Search NASA⌕ Search

SEARCH · Search NASA

Results for “power flow initialization”

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 OSTI2024

Newton-Raphson AC Power Flow Convergence Based on Deep Learning Initialization and Homotopy Continuation

Power flow forms the basis of many power system studies. With the increased penetration of renewable energy, grid planners tend to perform multiple power flow simulations under various operating conditions and not just selected snapshots at peak or light load conditions. Getting a converged AC power flow (ACPF) case remains a significant challenge for grid planners especially in large power grid networks. This paper proposes a two-stage approach to improve Newton-Raphson ACPF convergence and was applied to a 6102 bus Electric Reliability Council of Texas (ERCOT) system. The first stage utilizes a deep learning-based initializer with data re-training. Here a deep neural network (DNN) initializer is developed to provide better initial voltage magnitude and angle guesses to aid in power flow convergence. This is because Newton-Raphson ACPF is quite sensitive to the initial conditions and bad initialization could lead to divergence. The DNN initializer includes a data re-training framework that improves the initializer's performance when faced with limited training data. The DNN initializer successfully solved 3,285 cases out of 3,899 non-converging dispatch and performed better than random forest and DC power flow initialization methods. ACPF cases not solved in this first stage are then passed through a hot-starting algorithm based on homotopy continuation with switched shunt control. The hot-starting algorithm successfully converged 416 cases out of the remaining 614 non-converging ACPF dispatch. In conclusion, the combined two-stage approach achieved a 94.9% success rate, by converging a total of 3,701 cases out of the initial 3,899 unsolved cases.

Deep learning↗
DOE OSTI2024

A Machine Learning Initializer for Newton-Raphson AC Power Flow Convergence

Power flow computations are fundamental to many power system studies. Obtaining a converged power flow case is not a trivial task especially in large power grids due to the non-linear nature of the power flow equations. One key challenge is that the widely used Newton based power flow methods are sensitive to the initial voltage magnitude and angle estimates, and a bad initial estimate would lead to non-convergence. This paper addresses this challenge by developing a random-forest (RF) machine learning model to provide better initial voltage magnitude and angle estimates towards achieving power flow convergence. This method was implemented on a real ERCOT 6102 bus system under various operating conditions. By providing better Newton-Raphson initialization, the RF model precipitated the solution of 2,106 cases out of 3,899 non-converging dispatches. These cases could not be solved from flat start or by initialization with the voltage solution of a reference case. Finally, results obtained from the RF initializer performed better when compared with DC power flow initialization, Linear regression, and Decision Trees.

random forest↗
DOE OSTI2024

Taylor-Expansion-Based Robust Power Flow in Unbalanced Distribution Systems: A Hybrid Data-Aided Method

Traditional power flow methods often adopt certain assumptions designed for passive balanced distribution systems, thus lacking practicality for unbalanced operation. moreover, their computation accuracy and efficiency are heavily subject to unknown errors and bad data in measurements or prediction data of distributed energy resources (ders). to address these issues, this paper proposes a hybrid data-aided robust power flow algorithm in unbalanced distribution systems, which combines taylor series expansion knowledge with a data-driven regression technique. the proposed method initiates a linearization power flow model to derive an explicitly analytical solution by modified taylor expansion. to mitigate the approximation loss that surges due to the der integration and bad data, we further develop a data-aided robust support vector regression approach to estimate the errors efficiently. comparative analysis in the 13-bus and 123-bus ieee unbalanced feeders shows that the proposed hybrid algorithm achieves superior computational efficiency, with guaranteed accuracy and robustness against outliers.

data-driven↗
DOE OSTI2025

PowerModel-AI: A First On-the-Fly Machine-Learning Predictor for AC Power Flow Solutions

The real-time creation of machine-learning models via active or on-the-fly learning has attracted considerable interest across various scientific and engineering disciplines. These algorithms enable machines to build models autonomously while remaining operational. Through a series of query strategies, the machine can evaluate whether newly encountered data fall outside the scope of the existing training set. In this study, we introduce PowerModel-AI, an end-to-end machine learning software designed to accurately predict AC power flow solutions. We present detailed justifications for our model design choices and demonstrate that selecting the right input features effectively captures load flow decoupling inherent in power flow equations. Our approach incorporates on-the-fly learning, where power flow calculations are initiated only when the machine detects a need to improve the dataset in regions where the model’s suboptimal performance is based on specific criteria. Otherwise, the existing model is used for power flow predictions. This study includes analyses of five Texas A&M synthetic power grid cases, encompassing the 14-, 30-, 37-, 200-, and 500-bus systems. The training and test datasets were generated using PowerModels.jl, an open-source power flow solver/optimizer developed at Los Alamos National Laboratory, NM, USA.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Acceleration of Power System Dynamic Simulations Using a Deep Equilibrium Layer and Neural ODE Surrogate

The dominant paradigm for power system dynamic simulation is to build system-level simulations by combining physics-based models of individual components. The sheer size of the system along with the rapid integration of inverter-based resources exacerbates the computational burden of running time domain simulations. Here, in this paper, we propose a data-driven surrogate model based on implicit machine learningspecifically deep equilibrium layers and neural ordinary differential equationsto learn a reduced order model of a portion of the full underlying system. The data-driven surrogate achieves similar accuracy and reduction in simulation time compared to a physics-based surrogate, without the constraint of requiring detailed knowledge of the underlying dynamic models. This work also establishes key requirements needed to integrate the surrogate into existing simulation workflows; the proposed surrogate is initialized to a steady state operating point that matches the power flow solution by design.

Neural ordinary differential equations↗
DOE OSTI2025

Machine learning modeling and model predictive control of a closed-circuit reverse osmosis system

Closed-circuit reverse osmosis (CCRO) offers a flexible and energy-efficient alternative to conventional reverse osmosis by operating in a semi-batch mode that recycles brine, enabling higher recovery rates and reduced specific energy consumption (SEC). However, developing accurate, system-level dynamic models for CCRO remains challenging due to its nonlinear, multi-phase operation and sensitivity to variable feed water conditions. Traditional modeling approaches, such as NARMAX (nonlinear autoregressive moving average with exogenous inputs), often struggle to generalize across varying inlet feed concentrations, necessitating frequent parameter re-estimation and limiting their utility for real-time control applications. To address these limitations, we developed a long short-term memory (LSTM) neural network model trained on an extensive experimental data set from a CCRO pilot plant. The model accepts three inputs, feed flow rate, recirculation flow rate, and initial feed conductivity, and predicts three key outputs: reject conductivity, feed pump power draw, and recirculation pump power draw. We validated the LSTM model against experimental data, demonstrating its ability to distinguish between different feed conductivities and adapt to variable flow rates. Subsequently, we incorporated the LSTM model within a nonlinear model predictive control (MPC) scheme and conducted closed-loop simulations to optimize the integrated SEC (iSEC). In conclusion, the results project up to a 6% reduction in iSEC by using MPC to optimize performance over the entire experiment duration, without requiring any random excitation for data collection or parameter re-estimation.

Desalination↗
DOE OSTI2023

Probing initial baryon stopping and equation of state with rapidity-dependent directed flow of identified particles

In this work, using a (3 + 1)-dimensional hybrid framework with parametric initial conditions, we study the rapidity-dependent directed flow $v_1 (y)$ of identified particles, including pions, kaons, protons, and lambdas in heavy-ion collisions. Cases involving Au + Au collisions are considered, performed at $\sqrt {^SNN}$ ranging from 7.7 to 200 GeV. The dynamics in the beam direction is constrained using the measured pseudo-rapidity distribution of charged particles and the net proton rapidity distribution. Within this framework, the directed flow of mesons is driven by the sideward pressure gradient from the tilted source, and that of baryons mainly due to the initial asymmetric baryon distribution with respect to the beam axis driven by the transverse expansion. Our approach successfully reproduces the rapidity- and beam energy-dependence of $v_1$ for both mesons and baryons. We find that the $v_1 (y)$ of baryons has strong constraining power on the initial baryon stopping, and together with that of mesons, the directed flow probes the equation of state of the dense nuclear matter at finite chemical potentials.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

Constraints on initial baryon stopping and equation of state from directed flow

Our investigation focuses on the rapidity-dependent directed flow, v 1 (y), of identified hadrons in Au+Au collisions across a broad range of √S NN from 7.7 to 200 GeV. Employing a (3+1)-dimensional hybrid framework, our study successfully reproduces the characteristic features of the measured v 1 (y) for both mesons and baryons across the considered beam energies. Notably, our analysis reveals the constraining power of baryonic v 1 (y) on the initial baryon stopping mechanism. Together with mesonic v 1 (y), the directed flow serves as a crucial tool for probing the equation of state governing dense nuclear matter at finite chemical potentials.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Parameter Study of the Running-In Process for the Generic Pebble Bed Reactor (GPBR200)

The run-in period of a pebble bed reactor is complex and difficult to model given significant heterogeneity in core composition, power, and temperature. While it is understood that the initial composition of the core should eventually result in the same equilibrium core composition, the approach to equilibrium can vary significantly depending on factors such as start-up fuel enrichment and power ramp rate. To explore this, a high-fidelity model of the General Pebble Bed Reactor was used to vary power ramp schemes and start-up core compositions. It was found that both the initial core composition and power ramp rate had a significant impact on the flow rate of pebbles during early time steps, with higher ramp rates and low enrichment resulting in non-physical flow rates. Power ramp rate alone was found to dictated maximum pebble power peaking observed during the run-in process, with higher ramp rates resulting in greater peak pebble powers. Start-up fuel enrichment and power ramp were found to both impact total fuel consumption, although the impact of start-up fuel enrichment was generally secondary to ramp rate.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

Data-Efficient Strategies for Probabilistic Voltage Envelopes under Network Contingencies

This work presents an efficient data-driven method to construct probabilistic voltage envelopes (PVE) using power flow learning in grids with network contingencies. First, a network-aware Gaussian process (GP) termed Vertex-Degree Kernel (VDK-GP), developed in prior work, is used to estimate voltage–power functions for a few network configurations. The paper introduces a novel multi-task vertex degree kernel (MT-VDK) that amalgamates the learned VDK-GPs to determine power flows for unseen networks, with a significant reduction in the computational complexity and hyperparameter requirements compared to alternate approaches. Simulations on the IEEE 30-Bus network demonstrate the retention and transfer of power flow knowledge in both N-1 and N-2 contingency scenarios. The MT-VDK-GP approach achieves over 50 % reduction in mean prediction error for novel N-1 contingency network configurations in low training data regimes (50–250 samples) over VDK-GP. Additionally, MT-VDK-GP outperforms a hyper-parameter based transfer learning approach in over 75 % of N-2 contingency network structures, even without historical N-2 outage data. Furthermore, the proposed method demonstrates the ability to achieve PVEs using sixteen times fewer power flow solutions compared to Monte-Carlo sampling-based methods.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Development of a Distribution Optimal Power Flow Federate for Open-Source OEDI-SI Platform

Increasing numbers of distributed generators in the electric power distribution networks require developing a control strategy to optimize solutions in real time. Linearized optimal distribution flow development has seen growth and acceptance in the distribution systems literature for efficiently modeling the \glspl{opf} for distribution systems. This paper examines the implementation and integration procedure for linearized optimal distribution flow federate to \gls{oedisi} platform. Specifically, we discuss i) the usage of the \gls{oedisi} platform, ii) obtaining a tractable solution using developed \gls{opf} federate, and iii) validation of solutions and bench-marking the \gls{oedisi} platform with developed \gls{opf} federate using OpenDSS. In brief, we demonstrate how a general linearized optimal distribution flow federate can be developed and integrated with a co-simulation environment to mimic real-world examples. The efficacy of the proposed method is demonstrated using the IEEE 123-bus test system under different scenarios to obtain a tractable solution and compare its results.

Sadnan, Rabayet↗
DOE OSTI2024

Flow Sensor Test Article (F-STAr) - Water Testing and Commissioning Report for FY2024

Argonne National Laboratory’s Mechanism Engineering Test Loop (METL) facility is developing new experimental testing capabilities with the Flow Sensor Test Article (F-STAr). F-STAr’s mission is to provide high flowrate, sodium submersible testing capabilities to support the development of sodium cooled nuclear fission reactors. Figure 1 shows a model of F-STAr (left) and a photo of the partially assembled test article (right) with several of the main components labeled. The system currently includes a large pump with desired flowrates of up to 100 GPM; an immersion heater with a maximum power of 5 kW; an immersion cooler with an estimated maximum cooling power of 2 kW; and an experimental test section to support sub-test articles. Initially, F-STAr will be setup for testing flow sensors, specifically Eddy Current Flow Sensors (ECFS). However, the system can be reconfigured to support a wide range of testing needs. For example, F-STAr could be setup to complete testing of hydraulic components, heat exchangers, hydrodynamic bearings, seals, and more. This report will provide an update on the development of F-STAr with a focus on the water testing and commissioning activities during Fiscal Year 2024 (FY24). First, the report will focus on describing the pump development. Specifically, updates will be provided on troubleshooting the pump’s excessive vibrational issues. In addition to updates on the pump, pressure-flowrate performance testing in water will be described. Furthermore, updates will be provided on a pump shaft bushing development as well as rigging procedures. Future work and paths forward for F-STAr will be described. Finally, a short summary on the ECFS development will be described at the end of the report.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Influence of Oxygen Flow and Stoichiometry on Optical Properties and Damage Resistance of Hafnium Oxide Thin Films

Hafnium oxide (HfO 2 ) is predominantly used as a high-index material in multi-layer dielectric coatings for high-peak- and high-average-power lasers, but laser damage often initiates within the HfO 2 layers despite their wide bandgap. Oxygen deficiency during deposition can introduce vacancy-related sub-bandgap states and absorptive defects, lowering damage resistance. This study investigates how oxygen flow during HfO 2 deposition with ion beam sputtering (IBS) affects its stoichiometry, defect formation, and nanosecond laser-induced damage threshold (LIDT) and whether single-layer trends predict multilayer performance. Single layers were deposited at varying oxygen flows, characterized for optical and structural properties, and tested for the LIDT at 1064 nm and 355 nm. Increasing oxygen flow drove the layer toward near-stoichiometric HfO 2 , reduced the refractive index, and altered the density of surface pinhole-like features. The single-layer LIDT at 355 nm increased with oxygen, whereas the 1064 nm LIDT was comparatively less sensitive to oxygen flow, consistent with the wavelength-dependent roles of absorptive precursors and microstructural defects. In contrast, a HfO 2 -based high-reflector (HR) showed a higher LIDT at lower oxygen flow, indicating that the family of damage precursors changes between single layers and multilayers; in stacks, structural properties such as stress, gas entrapment and thermal dissipation may outweigh the isolated absorptive defects found in single layers. These results demonstrate that the optimal oxygen flow condition depends on both LIDT wavelength and film architecture. We identified, for single layers, a 15–35 sccm window for maximizing the 1064 nm LIDT and a high-flow optimum (45 sccm) for the 355 nm LIDT and, for 355 nm HR stacks, a distinct lower-flow regime (~10 sccm).

Optics and optical instruments↗
DOE OSTI2024

A Pseudoreversible Normalizing Flow for Stochastic Dynamical Systems with Various Initial Distributions

Here, we present a pseudoreversible normalizing flow method for efficiently generating samples of the state of a stochastic differential equation (SDE) with various initial distributions. The primary objective is to construct an accurate and efficient sampler that can be used as a surrogate model for computationally expensive numerical integration of SDEs, such as those employed in particle simulation. After training, the normalizing flow model can directly generate samples of the SDE’s final state without simulating trajectories. The existing normalizing flow model for SDEs depends on the initial distribution, meaning the model needs to be retrained when the initial distribution changes. The main novelty of our normalizing flow model is that it can learn the conditional distribution of the state, i.e., the distribution of the final state conditional on any initial state, such that the model only needs to be trained once and the trained model can be used to handle various initial distributions. This feature can provide a significant computational saving in studies of how the final state varies with the initial distribution. Additionally, we propose to use a pseudoreversible network architecture to define the normalizing flow model, which has sufficient expressive power and training efficiency for a variety of SDEs in science and engineering, e.g., in particle physics. We provide a rigorous convergence analysis of the pseudoreversible normalizing flow model to the target probability density function in the Kullback–Leibler divergence metric. Numerical experiments are provided to demonstrate the effectiveness of the proposed normalizing flow model.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Hydropower Market Game

SF-25-121 The Hydropower Market Game is an interactive educational software developed to teach fundamental concepts of hydropower generation, operations, and its role within electricity markets. Developed in Python and powered by Pygame, the game combines narrative-driven learning with progressive, hands-on levels where players explore topics such as the relationship between water flow and power generation, dam operations, pumped-storage systems, market interactions, and environmental constraints.Designed to support outreach and workforce development initiatives, the game provides an intuitive and engaging way to build energy literacy and raise awareness of hydropower’s contribution to the nation’s energy system.

Ploussard, Quentin [Argonne National Laboratory (A↗
DOE OSTI2024

Cascading Transformer Failure Probability Model Under Geomagnetic Disturbances

This paper develops a probabilistic model to assess the cascading failure of transformers in an electric power grid experiencing geomagnetic disturbances caused by a solar storm. We propose a model in which the probability of failure is a function of the intensity of the solar storm, the physical properties of the transformer, the geographical location of the transformer, and the flow of electrical power. We demonstrate the proposed model using the IEEE 14-bus system and several notional solar storms. The model quickly computes the initial and cascading failure probabilities of the transformers in the system as a first step towards quantifying the risks posed by future solar storms.

Shukla, Pratishtha↗
DOE OSTI2024

System design and analysis of thermal power dispatch systems for boiling water reactors

Nuclear power plants are crucial to meeting net zero emission goals and achieving energy sustainability. Integrating these plants with clean energy technologies such as high-temperature steam electrolysis (HTSE) may improve the efficiency and economic competitiveness of these plants. Here, the current study investigates the design and operation of a thermal power dispatch (TPD) system for coupling boiling water reactors (BWRs) to HTSE plants. The TPD system extracts a portion of the steam from the reactor’s main steam line and transfers its thermal energy to an HTSE plant through a power transport loop. A TPD system for 5 % steam extraction has been designed and the system performance during steady and transient operations has been analyzed. The TPD system dispatched a total of 197 MW thermal energy to the HTSE plant under nominal design conditions. Saturated steam at 7.17 MPa from the BWR plant was condensed and subcooled to a temperature of 168 °C, while a mass flow rate of 91.1 kg/s of superheated steam was dispatched to the HTSE plant. Furthermore, the system performance during transient operation showed a continuous transition from the initial hot standby mode to the nominal power dispatch level. The transient simulation results emphasized the importance of investigating component level performance for the TPD system design. The current results will guide future works on the development of integrated energy systems for hydrogen production or process heat applications.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Automation of Nanoparticle Synthesis Processes in a Plasma Environment Using LabVIEW

This work presents an automated control system for the synthesis of nanomaterials by plasma-enhanced chemical vapor deposition (PECVD), implemented using the LabVIEW software environment. The main objective of the study is to develop an integrated hardware-software platform that enables sequential control of the key stages of the PECVD process, including vacuum chamber preparation, pressure monitoring, working gas supply, plasma ignition, power matching, cyclic nanomaterial growth, and optical monitoring of nanoparticles in the plasma environment. The use of LabVIEW made it possible to integrate actuator control, experimental parameter acquisition, and realtime process visualization within a single automated system. The automated cycle begins with evacuation of the reaction chamber to a predefined base pressure. Transition to the next stage is permitted only after the specified pressure threshold has been reached, ensuring reproducible initial conditions for each experiment. The program then controls the supply of the working gas through mass flow controllers (MFCs). In this work, two gas-flow control modes were considered: analog control using a 0-5 V voltage signal and digital communication via RS-232 interface. It was shown that the analog approach requires accurate scaling of the control voltage, since applying 5 V corresponds to full-scale opening of the controller and results in the maximum gas flow. In contrast, the RS232 interface enables the gas flow rate to be specified directly in sccm, improving the accuracy, flexibility, and convenience of gas-environment control. After pressure stabilization, LabVIEW initiates RF plasma ignition and executes the RF matching algorithm aimed at minimizing reflected power and improving the stability of the plasma process. A separate software module implements the cyclic nanomaterial growth mode, in which the plasma-on time, plasma duration, and total number of synthesis cycles are predefined. This approach makes it possible to control material accumulation on the substrate and to correlate the process parameters with the morphological characteristics of the resulting nanostructures. The final module of the system is designed for optical monitoring of the nanoparticle cloud density in dusty plasma. For this purpose, the change in the intensity of laser radiation passing through the plasma region is recorded using a photodetector and a Keithley 2401 measuring unit connected to LabVIEW via RS-232 interface. The difference between the initial and modified optical signal intensity is used as a diagnostic parameter characterizing the formation and temporal evolution of nanoparticles. The developed system demonstrates that LabVIEW can be effectively applied not only for the automation of individual instruments, but also for the implementation of a complete digital control cycle for PECVD-based nanomaterial synthesis.

PECVD↗