Search NASA⌕ Search

SEARCH · Search NASA

Results for “power processing unit”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Development of a High-Efficiency Hybrid Dry Cooler System for sCO 2 Power Cycles in CSP Applications

This project addressed a major gap in supercritical CO 2 (sCO 2 ) power cycle research by focusing on the pre-cooler, a component that had received little attention despite its significant impact on cycle efficiency and plant economics. The team developed a compact dry cooler using brazed/diffusion-bonded microchannel passages paired with formed air-side fins, advancing the technology from TRL-2 toward commercial readiness. Compared to conventional fin-tube coolers, the design cuts installation footprint by roughly half for 10+ MWth systems while achieving better heat transfer and lower approach temperatures, translating into a projected LCOE reduction from 6.04 ¢/kWh to between 5.85 and 5.94 ¢/kWh. While fabrication of an aluminum MW-scale prototype revealed brazing and sealing challenges at larger scales, lessons learned informed a subsequent 1 MWth unit that was successfully built and delivered for integration into Sandia's Gen3 Particle Pilot Plant, advancing the technology to TRL-7 with a path toward TRL-8 pending successful testing.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2025

A VSC-HVDC-Assisted Black-Start Strategy in Bulk Power Systems a Case Study in San Diego

With the worldwide growth in deploying high-voltage direct current (HVDC) transmission systems, their ability to facilitate black-start (BS) restoration has been a research topic of interest. In this context, voltage source converter (VSC)-HVDC is regarded as a BS resource, and this paper proposes a VSC-HVDC-assisted parallel BS restoration strategy in bulk power systems. The proposed strategy consists of two stages: 1) determination of the VSC and generator startup sequence and 2) load restoration simulation. In the first stage, the entire blackout system is sectionalized into multiple subsystems. Each subsystem includes a VSC-HVDC station or traditional BS unit, it independently determines its generator startup timeline and the energization timelines for buses and lines. The second stage involves load restoration, conceptualized as a modified unit commitment problem, with the timelines established in the first stage work as critical inputs. The proposed BS restoration strategy is tested on the San Diego power system to simulate the 2011 Southwest blackout. The simulation results validate the effectiveness of using VSC-HVDC links as a BS resource which not only speeds up the restoration process but also reduces both energy and economic losses.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Thermal Integration of Advanced Nuclear Reactors with a Reference Refinery, Methanol Synthesis, and a Wood Pulp Plant (Rev.1)

The present report is intended to provide process flow diagrams (PFDs) and energy and mass balance data sheets for a U.S. industrial sector subset with which nuclear heat and power could be integrated—a subset that includes the oil refining, methanol and pulp and paper industries. Coupling options for integrating nuclear energy into these industries are quantitatively outlined for reference systems, and future work will extend this analysis in greater detail. Opportunities for integrating small modular nuclear reactors (SMNRs) were investigated for each of the industrial process configurations. Aspen HYSYS and Cycle-Tempo models for a high-temperature gas-cooled reactor were developed to evaluate the proposed integration. This introductory evaluation provides a general description and assessment of the operating principles, reactor coolant core outlet temperature, and reactor size to be integrated with industry. The industrial processes of oil refining and the production of methanol, pulp and paper were simulated by using Aspen HYSYS, Aspen Plus, and the PRELIM (Petroleum Refinery Life Cycle Inventory Model) tool to develop process details. Cycle-Tempo models then extend the process modeling results to obtain net energy demands (e.g., heat, steam, and electricity) when accounting for process steam and waste heat recovery. This information is intended to foster the analysis of integrating an SMNR to decarbonize industrial facilities. The SMNR would provide reliable, competitive, and sustainable clean energy while reducing carbon emissions and other environmental impacts, such as water withdrawals, consumption, and contamination. The refining industry, exhibited in Figure ES1, is a leading consumer of fossil -fuel-based heat, power, and hydrogen in the U.S. industrial sector, generating over 164 million metric tons (MMT) of CO 2 emissions in 2023. The overall mass and energy pertaining to a generalized complex refinery in the United States is reflected in Figure ES1, along with energy metrics regarding integration with a nuclear power plant (NPP). Data sheets were developed to indicate the energy requirements for the overall refinery and each refinery process. The data sheet for the overall refinery is shown in Table ES2.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

Technical and Economic Assessment and Gap Analysis of Advanced Nuclear Reactor Integration with a Reference Methanol Synthesis Plant

Efforts continue to identify the most-economic methods to decarbonize several sectors of the United States (U.S.) economy. Industrial processes such as synfuel synthesis and high value commodity chemicals rely heavily on energy-dense and easily stored and transported fossil fuels, which power and feed their operations. Steam methane reforming (SMR) is a widely used process for producing methanol. In this process, methane (CH 4 ) from natural gas (NG) reacts with steam (H 2 O) over a catalyst at high temperatures (700°1,000°C) to produce syngas, a mixture of hydrogen (H 2 ) and carbon monoxide (CO). The syngas is then converted into methanol (CH 3 OH) through a second catalytic reaction. This method is known for being an efficient and commonly employed pathway for industrial methanol production. The high-temperature heat needed for SMR, which is currently used in the natural-gas-to-methanol process, cannot be supplied by small modular nuclear reactor (SMNR) direct heating; the temperatures required for the SMR process exceed those of the main steam produced by near-market high-temperature gas reactors (HTGRs). For the conventional methanol process, this leaves possible nuclear-integration opportunities that include: (1) blending nuclear hydrogen into the SMR NG fuel, or (2) assessing alternative synthesis routes leveraging nuclear capabilities and steam electrolysis outputs. In the reference methanol plant, SMR provides the methanol-synthesis reactor with H 2 and co. In Case (2), the state-of-the-art reverse water gas shift (RWGS) pathway achieves the same, sourcing carbon from an industrial CO 2 source.

08 - HYDROGEN↗
DOE OSTI2025

Denoising Seismograms in the Time Domain Using a Deep Learning Model

Deep learning has emerged as a transformative tool for enhancing the extraction of reliable information from seismograms, addressing the increasing demand for precise and efficient seismic data analysis. We introduce an innovative encoder–decoder deep learning model, named WaveDenoiser, designed for noise reduction in the time domain, thereby eliminating the need for spectrogram computations that have been used for existing deep learning tools and significantly improving processing speed. Utilizing the benchmark dataset that is Stanford Earthquake Dataset, we developed three models of varying sizes: base, medium, and large. Notably, the large (referred to as WaveDenoiser) model demonstrated superior performance, achieving a median signal‐to‐noise ratio improvement of 8.8 dB on in‐distribution unseen data (in the same geographic region) and 7.7 dB on out‐distribution unseen data (in a new geographic region), outpacing both the base and medium models. Further evaluation of the WaveDenoiser model revealed a reduction in median arrival‐time errors by 0.02 s for P waves and 0.01 s for S waves when processing waveforms prior to phase picking using PhaseNet on in‐distribution unseen data. When tested on out‐distribution unseen data, the model also effectively reduced the P‐wave median arrival‐time error by 0.02 and 0.01 s in median arrival‐time error for S waves. Importantly, the application of WaveDenoiser resulted in a significant reduction of phase picking outliers by 1.1% to 3.6% for both P and S waves. In addition, we achieved over five times acceleration in processing speed compared with the seisBench implementation of DeepDenoiser. Our findings underscore the potential of WaveDenoiser as a powerful tool for improving seismic data analysis and processing efficiency.

P-waves↗
DOE OSTI2026

Software Control Program For Transportable Microgrid State-of-charge Balancing And Frequency Stability Controls

A deterministic state-of-charge (SOC) balancing approach software control code is introduced as an integral secondary management to primary control layer of an islanded small microgrid or nanogrid system made up of multiple grid-forming inverter/battery/solar combination systems, where each set of batteries with each inverter are on independent DC buses (i.e. non-paralleled on the DC sides). A DERMS-level control approach, algorithm and automation controller program was developed to improve coordination and enable microgrid asset compliance and SOC balancing, enabling provision of a system-level power stability support architecture, load support, and asset scalability. The architecture is configured to treat each unit or micro/nano-grid as a node in a microgrid network, allowing for autonomous DERMS control regarding load and SOC balancing and power stability. As the network grows with the addition of units, greater coordination efforts may be required. The ideal small network microgrid ranges from 2-10 inverter/battery units before additional control parameters must be considered in the existing architecture. The control approach focuses on a deterministic state-of-charge analysis as the primary level control process followed by a secondary control loop using a forced frequency-watt droop strategy to conform off-the-shelf components into behaving under a leader-follower configuration. Adopting this control scheme has been shown to allow for a balanced, unit-coordinated microgrid network, enabling stable power flow. The deterministic state-of-charge approach is introduced as an integral primary control layer of an islanded small network microgrid. A standard strategy for SOC balancing is implementing a battery management system (BMS) to control SOC on the DC side. An alternative approach is to determine how to coordinate sending and receiving power on the AC side with multiple units. The latter approach assesses all the integrated units in the microgrid network. Once the individual units are identified, further system data is required to calculate each unit's total kWh, provided information about its capability to supply or consume kWh and availability. The secondary control layer in the multi-layered small network microgrid methodology uses the primary layer’s decision to initiate frequency setpoint changes, initializing the SOC balancing. The secondary control layer considers numerous system-dependent variables to enable a charging and discharging profile based on adjustable frequency setpoints. The combined architecture will result in stable, coordinated power flow enhancing an AC microgrid's functionalities.

Myers, KurtS [Idaho National Laboratory (INL), Ida↗
DOE OSTI2025

Benchmarking thermal energy storage cost for industrial process heat

Process heat accounts for roughly half of industrial energy demand, and currently 95% of process heat is derived from the combustion of natural gas, oil, and coal. Electrification of industrial heating could be an alternative, potentially expanding locations suitable for manufacturing; however, industrial facility owners may desire energy storage to stabilize energy costs. In this work, the economic benefits of pairing thermal storage with electrified process heat to reduce the average price paid for energy are analyzed. Cost savings focus on energy arbitrage, or leveraging flexible energy pricing schemes, alone. The cost of natural gas combustion across decades (2019-2060) is compared to the costs of electricity and thermal energy storage in four United States Independent System Operator (ISO) regions. Systems installed today may not yield positive net present value (NPV) compared to the use of natural gas. However, using estimated electricity prices, systems installed in 2030 using arbitrage alone could be profitable when compared to natural gas in some regions of the U.S. Furthermore, if capital expenditures could be reduced by 50% for sensible thermal storage systems by 2030, profitable systems are found across all regions. This implies that electrification of industrial process heat, when paired with inexpensive thermal energy storage systems, could be less expensive than brownfield natural gas systems, using arbitrage as the only source of revenue and without a dependency on any future policy drivers such as pricing externalities that could further incentivize the electrification of industrial process heat.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Large area transparent refractory aerogels with high solar thermal performance

Application of transparent silica aerogels in low-temperature solar thermal systems has led to major improvements in performance. In high temperature concentrating solar thermal (CST) systems, aerogels have yet to demonstrate the necessary scalability, durability, and performance to support their widespread deployment. Here, large-area transparent refractory aerogel tiles are synthesized and shown to achieve a record-high receiver figure-of-merit (FOM) at high temperatures. The work leverages a scaled-up process for sol–gel synthesis to control the density of the aerogels for improved solar transmittance and adapts a previous atomic layer deposition (ALD) technique with the aid of predictive reaction-transport modeling. After aging for 10 days at 700 °C, the large-area tiles exhibit a solar-weighted transmittance of 95.6 % and a thermal emittance of 0.31, corresponding to a FOM of 80 % at 100 suns and 700 °C. The observed sintering rates at 700 °C are comparably low to earlier one-inch aerogels, suggesting long-term stability under relevant operating conditions. Furthermore, the study indicates that refractory aerogels are scalable materials for efficient photothermal conversion at high temperatures.

Aerogels↗
DOE OSTI2024

Design-Point Techno-Economics of Brayton Cycle PTES for Combined Heat and Power

Pumped thermal energy storage (PTES) systems are grid batteries that use heat pumps to create both hot and cold thermal energy stores when there is excess electricity and then use a power cycle to convert the thermal energy into electricity when there is demand for electricity. In normal operation, Joule-Brayton PTES discharges low-grade heat at temperatures useful for thermal energy consumers like district and industrial heating. Furthermore, PTES designs, like conventional combined heat and power (CHP) technology, can be modified to sacrifice some round-trip efficiency RTE to increase the temperature of heat rejection. This paper uses design-point performance and cost models that provide a detailed understanding of the efficiency and cost trade-offs of rejecting heat at various temperatures in ideal-gas Brayton PTES configurations. First, we keep the heat rejection in its nominal location in the PTES system: in the discharge cycle after the low-pressure exit of the recuperator before the cold-storage heat exchanger. Next, we move the heat rejection to the discharge turbine exit. We define design-point metrics that isolate both the cost and performance penalty associated with the hotter heat rejection and attribute it exclusively to the heat economic metrics. Finally, we estimate the performance of electric heater technology to generate heat at equivalent temperatures. We find that the levelized cost of heat, including the cost of thermal energy storage (TES) buffering the PTES and heat off-taker, compares favorable versus electric technologies and is less than the cost of natural gas for low temperature scenarios and competitive with the cost of natural gas in some regions of the contiguous United States in high temperature scenarios.

CHP↗
DOE OSTI2025

A novel approach for large-scale wind energy potential assessment

Increasing wind energy generation is central to grid decarbonization, yet methods to estimate wind energy potential are not standardized, leading to inconsistencies and even skewed results. This study aims to improve the fidelity of wind energy potential estimates through an approach that integrates geospatial analysis and machine learning (i.e., Gaussian process regression). We demonstrate this approach to assess the spatial distribution of wind energy capacity potential in the Contiguous United States (CONUS). We find that the capacity-based power density ranges from 1.70 MW/km2 (25th percentile) to 3.88 MW/km2 (75th percentile) for existing wind farms in the CONUS. The value is lower in agricultural areas (2.73 ± 0.02 MW/km2, mean ± 95 % confidence interval) and higher in other land cover types (3.30 ± 0.03 MW/km2). Notably, advancements in turbine manufacturing could reduce power density in areas with lower wind speeds by adopting low specific-power turbines, but improve power density in areas with higher wind speeds (>8.35 m/s at 120m above the ground), highlighting opportunities for repowering existing wind farms. Wind energy potential is shaped by wind resource quality and is regionally characterized by land cover and physical conditions, revealing significant capacity potential in the Great Plains and Upper Texas. The results indicate that areas previously identified as hot spots using existing approaches (e.g., the west of the Rocky Mountains) may have a limited capacity potential due to low wind resource quality. Improvements in methodology and capacity potential estimates in this study could serve as a new basis for future energy systems analysis and planning.

Dai, Tao↗
DOE OSTI2024

Technical and Economic Assessment and Gap Analysis of Advanced Nuclear Reactor Integration with a Reference Methanol Synthesis Plant

Efforts continue to identify the most-economic methods to decarbonize several sectors of the United States (U.S.) economy. Industrial processes such as synfuel synthesis and high value commodity chemicals rely heavily on energy-dense and easily stored and transported fossil fuels, which power and feed their operations. Steam methane reforming (SMR) is a widely used process for producing methanol. In this process, methane (CH 4 ) from natural gas (NG) reacts with steam (H 2 O) over a catalyst at high temperatures (700-1,000°C) to produce syngas, a mixture of hydrogen (H 2 ) and carbon monoxide (CO). The syngas is then converted into methanol (CH 3 OH) through a second catalytic reaction. This method is known for being an efficient and commonly employed pathway for industrial methanol production. The high-temperature heat needed for SMR, which is currently used in the natural-gas-to-methanol process, cannot be supplied by small modular nuclear reactor (SMNR) direct heating; the temperatures required for the SMR process exceed those of the main steam produced by near-market high-temperature gas reactors (HTGRs). For the conventional methanol process, this leaves possible nuclear-integration opportunities that include: (1) blending nuclear hydrogen into the SMR NG fuel, or (2) assessing alternative synthesis routes leveraging nuclear capabilities and steam electrolysis outputs. In the reference methanol plant, SMR provides the methanol-synthesis reactor with H 2 and co. In Case (2), the state-of-the-art reverse water gas shift (RWGS) pathway achieves the same, sourcing carbon from an industrial CO 2 source.

08 HYDROGEN↗
DOE OSTI2026

Digital Grid Twin–Direct Communication Scheme Test Bed for Assessing Relay-to-Relay Radio Antenna and Optical Fiber Performance and Misoperations

This study introduces a novel “Digital Grid Twin–Direct Communication Scheme” test bed. This advanced platform evaluates point-to-point communication between transmitter and receiver relays with optical fiber and radio omnidirectional antenna systems, implemented at the Advanced Protection lab in the Grid Research Innovation and Development Center at Oak Ridge National Laboratory. The increased diversity of energy sources has led to more protective relay misoperations. In North America, microgrid protection schemes now use point-to-point communication along distribution lines between relays to implement advanced logic in nonradial grids that include both high- and low-inertia generators. This trend challenges utilities to minimize misoperations while ensuring rapid fault clearance and accurate selectivity coordination between primary and backup relays. This study assesses relay-to-relay communication schemes by introducing an advanced testing platform based on a digital grid twin protection test bed using a synchronized time source system. The platform evaluates the communication system using radio antennas or optical fiber links by integrating protective relays that operate breakers within the digital twin and record relay events and communication signals. In the experiments, transmitter and receiver relays were configured with inverse time overcurrent and breaker trip detection logic to assess the total time of the communication protection schemes based on the sum of the relay protection element operating time, radio latency, propagation delay, baud rate delay, and relay processing time. These delays were derived from recorded relay events and communication signals from the interface of a real-time simulator set as a digital grid twin. The test bed successfully simulated various electrical faults along a distribution line while ensuring effective and reliable point-to-point communication between transmitter and receiver relays. The radio antenna communication system exhibited latency because of the radio. This latency depends on the baud rate setting and type of radio application; in general, the higher the baud rate, the lower the radio latency. The measured radio latency (for Mirrored Bits with an encryption card at 9,600 bps) was about 9–10 ms. Additionally, calculated propagation delay per mile for radio antennas and optical fiber was 5.36 µs/mi and 8.04 µs/mi, respectively. Optical fiber communication did not demonstrate radio latency. Instead, the protection element operating time depends mainly on the protection logic function set in the relay, and the relay processing time depends on the processing rate of the relay in samples per power system cycle.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

Investigation of Flux Spreading in a Light-Trapping, Planar-Cavity Receiver for Enclosed Solar Particle Heating

Concentrating solar thermal power (CSP) technology development has recently focused on increasing the operating temperatures to accommodate high efficiency power cycles and thermochemical processes. Inert solid particles as heat transfer media enable solar receivers to operate above 700 degrees Celsius resulting in increased system thermal efficiency compared to the conventional molten salt based CSP system. An open-cavity falling-particle solar receiver that can efficiently heat particles by direct heating from concentrated solar radiation faces challenges with large particle losses from wind and unable to support thermochemical reactions. A light-trapping, planar cavity reiver (LTPCR) where particles are indirectly heated can significantly minimize the particle losses during the operation, support thermochemical reactions, and offer scalability potential. The LTPCR features an array of vertical planar receiver/absorber panels arranged within a cavity configuration. Concentrated solar radiation from heliostats is focused onto the receiver walls, where heat is indirectly transferred to solid particles flowing inside the receiver channels. Heat transfer occurs through direct contact between the receiver panel walls and particles, and can be enhanced by fluidizing particles with air. This fluidization increases particle-wall contact and extends particle residence time, maximizing heat transfer efficiency. The unique vertical planar receiver structure originated from a near-blackbody tubular light absorber, effectively distributing the incoming solar beam spread across the panel walls and trapping light. This flux spreading effect, driven by cosine projection, converts high incident solar flux into a lower, more uniform heat flux on the panel walls. This redistribution enhances heat transfer efficiency between particle-wall or reaction gases-wall, while preventing localized overheating of the receiver panel. Indirect planar cavity solar receivers completely separate solid particles from the ambient environment that can greatly reduce the thermal losses in heated particles resulting in high efficiency at high temperatures above 700 degrees Celsius. This design ensures no particle losses to the environment during the operation while open-cavity designs can experience significant particle losses from wind. An experimental investigation was conducted to observe flux spreading on the receiver panel wall. A lab-scale prototype planar receiver, fabricated using Haynes 230 alloy, was tested under direct concentrated solar radiation using the high-flux solar furnace (HFSF) facility at NREL. The experiment was performed under normal peak radiative heat fluxes ranging from 800 to 1900 kW/m2. A temperature distribution on the panel wall was measured using a thermal imaging camera (FLIR A 6600). To prevent overheating at the receiver front tip, prism-shaped heat shields (Zircar UNIFROM C1) were placed in front of the receiver, and their influence on flux spreading was also studied. Absorbed flux distribution on the panel wall was modeled using SolTrace. The total solar power and flux distributions delivered from HFSF were determined based on the heliostat mirror optical properties, direct normal irradiance (DNI) on the on-sun testing days, peak flux measurement during the on-sun testing, and shutter/attenuator settings Due to the large incident angles of the solar beam on the panel wall, the angular optical properties of Haynes 230 alloy and Zircar heat shields were incorporated into the model. This flux distribution model was then integrated into a computational fluid dynamics (CFD) simulation to predict the receiver panel wall temperature, which was compared with the experimental measurements. Both prediction and measurements identified a temperature hotspot at the backside of the panel, indicating that the incident solar beam can fully reach to the rear of the receiver. The heat shields positioned at the front of the receiver effectively reduced the excessive temperature rise at the receiver front tip. Overall, the temperature was well distributed over the panel wall, with a minor hotspot at the back of the receiver. The model slightly overpredicted the temperature, possibly due to discrepancies in optical properties of the panel and an underprediction of thermal loss in the receiver. The advancement of the particle LTPCR offers a viable alternative to open-cavity receivers by addressing particle loss issues. Additionally, it presents a pathway for enabling solar thermochemical processes, extending CSP technology beyond power generation to fuel and chemical production.

14 SOLAR ENERGY↗
DOE OSTI2025

Constraining nuclear mass models using 𝑟-process observables with multiobjective optimization

Modeling nuclear masses, particularly for nuclei far from stability, remains a key objective in nuclear physics. One contemporary approach is machine learning (ML), which trains on experimental data, but can suffer large errors when extrapolating toward neutron-rich species. In nature, such masses shape observables for the rapid neutron capture process (𝑟 process), which in principle could inform ML models. Here, we introduce a multiobjective optimization approach using the Pareto front algorithm. We show that this technique, capable of identifying models that generate 𝑟-process abundances aligning with both solar and stellar data, is a promising method to select ML models with reliable extrapolation power.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Decarbonizing Industrial Heat and Electricity Applications Using Advanced Nuclear Energy

Idaho National Laboratory (INL) is investigating the technical pathways to assist industrial heat and electricity users to meet their decarbonization goals through integration with advanced nuclear power plants (NPPs). This project will deliver a library of process models and accompanying documents that guide specific industries in choosing potential nuclear technologies based on their needs. Considerations in providing this guidance include specific hazards from the industrial facility, heat transport requirements and associated technologies, and feasibility with site-specific demand profiles. The library of facility process models will be based on real data from industrial facilities in the United States. The industrial processes will be identified in this project based on the following: (1) operational heat characteristics that nuclear systems can provide, (2) sufficient energy requirements to merit the capital investment for nuclear plant construction, and (3) environmental benefits of replacing existing energy production with carbon-free nuclear power. Other decarbonization opportunities considered are the addition of nuclear-powered electrolysis processes or high-temperature electric heating where the thermal requirements exceed nuclear generation conditions. In addition to assessing the technical feasibility, INL is evaluating the impact of hazards introduced by the industrial facilities on the siting requirements of advanced NPPs. Site characterization of an industrial plant is essential to determine the feasibility and suitable integration methods for each industry. The assessment of siting and technical data will reveal opportunities for a single-use nuclear integration as well as integration of multiple industrial facilities with a single NPP.

02 PETROLEUM↗
DOE OSTI2025

Hardware acceleration for HPS algorithms in two and three dimensions

We provide a flexible, open-source framework for hardware acceleration, namely massively-parallel execution on general-purpose graphics processing units (GPUs), applied to the hierarchical Poincaré–Steklov (HPS) family of algorithms for building fast direct solvers for linear elliptic partial differential equations. To take full advantage of the power of hardware acceleration, we propose two variants of HPS algorithms to improve performance on two- and three-dimensional problems. In the two-dimensional setting, we introduce a novel recomputation strategy that minimizes costly data transfers to and from the GPU; in three dimensions, we modify and extend the adaptive discretization technique of Geldermans and Gillman [1] to greatly reduce peak memory usage. We provide an open-source implementation of these methods written in JAX, a high-level accelerated linear algebra package, which allows for the first integration of a high-order fast direct solver with automatic differentiation tools. We conclude with extensive numerical examples showing our methods are fast and accurate on two- and three-dimensional problems.

Fast direct solvers↗