exactly when the predict would be most valuable. Each spacecraft operation tends to show up as a transient in the data, and if the transients are misaligned, using the predict can actually harm forecasting performance. To address this problem, the approach here expresses the predict in terms of a baseline function superposed with one or more transient functions. These transients serve as signal templates, which can be relocated in time and space against the signal background. One then has the ability to reconstruct a signal regardless of the precise timing of the transients. During operation, one applies the actual start times of spacecraft activities as they occur, and produces a reconstructed, accurate predict in real-time. This general approach is valid under two important conditions. First, the transients themselves must be time-invariant. Second, the transients must be reasonably consistent with respect to different operating points. Both of these assumptions are generally valid, but in the case of a complicated system with numerous types of overlapping transients, this approach may not be effective. Fortunately, there tend to be few transients in spacecraft telemetry of sensor quantities or low-level health and status information because these signals rarely reflect multiple different types of operation. Furthermore, if the predict is at least reasonably close to actual operation, the shift either in time, or in the operating point when the transient occurs - is likely to be small. The proposed approach considers three ways to recognize a transient. The first and most reliable is to use a different signal that identifies operating mode - often the transient will be correlated to a change in operating mode, and this change can usually be detected positively from discrete signals in spacecraft telemetry. If there is no useful mode signal, the second option is to identify the transient template by hand, and detect the onset of the transient using a curve-fitting approach. Finally, if one elects not to choose by hand, there is an option for automatic selection. This technique has been applied to sensor data from several JPL missions and industrial applications, demonstrating an improvement in accurate prediction of future behavior and early detection of system problems. This technique is applicable to practically any time-varying, quantitative sensor measurement.

Mackey, Ryan↗
NASA NTRS2010

Measuring and Modeling Solar Flares with SDO EVE

The Flare Irradiance Spectral Model (FISM) currently provides estimations of the solar Vacuum Ultra-Violet (VUV; 0.1-190 nm) irradiance from 1947 to present at 1 nm bins and 60 second temporal resolution. The accuracy of FISM is limited to 1nm because the TIMED SEE Level 3 and Level 3A measurement, for which the FISM relationships were based on, were at Inm. With the launch of the Solar Dynamics Observatory (SDO) in Feb 2010, the EUV Variability Experiment (EVE) will provide more accurate, increased spectral resolution (0.1 nm) measurements from 6 nm to 106 nm every 10-seconds and almost 100% duty cycle. EVE also observes additional broadband measurements for 0.1-7 nm, some EUV bands from 16 nm to 40 nm, and Hydrogen Lyman alpha at 121.6 nm. Along with the better spectral resolution of EVE leading to similar spectral resolution improvements in FISM, other significant improvements provide by SDO/EVE measurements will be multi-thermal gradual phase decay, impulsive phase center-to-limb variation characterization, non-flare sub-daily variations, and multi-thermal proxies. SDO/EVE observations of various solar flares with different characteristics will be presented, with the discussion theme focused on how EVE observations will help improve the accuracy of the FISM in the future. The final topic will present the updated FISM website that now has easier access to the FISM results and produce real-time updates that are useful for Space Weather studies.

Chamberlin, Phillip C.↗
NASA NTRS2011

Flight Dynamics Aspects of a Large Civil Tiltrotor Simulation Using Translational Rate Command

An in-depth analysis of a Large Civil Tiltrotor simulation with a Translational Rate Command control law that uses automatic nacelle deflections for longitudinal velocity control and lateral cyclic for lateral velocity control is presented. Results from piloted real-time simulation experiments and offline time and frequency domain analyses are used to investigate the fundamental flight dynamic and control mechanisms of the control law. The baseline Translational Rate Command conferred handling qualities improvements over an attitude command attitude hold control law but in some scenarios there was a tendency to enter PIO. Nacelle actuator rate limiting strongly influenced the PIO tendency and reducing the rate limits degraded the handling qualities further. Counterintuitively, increasing rate limits also led to a worsening of the handling qualities ratings. This led to the identification of a nacelle rate to rotor longitudinal flapping coupling effect that induced undesired pitching motions proportional to the allowable amount of nacelle rate. A modification that applied a counteracting amount of longitudinal cyclic proportional to the nacelle rate significantly improved the handling qualities. The lateral axis of the Translational Rate Command conferred Level 1 handling qualities in a Lateral Reposition maneuver. Analysis of the influence of the modeling fidelity on the lateral flapping angles is presented. It is showed that the linear modeling approximation is likely to have under-predicted the side-force and therefore under-predicted the lateral flapping at velocities above 15 ft/s. However, at lower velocities, and therefore more weakly influenced by the side force modeling, the accelerations that the control law commands also significantly influenced the peak levels of lateral flapping achieved.

Lawrence, Ben↗
NASA NTRS2011

Radiation Belt Environment Model: Application to Space Weather and Beyond

Understanding the dynamics and variability of the radiation belts are of great scientific and space weather significance. A physics-based Radiation Belt Environment (RBE) model has been developed to simulate and predict the radiation particle intensities. The RBE model considers the influences from the solar wind, ring current and plasmasphere. It takes into account the particle drift in realistic, time-varying magnetic and electric field, and includes diffusive effects of wave-particle interactions with various wave modes in the magnetosphere. The RBE model has been used to perform event studies and real-time prediction of energetic electron fluxes. In this talk, we will describe the RBE model equation, inputs and capabilities. Recent advancement in space weather application and artificial radiation belt study will be discussed as well.

Fok, Mei-Ching H.↗
NASA NTRS2004

Analysis of SSEM Sensor Data Using BEAM

A report describes analysis of space shuttle main engine (SSME) sensor data using Beacon-based Exception Analysis for Multimissions (BEAM) [NASA Tech Briefs articles, the two most relevant being Beacon-Based Exception Analysis for Multimissions (NPO- 20827), Vol. 26, No.9 (September 2002), page 32 and Integrated Formulation of Beacon-Based Exception Analysis for Multimissions (NPO- 21126), Vol. 27, No. 3 (March 2003), page 74] for automated detection of anomalies. A specific implementation of BEAM, using the Dynamical Invariant Anomaly Detector (DIAD), is used to find anomalies commonly encountered during SSME ground test firings. The DIAD detects anomalies by computing coefficients of an autoregressive model and comparing them to expected values extracted from previous training data. The DIAD was trained using nominal SSME test-firing data. DIAD detected all the major anomalies including blade failures, frozen sense lines, and deactivated sensors. The DIAD was particularly sensitive to anomalies caused by faulty sensors and unexpected transients. The system offers a way to reduce SSME analysis time and cost by automatically indicating specific time periods, signals, and features contributing to each anomaly. The software described here executes on a standard workstation and delivers analyses in seconds, a computing time comparable to or faster than the test duration itself, offering potential for real-time analysis.

Zak, Michail↗
NASA NTRS2011

Evaluating the Application of Multi-Satellite Observations in Hydrologic Modeling

When monitoring local or regional hydrosphere dynamics for applications such as agricultural productivity or drought and flooding events, it is necessary to have accurate, high-resolution estimates of terrestrial water and energy storages. Though in-situ observations provide reliable estimates of hydrologic states and fluxes, they are only capable of accurately capturing the dynamics at relatively discrete points in space and time, which makes them inadequate for characterizing the variability of the water budget across scales. In contrast, satellite-based remote sensing is ideal for providing observations of hydrological states and fluxes because it provides spatially-distributed observations at spatial and temporal scales required for regional land surface process modeling. Due to the continued progress in algorithm development and emerging satellite technology, we now have near-real time monitoring of several components of the water cycle including precipitation, soil moisture, lake and river height, terrestrial water storage, snow cover, and evapotranspiration. As these data become more readily available, their application to hydrologic modeling is becoming more common, however there remains little consensus on the most appropriate method for optimal integration and evaluation in regard to hydrological applications. Here we present two case studies operationally applying several remotely sensed products from AMSR-E, GRACE, and MODIS and discuss assimilation strategies, ease of integration and interpretation, and methods for quantifying the success of the application methodology.

Bolten, John↗
NASA NTRS2011

Dynamic Weather Routes: A Weather Avoidance Concept for Trajectory-Based Operations

The integration of convective weather modeling with trajectory automation for conflict detection, trial planning, direct routing, and auto resolution has uncovered a concept that could help controllers, dispatchers, and pilots identify improved weather routes that result in significant savings in flying time and fuel burn. Trajectory automation continuously and automatically monitors aircraft in flight to find those that could potentially benefit from improved weather reroutes. Controllers, dispatchers, and pilots then evaluate reroute options to assess their suitability given current weather and traffic. In today's operations aircraft fly convective weather avoidance routes that were implemented often hours before aircraft approach the weather and automation does not exist to automatically monitor traffic to find improved weather routes that open up due to changing weather conditions. The automation concept runs in real-time and employs two keysteps. First, a direct routing algorithm automatically identifies flights with large dog legs in their routes and therefore potentially large savings in flying time. These are common - and usually necessary - during convective weather operations and analysis of Fort Worth Center traffic shows many aircraft with short cuts that indicate savings on the order of 10 flying minutes. The second and most critical step is to apply trajectory automation with weather modeling to determine what savings could be achieved by modifying the direct route such that it avoids weather and traffic and is acceptable to controllers and flight crews. Initial analysis of Fort Worth Center traffic suggests a savings of roughly 50% of the direct route savings could be achievable.The core concept is to apply trajectory automation with convective weather modeling in real time to identify a reroute that is free of weather and traffic conflicts and indicates enough time and fuel savings to be considered. The concept is interoperable with today's integrated FMS/datalink. Auxiliary(lat/long) waypoints define a minimum delay reroute between current position and a downstream capture fix beyond the weather. These auxiliary waypoints can be uplinked to equipped aircraft and auto-loaded into the FMS. Alternatively, for unequipped aircraft, auxiliary waypoints can be replaced by nearby named fixes, but this could reduce potential savings. The presentation includes an overview of the automation approach and focuses on several cases in terms of potential savings, reroute complexity, best auxiliary waypoint solution vs. named fix solution, and other metrics.

McNally, B. David↗
NASA NTRS2012

Near real-time traffic routing

A near real-time physical transportation network routing system comprising: a traffic simulation computing grid and a dynamic traffic routing service computing grid. The traffic simulator produces traffic network travel time predictions for a physical transportation network using a traffic simulation model and common input data. The physical transportation network is divided into a multiple sections. Each section has a primary zone and a buffer zone. The traffic simulation computing grid includes multiple of traffic simulation computing nodes. The common input data includes static network characteristics, an origin-destination data table, dynamic traffic information data and historical traffic data. The dynamic traffic routing service computing grid includes multiple dynamic traffic routing computing nodes and generates traffic route(s) using the traffic network travel time predictions.

Yang, Chaowei↗
NASA NTRS2012

Elastic Model Transitions Using Quadratic Inequality Constrained Least Squares

A technique is presented for initializing multiple discrete finite element model (FEM) mode sets for certain types of flight dynamics formulations that rely on superposition of orthogonal modes for modeling the elastic response. Such approaches are commonly used for modeling launch vehicle dynamics, and challenges arise due to the rapidly time-varying nature of the rigid-body and elastic characteristics. By way of an energy argument, a quadratic inequality constrained least squares (LSQI) algorithm is employed to e ect a smooth transition from one set of FEM eigenvectors to another with no requirement that the models be of similar dimension or that the eigenvectors be correlated in any particular way. The physically unrealistic and controversial method of eigenvector interpolation is completely avoided, and the discrete solution approximates that of the continuously varying system. The real-time computational burden is shown to be negligible due to convenient features of the solution method. Simulation results are presented, and applications to staging and other discontinuous mass changes are discussed

Orr, Jeb S.↗
NASA NTRS2004

Sensor Web Dynamic Measurement Techniques and Adaptive Observing Strategies

Sensor Web observing systems may have the potential to significantly improve our ability to monitor, understand, and predict the evolution of rapidly evolving, transient, or variable environmental features and events. This improvement will come about by integrating novel data collection techniques, new or improved instruments, emerging communications technologies and protocols, sensor mark-up languages, and interoperable planning and scheduling systems. In contrast to today's observing systems, "event-driven" sensor webs will synthesize real- or near-real time measurements and information from other platforms and then react by reconfiguring the platforms and instruments to invoke new measurement modes and adaptive observation strategies. Similarly, "model-driven" sensor webs will utilize environmental prediction models to initiate targeted sensor measurements or to use a new observing strategy. The sensor web concept contrasts with today's data collection techniques and observing system operations concepts where independent measurements are made by remote sensing and in situ platforms that do not share, and therefore cannot act upon, potentially useful complementary sensor measurement data and platform state information. This presentation describes NASA's view of event-driven and model-driven Sensor Webs and highlights several research and development activities at the Goddard Space Flight Center.

Talabac, Stephen J.↗
NASA NTRS2014

Development and Implementation of Dynamic Scripts to Execute Cycled WRF/GSI Forecasts

Automating the coupling of data assimilation (DA) and modeling systems is a unique challenge in the numerical weather prediction (NWP) research community. In recent years, the Development Testbed Center (DTC) has released well-documented tools such as the Weather Research and Forecasting (WRF) model and the Gridpoint Statistical Interpolation (GSI) DA system that can be easily downloaded, installed, and run by researchers on their local systems. However, developing a coupled system in which the various preprocessing, DA, model, and postprocessing capabilities are all integrated can be labor-intensive if one has little experience with any of these individual systems. Additionally, operational modeling entities generally have specific coupling methodologies that can take time to understand and develop code to implement properly. To better enable collaborating researchers to perform modeling and DA experiments with GSI, the Short-term Prediction Research and Transition (SPoRT) Center has developed a set of Perl scripts that couple GSI and WRF in a cycling methodology consistent with the use of real-time, regional observation data from the National Centers for Environmental Prediction (NCEP)/Environmental Modeling Center (EMC). Because Perl is open source, the code can be easily downloaded and executed regardless of the user's native shell environment. This paper will provide a description of this open-source code and descriptions of a number of the use cases that have been performed by SPoRT collaborators using the scripts on different computing systems.

Zavodsky, Bradley↗
NASA NTRS2014

Initial Evaluations of LoC Prediction Algorithms Using the NASA Vertical Motion Simulator

Flying near the edge of the safe operating envelope is an inherently unsafe proposition. Edge of the envelope here implies that small changes or disturbances in system state or system dynamics can take the system out of the safe envelope in a short time and could result in loss-of-control events. This study evaluated approaches to predicting loss-of-control safety margins as the aircraft gets closer to the edge of the safe operating envelope. The goal of the approach is to provide the pilot aural, visual, and tactile cues focused on maintaining the pilot's control action within predicted loss-of-control boundaries. Our predictive architecture combines quantitative loss-of-control boundaries, an adaptive prediction method to estimate in real-time Markov model parameters and associated stability margins, and a real-time data-based predictive control margins estimation algorithm. The combined architecture is applied to a nonlinear transport class aircraft. Evaluations of various feedback cues using both test and commercial pilots in the NASA Ames Vertical Motion-base Simulator (VMS) were conducted in the summer of 2013. The paper presents results of this evaluation focused on effectiveness of these approaches and the cues in preventing the pilots from entering a loss-of-control event.

Pilot Cue↗
NASA NTRS2013

Real-Time Minimization of Tracking Error for Aircraft Systems

This technology presents a novel, stable, discrete-time adaptive law for flight control in a Direct adaptive control (DAC) framework. Where errors are not present, the original control design has been tuned for optimal performance. Adaptive control works towards achieving nominal performance whenever the design has modeling uncertainties/errors or when the vehicle suffers substantial flight configuration change. The baseline controller uses dynamic inversion with proportional-integral augmentation. On-line adaptation of this control law is achieved by providing a parameterized augmentation signal to a dynamic inversion block. The parameters of this augmentation signal are updated to achieve the nominal desired error dynamics. If the system senses that at least one aircraft component is experiencing an excursion and the return of this component value toward its reference value is not proceeding according to the expected controller characteristics, then the neural network (NN) modeling of aircraft operation may be changed.

Garud, Sumedha↗
NASA NTRS2013

File Specification for GEOS-5 FP (Forward Processing)

The GEOS-5 FP Atmospheric Data Assimilation System (GEOS-5 ADAS) uses an analysis developed jointly with NOAA's National Centers for Environmental Prediction (NCEP), which allows the Global Modeling and Assimilation Office (GMAO) to take advantage of the developments at NCEP and the Joint Center for Satellite Data Assimilation (JCSDA). The GEOS-5 AGCM uses the finite-volume dynamics (Lin, 2004) integrated with various physics packages (e.g, Bacmeister et al., 2006), under the Earth System Modeling Framework (ESMF) including the Catchment Land Surface Model (CLSM) (e.g., Koster et al., 2000). The GSI analysis is a three-dimensional variational (3DVar) analysis applied in grid-point space to facilitate the implementation of anisotropic, inhomogeneous covariances (e.g., Wu et al., 2002; Derber et al., 2003). The GSI implementation for GEOS-5 FP incorporates a set of recursive filters that produce approximately Gaussian smoothing kernels and isotropic correlation functions. The GEOS-5 ADAS is documented in Rienecker et al. (2008). More recent updates to the model are presented in Molod et al. (2011). The GEOS-5 system actively assimilates roughly 2 × 10(exp 6) observations for each analysis, including about 7.5 × 10(exp 5) AIRS radiance data. The input stream is roughly twice this volume, but because of the large volume, the data are thinned commensurate with the analysis grid to reduce the computational burden. Data are also rejected from the analysis through quality control procedures designed to detect, for example, the presence of cloud. To minimize the spurious periodic perturbations of the analysis, GEOS-5 FP uses the Incremental Analysis Update (IAU) technique developed by Bloom et al. (1996). More details of this procedure are given in Appendix A. The assimilation is performed at a horizontal resolution of 0.3125-degree longitude by 0.25- degree latitude and at 72 levels, extending to 0.01 hPa. All products are generated at the native resolution of the horizontal grid. The majority of data products are time-averaged, but four instantaneous products are also available. Hourly data intervals are used for two-dimensional products, while 3-hourly intervals are used for three-dimensional products. These may be on the model's native 72-layer vertical grid or at 42 pressure surfaces extending to 0.1 hPa. This document describes the gridded output files produced by the GMAO near real-time operational FP, using the most recent version of the GEOS-5 assimilation system. Additional details about variables listed in this file specification can be found in a separate document, the GEOS-5 File Specification Variable Definition Glossary. Documentation about the current access methods for products described in this document can be found on the GMAO products page: http://gmao.gsfc.nasa.gov/products/.

GSI↗
NASA NTRS2013

File Specification for GEOS-5 FP-IT (Forward Processing for Instrument Teams)

The GEOS-5 FP-IT Atmospheric Data Assimilation System (GEOS-5 ADAS) uses an analysis developed jointly with NOAA's National Centers for Environmental Prediction (NCEP), which allows the Global Modeling and Assimilation Office (GMAO) to take advantage of the developments at NCEP and the Joint Center for Satellite Data Assimilation (JCSDA). The GEOS-5 AGCM uses the finite-volume dynamics (Lin, 2004) integrated with various physics packages (e.g, Bacmeister et al., 2006), under the Earth System Modeling Framework (ESMF) including the Catchment Land Surface Model (CLSM) (e.g., Koster et al., 2000). The GSI analysis is a three-dimensional variational (3DVar) analysis applied in grid-point space to facilitate the implementation of anisotropic, inhomogeneous covariances (e.g., Wu et al., 2002; Derber et al., 2003). The GSI implementation for GEOS-5 FP-IT incorporates a set of recursive filters that produce approximately Gaussian smoothing kernels and isotropic correlation functions. The GEOS-5 ADAS is documented in Rienecker et al. (2008). More recent updates to the model are presented in Molod et al. (2011). The GEOS-5 system actively assimilates roughly 2 × 10(exp 6) observations for each analysis, including about 7.5 × 10(exp 5) AIRS radiance data. The input stream is roughly twice this volume, but because of the large volume, the data are thinned commensurate with the analysis grid to reduce the computational burden. Data are also rejected from the analysis through quality control procedures designed to detect, for example, the presence of cloud. To minimize the spurious periodic perturbations of the analysis, GEOS-5 FP-IT uses the Incremental Analysis Update (IAU) technique developed by Bloom et al. (1996). More details of this procedure are given in Appendix A. The analysis is performed at a horizontal resolution of 0.625-degree longitude by 0.5-degree latitude and at 72 levels, extending to 0.01 hPa. All products are generated at the native resolution of the horizontal grid. The majority of data products are time-averaged, but four instantaneous products are also available. Hourly data intervals are used for two-dimensional products, while 3-hourly intervals are used for three-dimensional products. These may be on the model's native 72-layer vertical grid or at 42 pressure surfaces extending to 0.1 hPa. This document describes the gridded output files produced by the GMAO near real-time operational GEOS-5 FP-IT processing in support of the EOS instrument teams. Additional details about variables listed in this file specification can be found in a separate document, the GEOS-5 File Specification Variable Definition Glossary.

3D Var↗
DOE OSTI2025

Nature-GL: A Revolutionary Learning Paradigm Unleashing Nature’s Power in Real-World Spatial-Temporal Graph Learning

Spatial-Temporal Graph Learning (ST-GL) is a prominent research area due to its unique capability to effectively learn real-world graphs. Applications of ST-GL pose stringent and various demands on not only real-time inference with low energy cost and high ac- curacy but also fast training. Unfortunately, as Moore’s Law approaches its limits and ST-GL model complexity drastically grows, the gap between digital hardware’s computational power and ST- GL application demands is widening. In response, this paper introduces Nature-GL, a nature-powered graph learning paradigm that exploits the principle of entropy increase to advance graph learning. In particular, Nature-GL transforms both the training and inference of real-valued ST-GL into electron-speed natural anneal- ing processes of a parameterized dynamical system that represents the target graphs. Experimental results across four real-world ap- plications with six datasets demonstrate that Nature-GL achieves orders-of-magnitude speedups in both training and inference, delivering higher accuracy compared to Graph Neural Networks.

Liu, Chuan [University of Rochester]↗
DOE OSTI2025

Predicting Initial Trans-Membrane Pressure for Optimized Operations in UF Unit Using Random Forest

With the growing scarcity of freshwater, innovative process design mechanisms like Reverse Osmosis (RO) are increasingly gaining attention among water treatment utilities to address the rising demand. Ensuring reliable water production necessitates efficient resource utilization, minimizing downtime in (ultra-filtration) UF systems. Recent advancements in machine learning (ML) have enabled the development of accurate data-driven models for Model Predictive Control (MPC), often requiring minimal prior knowledge of underlying physical processes. In this study, we present predictive regression models based on Random Forest (RF) and Auto-Regressive (AR) approaches to forecast the initial Trans-Membrane Pressure (TMP) for each filtration cycle in data generated by Direct Potable Reuse (DPR) systems. The proposed RF-based model demonstrates superior performance compared to baseline methods, including historical mean, Last Observation Carried Forward (LOCF), and naïve AR models, across various forecasting horizons in terms of root mean square error (RMSE) metric. To evaluate how different classes of process variables contribute to TMP dynamics over time, we examine the feature importance of independent covariates across multiple forecast horizons. This analysis provides insight into the temporal relevance of operational and sensor-derived features, guiding control and monitoring strategies. Additionally, the impact of hyperparameter tuning on TMP prediction performance is studied for both direct and recursive RF modelling approaches across increasing forecast horizons. Accurate prediction of initial TMP is critical for optimizing RO operations, as it enables the development of robust modelling frameworks by accurately estimating membrane fouling trends, thereby enhancing process efficiency and long-term reliability. The demonstrated efficacy of the RF-based approach highlights its potential as a tool for real-time decision-making in water treatment systems, paving the way for advanced process optimization and sustainable water resource management.

Mukherjee, Subrata [ORNL] (ORCID:0000000309930338)↗