SEARCH · Engineering Papers

Results for “lead optimization”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 487 records · Page 27

All sources NASA NTRS DOE OSTI

NASA NTRS1994

Integrated design of structures, controls, and materials

In this talk we shall discuss algorithms and CAD tools for the design and analysis of structures for high performance applications using advanced composite materials. An extensive mathematical theory for optimal structural (e.g., shape) design was developed over the past thirty years. Aspects of this theory have been used in the design of components for hypersonic vehicles and thermal diffusion systems based on homogeneous materials. Enhancement of the design methods to include optimization of the microstructure of the component is a significant innovation which can lead to major enhancements in component performance. Our work is focused on the adaptation of existing theories of optimal structural design (e.g., optimal shape design) to treat the design of structures using advanced composite materials (e.g., fiber reinforced, resin matrix materials). In this talk we shall discuss models and algorithms for the design of simple structures from composite materials, focussing on a problem in thermal management. We shall also discuss methods for the integration of active structural controls into the design process.

Blankenship, G. L.↗

DOE OSTI2020

Enhanced magnetocaloric performance in manganite bilayers

Bilayer films of La 0.8 Sr 0.2 MnO 3 and La 0.7 Ca 0.3 MnO 3 with different layer thicknesses and stacking sequences were synthesized via pulsed laser deposition. The magnetic properties and magnetocaloric effects were systematically investigated. We found that the phase transition associated with the layers tends to merge together at an optimal thickness and a stacking sequence. The operating temperature span of the entropy change broadens at the expense of its magnitude, leading to an enhancement of the refrigerant capacity by over 40%. The optimized bilayer film possesses a refrigerant capacity of ~6.0J/kg, compared to a value of ~4.2J/kg for the single-phase films. The phase transition broadening induced by epitaxial strain is responsible for the enhanced window of operating temperature as well as refrigerant capacity. Finally, the design principle developed here may be applied to ferroelastic and ferroelectric materials to enhance elastocaloric and electrocaloric effects.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Operation at Reduced Atmospheric Pressure and Concept of Reliability Redundancy for Optimized Design of Insulation Systems

Electrified transportation is calling for insulation design criteria that is adequate to provide elevated levels of power density, power dynamics and reliability. Increasing voltage levels are expected to cause accelerated intrinsic and extrinsic aging effects which will not be easily predictable at the design stage due to a lack of suitable modeling. Designing reliable insulation systems would require finding solutions able to control accelerated aging due to an unpredictable increase of intrinsic stresses and the onset of extrinsic stresses as partial discharges. This paper proposes the concept of reliability redundancy for the insulation design of aerospace electrical asset components, which is also validated at lower-than-standard atmospheric pressure. The principle is that extrinsic-aging-free design might be achieved upon determining the aging stress or abnormal service stresses distribution and being sure that aging will not generate conditions that can incept extrinsic aging (partial discharges) during operation life. However, such information is never, in practice, fully available to insulation system designers. Hence, especially in critical applications such as electrified aircraft, aerospace, and combat ships a further level of reliability should be added to a partial-discharge-free design, which can consist of the use of corona-resistant materials and/or of life models able to consider the accelerated aging effect of partial discharges (or any other type of extrinsic-accelerated aging factor). Innovative life modeling considering both extrinsic and intrinsic aging stresses, insulating material testing to estimate model parameters, and a metric for quantifying the extent of corona (or partial discharge) resistance can lead to establishing feasibility and limit conditions for optimized or fully reliability-redundant design. It is shown in the paper that if an extrinsic-aging-free design is not feasible, and it is therefore replaced by a redundant design, a further level of reliability redundancy can be provided by effective condition monitoring plans.

Montanari, Gian Carlo↗

NASA NTRSDate not supplied

Advances in Design Capabilities for Planetary Missions from the NASA Entry Systems Modeling and Instrumentation Portfolio

The Entry Systems Modeling project (ESM) is supported by both the NASA Space Technology and the Science Mission Directorates and focuses on developing simulation tools and validated models for characterizing the performance of entry systems tailored to planetary destinations across the Solar System. ESM is organized into six technical capability areas that together address all relevant factors related to spacecraft entry, as well as some aspects of descent: Thermal Protection System (TPS) Materials; Aerothermodynamics; Entry & Descent Vehicle Dynamics; Guidance, Navigation, and Control; Vehicle Systems Analysis; and Advanced Tools and Numerical Methods. Development within the capability areas is undertaken explicitly with a focus on transition and infusion to science missions, human exploration missions, and commercial space activities. The present talk details developments that specifically impact science missions, including simulation tool capabilities that aid in mission design and model development to understand entry system performance at a given destination. Examples of the successful infusion and transition of such project outcomes to science missions also are provided. Several simulation tool development efforts within ESM have resulted in new design capabilities for missions. One such outcome is improved toolsets for mission trajectory and concept of operations design. Specifically, an initiative to couple a leading tool for entry, ascent/descent, and orbital trajectory optimization (Program to Optimize Simulated Trajectories II or POST2) to those used within the Agency for interplanetary trajectory optimization (Copernicus and Monte) has made substantial progress, with the outcomes to date promising to allow efficient trajectory optimization across mission phases. Additionally, toolchains for the evaluation of vehicle performance during entry and descent have been developed that allow assessment of multi-dimensional aeroheating on detailed vehicle geometries, characterization of deployment and inflation of parachutes, and assessment of vehicle dynamic stability during descent. These capabilities are achieved by coupling diverse sets of physics together – material response, computational fluid dynamics, radiation, and vehicle dynamics – to suitably describe complex entry and descent phenomena. Several model development and validation efforts for specific destinations and entry regimes also are underway within the ESM project. For instance, new experimental capabilities to validate radiation models at low densities/high altitudes recently have been established with project support, specifically the Low-Density Shock Tube (LDST) at the NASA Ames Research Center Electric Arc Shock Tube (EAST) facility. The LDST is being leveraged to develop improved models of shock layer kinetics and radiation in Titan atmospheres, while future studies will be conducted in the LDST and the existing high velocity shock tube to provide validation data for radiation models of Venus, Ice Giants, and Mars atmospheres. Models describing the aerothermal and thermo-structural performance of Thermal Protection System (TPS) materials has been another focus, with multiscale modeling activities on-going for the two leading TPS materials applicable to a range of entry conditions and science missions: the Phenolic-Impregnated Carbon Ablator (PICA) and woven materials like 3D Mid-Density Carbon Phenolic (3MDCP). A continual effort is made to infuse and transition outcomes from ESM simulation tool and model development activities into relevant science missions. Significant progress has been made on this front, with missions such as Dragonfly, DAVINCI, and Mars Missions benefitting from project outcomes. The groundwork also is being laid to provide insights into forward looking missions to Gas/Ice Giants as well as for potential sample returns.

Justin Haskins↗

DOE OSTI2025

Electric Drive Technologies Consortium (EDTC)/ Cost competitive, high-Performance, highly Reliable (CPR) Power Devices on 4H-SiC (Final Report)

4H-Silicon carbide (4H-SiC) is a wide bandgap semiconductor that offers superior material properties over silicon, including higher critical electric field, thermal conductivity, and electron saturation velocity. These advantages make 4H-SiC highly attractive for high-voltage, high-efficiency power electronics. However, realizing the full potential of SiC requires device technologies that are not only high-performing but also manufacturable and reliable under real-world operating conditions. This report summarizes the outcomes of a five-year R&D effort funded by the U.S. Department of Energy (DOE) under the Electric Drive Technologies Consortium (EDTC), focused on developing cost-competitive, high-performance, and highly reliable (CPR) power devices on 4H-SiC substrates. The program targeted scalable and manufacturable 1.2 kV-class SiC MOSFETs optimized for next-generation electric vehicles, renewable energy systems, and industrial power conversion. The project delivered transformative advancements in SiC power device performance and ruggedness. Particularly, Specific on-resistance (R on,sp ) was reduced by up to 37%, from ~4.0 m$\Omega \cdot$cm 2 in earlier designs to an industry-leading 2.40 m$\Omega \cdot$cm 2 , driven by optimized doping, refined JFET widths, and layout engineering. Breakdown voltages (BV) exceeded 1600 V, marking improvement over legacy baselines, and demonstrating the robustness of newly implemented junction profiles and edge terminations. Short-circuit withstand time (SCWT) saw a remarkable 4$\times$ increase, from ~2 $\mu$s to over 8 $\mu$s, achieved through the successful deployment of deep P-well structures (~1.8–2.0 $\mu$m) via channeling implantation. This innovative process breakthrough enabled precise junction formation without MeV-class implantation tools, reduced leakage under high field stress, and allowed even the shortest-channel devices (down to 0.3 $\mu$m) to achieve both high BV and excellent ruggedness—breaking the traditional trade-off between conduction efficiency and blocking capability. Several novel architectures pushed the performance envelope further. JBSFETs—featuring embedded Schottky portions—eliminated bipolar degradation and drastically reduced third-quadrant leakage, while Ladder MOSFETs introduced a clever orthogonal conduction path that achieved a 15.4% reduction in R on,sp over standard linear designs. Switching performance reached new benchmarks: short-channel devices showed a 31% reduction in total switching energy compared to 0.5 $\mu$m counterparts, while maintaining manageable gate drive requirements. Layout-optimized structures not only improved transconductance but also accelerated switching transitions, pointing to real-world benefits in converter-level efficiency. The devices also passed rigorous reliability validation. Stress-tested across TDDB, HTGB, HTRB, HVP, and burn-in, the devices screened under 30 V/10 hr and 43 V/1 s protocols consistently exhibited tighter lifetime distributions and long-term oxide robustness. These screening techniques proved effective in identifying latent defects and ensuring deployment-grade reliability. Meanwhile, advanced 3D TCAD simulations revealed and resolved electric field hotspots—particularly in HEXFET corners—where fields exceeding 4.8 MV/cm were mitigated through geometry-aware layout corrections. Overall, the results of this project demonstrate a manufacturable and scalable SiC power device platform that addresses key DOE performance targets for efficient, robust, and reliable 1.2kV 4H-SiC Power Devices. The developed technologies represent a meaningful step forward in the commercial readiness of high-voltage SiC solutions and provide a strong foundation for continued advancement in wide bandgap power electronics.

42 ENGINEERING↗

NASA NTRS1991

Optimal control of systems with capacity: Related noises

In the ordinary theory of optimal control (LQR and Kalman filter), the variances of the actuators and the sensors are assumed to be known (not related to the capacities of the devices). This assumption is not true in practice. Generally, a device with greater capacity to exert actuating forces and a sensor capable of sensing greater sensing range will generate noise of greater power spectral density. When the ordinary theory of optimal control is used to estimate the errors of the outputs in such cases it will lead to faulty results, because the capacities of such devices are unknown before the system is designed. The performance of the system designed by the ordinary theory will not be optimal as the variances of the sensors and the actuators are neither known nor constant. The interaction between the control system and structure could be serious because the ordinary method will lead to greater feedback (Kalman gain) matrices. Methods which can optimize the performance of systems when noises of the actuators and the sensors are related to their capacities are developed. These methods will result in smaller feedback (Kalman gain) matrix.

Ruan, Milfang↗

NASA NTRS2011

Tissue Doppler Imaging Combined with Advanced 12-Lead ECG Analysis Might Improve Early Diagnosis of Hypertrophic Cardiomyopathy in Childhood

Optimization of early diagnosis of childhood hypertrophic cardiomyopathy (HCM) is essential in lowering the risk of HCM complications. Standard echocardiography (ECHO) has shown to be less sensitive in this regard. In this study, we sought to assess whether spatial QRS-T angle deviation, which has shown to predict HCM in adults with high sensitivity, and myocardial Tissue Doppler Imaging (TDI) could be additional tools in early diagnosis of HCM in childhood. Methods: Children and adolescents with familial HCM (n=10, median age 16, range 5-27 years), and without obvious hypertrophy but with heredity for HCM (n=12, median age 16, range 4-25 years, HCM or sudden death with autopsy-verified HCM in greater than or equal to 1 first-degree relative, HCM-risk) were additionally investigated with TDI and advanced 12-lead ECG analysis using Cardiax(Registered trademark) (IMED Co Ltd, Budapest, Hungary and Houston). Spatial QRS-T angle (SA) was derived from Kors regression-related transformation. Healthy age-matched controls (n=21) were also studied. All participants underwent thorough clinical examination. Results: Spatial QRS-T angle (Figure/ Panel A) and septal E/Ea ratio (Figure/Panel B) were most increased in HCM group as compared to the HCM-risk and control groups (p less than 0.05). Of note, these 2 variables showed a trend toward higher levels in HCM-risk group than in control group (p=0.05 for E/Ea and 0.06 for QRS/T by ANOVA). In a logistic regression model, increased SA and septal E/Ea ratio appeared to significantly predict both the disease (Chi-square in HCM group: 9 and 5, respectively, p less than 0.05 for both) and the risk for HCM (Chi-square in HCM-risk group: 5 and 4 respectively, p less than 0.05 for both), with further increased predictability level when these 2 variables were combined (Chi-square 10 in HCM group, and 7 in HCM-risk group, p less than 0.01 for both). Conclusions: In this small material, Tissue Doppler Imaging and spatial mean QRS-T angle deviation, particularly when combined, appear to be sensitive in predicting the risk for developing childhood HCM. Large-scale, prospective studies are needed to confirm this hypothesis.

Femlund, E.↗

DOE OSTI2020

Blading Phase–Pure Formamidinium–Alloyed Perovskites for High–Efficiency Solar Cells with Low Photovoltage Deficit and Improved Stability

Currently, blade-coated perovskite solar cells (PSCs) with high power conversion efficiencies (PCEs), i.e. greater than 20%, normally employ methylammonium lead tri-iodide with a sub-optimal bandgap. Alloyed perovskites with formamidinium (FA) cation have narrower bandgap and thus enhance device photocurrent. However, FA-alloyed perovskites show low phase stability and high moisture sensitivity. Here, we report that incorporating 0.83 molar percent organic halide salts (OHs) into the perovskite inks enables phase-pure, highly crystalline FA-alloyed perovskites with extraordinary optoelectronic properties. Here, the OH molecules modulate the crystal growth, enhance the phase stability, passivate ionic defects at the surface and/or grain boundaries, and enhance the moisture stability of the perovskite film. We demonstrate a high efficiency of 22.0% under 1 sun illumination for blade-coated PSCs with an open-circuit voltage of 1.18 V, corresponding to a very small voltage deficit of 0.33 V, and significantly improved operational stability with 96% of the initial efficiency retained under one sun illumination for 500 hours.

14 SOLAR ENERGY↗

DOE OSTI2020

Optimized Lie–Trotter–Suzuki decompositions for two and three non-commuting terms

Lie–Trotter–Suzuki decompositions are an efficient way to approximate operator exponentials exp ( t H ) when H is a sum of n (non-commuting) terms which, individually, can be exponentiated easily. They are employed in time-evolution algorithms for tensor network states, digital quantum simulation protocols, path integral methods like quantum Monte Carlo, and splitting methods for symplectic integrators in classical Hamiltonian systems. Here, we provide optimized decompositions up to order t 6 . The leading error term is expanded in nested commutators (Hall bases) and we minimize the 1-norm of the coefficients. For n = 2 terms, several of the optima we find are close to those in McLachlan (1995). Generally, our results substantially improve over unoptimized decompositions by Forest, Ruth, Yoshida, and Suzuki. We explain why these decompositions are sufficient to efficiently simulate any one- or two-dimensional lattice model with finite-range interactions. This follows by solving a partitioning problem for the interaction graph.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2026

Understanding Adsorption and Reactions at Aqueous Oxide Interfaces with Neural Network Potential Molecular Dynamics

Chemical processes at metal oxide−water interfaces are of central importance in geochemistry, biology, and energy technologies. A better understanding of these processes would allow us to make a significant step toward optimizing and controlling them, which could in turn lead to broader impacts. Computational modeling is indispensable to accomplishing this task because complexity and disorder often make it difficult to extract atomistic information from experiments. Balancing computational cost and accuracy, simulation schemes based on efficient machine learning representations of the potential energy surface (PES) predicted by ab initio calculations have become increasingly popular over the past decade. In particular, several studies have demonstrated the ability of machine learning models to accurately reproduce the complex ab initio PESs of aqueous oxide interfaces, allowing simulations of systems and processes that are not accessible with ab initio methods. In this Account, we review our recent efforts to understand adsorption processes and reactions at aqueous oxide interfaces using deep potential molecular dynamics (DPMD), a simulation scheme employing deep neural networks (DNNs), which has proven to be quite successful in accurately describing many different systems in the condensed phase. After summarizing the DPMD methodology, we first review our work on the acid−base chemistry of oxide surfaces in contact with water, a fundamental characteristic that controls proton transfer and surface charge at the interface. We focus on the aqueous interface of rutile IrO 2 , an oxide material thus far considered the best catalyst for the oxygen evolution reaction (OER). We show that this interface is characterized by a large fraction of dissociated water and a strong Brønsted acidity of the surface sites, in good agreement with the experimentally measured value of the point of zero proton charge. In our second example, we investigate how the adsorption of organic species from ambient air or water affects the structure and wettability of the aqueous interfaces of TiO 2 , a prototypical photocatalytic material. This is a question that is relevant to understanding the UV-induced hydrophilicity of TiO 2 surfaces, a property at the basis of self-cleaning windows and related applications. Specifically focusing on formic and acetic acids, the two most common atmospheric organic acids, our simulations reveal that these acids control the wettability of TiO 2 largely through acid−base chemistry at the interface rather than chemisorption on the oxide surface, a finding that could help improve the design of self-cleaning surfaces and photocatalytic devices. Finally, we review our recent study of methanol at TiO 2 −water interfaces, a system whose interest is largely motivated by the role of methanol in enhancing photocatalytic hydrogen evolution on TiO 2 . Our simulations provide mechanistic insights into the coupled roles of the organic adsorbate and water at the TiO 2 interface, with implications for how methanol enhances the activity of H 2 evolution.

adsorption↗

DOE OSTI2020

Mechanistic Understanding and Rational Design of Quantum Dot/Mediator Interfaces for Efficient Photon Upconversion

The semiconductor-nanocrystal-sensitized, three-component upconversion system has made great strides over the past 5 years. The three components (i.e., triplet photosensitizer, mediator, and emitter) each play critical roles in determining the input and output photon energy and overall quantum efficiency (QE). The nanocrystal photosensitizer converts the absorbed photon into singlet excitons and then triplet excitons via intersystem crossing. The mediator accepts the triplet exciton via either direct Dexter-type triplet energy transfer (TET) or sequential charge transfer (CT) while extending the exciton lifetime. Through a second triplet energy-transfer step from the mediator to the emitter, the latter is populated in its lowest excited triplet state. Triplet–triplet annihilation (TTA) between two triplet emitters generates the emitter in its bright singlet state, which then emits the upconverted photon. Quantum dots (QD) have a tunable band gap, large extinction coefficient, and small singlet–triplet energy losses compared to metal–ligand charge-transfer complexes. This high triplet exciton yield makes QDs good candidates for photosensitizers. In terms of driving triplet energy transfer, the triplet energy of the mediator should be slightly lower than the triplet exciton energy of the QD sensitizer for a downhill energy landscape with minimal energy loss. The same energy cascade is also required for the transfer from the mediator to the emitter. Lastly, the triplet energy of the emitter must be slightly larger than one-half of its singlet energy to ensure that TTA is exothermic. Optimization of the sensitizer, mediator, and emitter will lead to an increase in the anti-Stokes shift and the total quantum efficiency. Evaluating each individual step’s efficiency and kinetics is necessary for the understanding of the limiting factors in existing systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2025

A Unique Case of the “Goldilocks Rule” in Solid-State Electrolytes: Two Are Good, Four Are Too Many

We report the syntheses of two new series of methacrylate monomers with different backbones: ureidopyrimidinone (PU) and boron-substituted urea pyrimidine (U), which enhance both the mechanical and electrochemical properties of the solid-state electrolyte (SSE) while improving the cycle life of lithium iron phosphate (LiFePO 4 , LFP) cathodes. The PU backbone is characterized by four hydrogen bonds (H-bonds), while the U backbone bears only two. Importantly, our research reveals that two H-bonds in these monomers are optimal; in contrast, four are excessive. The exceptional mechanical properties and processability of the SSE with the U series additives, resulting from the optimal H-bonds, were unexpectedly achieved. This leads to the establishment of a “Goldilocks rule” for additive design. The key strategies include: 1) reducing hydrogen-bonding (H-bonding) sites by changing pyrimidinone to pyrimidine and 2) shifting from intermolecular to intramolecular H-bonding and π−π bonding. Furthermore, this reduction in H-bonding also offers significant advantages in processability. The advancement can be extended to electrode fabrication, making the manufacturing of all-solid-state batteries more practical and efficient.

hydrogen bonding↗

DOE OSTI2026

Routes to high-performance operation in Wendelstein 7-X: turbulence suppression with shaping of the density profile

Steep density gradients generally lead to improved plasma performance in the neoclassically optimized stellarator W7-X. This is evident in the global energy confinement time as well as in the ion temperature and can be explained by a strong reduction of the ion temperature gradient turbulence. Such conditions can experimentally be realized by several methods: injection of cryogenic hydrogen pellets, appropriate combination of neutral beam and electron cyclotron resonance heating (ECRH) and, in some cases, with low power of ECRH after preconditioning of the first wall. The duration of the improved phases is determined by the ability to sustain the steep density gradient, by technical limitations of the involved systems and, eventually, by the plasma stability. This paper gives an overview of relevant experimental results and presents example discharges where the improved confinement conditions could be extended to multiple seconds: up to 4 s using neutral beam injection and from 14 to 40 s with steady state pellet injection. In these plasmas the turbulent thermal diffusivity is reduced by a factor of 3 to 4 in a broad radial range, which allows high ion temperatures of up to 3 keV at the densities of about 1.5 • 10 20 m −3 .

high performance plasma↗

DOE OSTI2025

Tracing Long-Lived Atomic Coherences Generated via Molecular Conical Intersections

Accessing coherences is key to fully understand and control ultrafast dynamics of complex quantum systems like molecules. Most photochemical processes are mediated by conical intersections, which generate coherences between electronic states in molecules. We show with accurate calculations performed on gas-phase methyl iodide that electronic coherences of spin-orbit-split states persist in atomic iodine after dissociation. Our simulation predicts a maximum magnitude of vibronic coherence in the molecular regime of 0.75% of the initially photoexcited state population. Upon dissociation, one-third of this coherence magnitude is transferred to a long-lived atomic coherence where vibrational decoherence can no longer occur. To trace these dynamics, we propose a tabletop experimental approach—heterodyned attosecond four-wave-mixing spectroscopy. This technique can temporally resolve small electronic coherence magnitudes and reconstruct the full complex coherence function via phase cycling. Hence, heterodyned attosecond four-wave-mixing spectroscopy leads the way to a complete understanding and optimal control of spin-orbit-coupled electronic states in photochemistry.

Chemical Physics↗

DOE OSTI2023

Cryogenic System Preliminary Design for a 0.5m-Long, Conduction-Cooled Nb3Sn Undulator Magnet Prototype

There are several NbTi superconducting undulator (SCU) magnets currently in operation at the Advanced Photon Source (APS) at Argonne National Laboratory (ANL). The development on Nb3Sn-based superconducting undulator magnets at APS is underway due to the potential to further enhance the performance of the SCUs. Superconducting undulator magnets need to keep temperature gradients minimized in order to retain thermal and operating current margin. We have designed the thermal links for efficient heat conduction using 3D finite element analysis (FEA) simulation in COMSOL Multiphysics software, which was later used for the calculation of the temperature distribution across a 0.5 m long, conduction-cooled, Nb3Sn undulator magnet prototype that includes both conductive heat transfer and radiative heating components. We have modelled the evolution of the thermal properties of the magnet winding as well as other cold parts during cool-down from ambient temperature with the operation of a SHI RDK-415D cryocooler to examine the estimated time that is needed for the cooldown as well as the baseline temperature we could achieve. A key result was that a maximum coil delta T = 0.11 K along the designed thermal links and a temperature range of 3.42 K-3.53 K for the winding were predicted at steady state. A pair of G10 support rods for carrying the magnet as well as a pair of current leads made of metals were also designed and optimized for the testing system, targeting for an operation temperature of 4 K and a coil current of 950 A DC.

Zhang, Danlu↗

DOE OSTI2022

Automatic Generation of High-Performance Convolution Kernels on ARM CPUs for Deep Learning

In this work, we present FastConv, a template-based code auto-generation open source library that can automatically generate high-performance deep learning convolution kernels of arbitrary matrices/tensors shapes. FastConv is based on the Winograd algorithm, which is reportedly the highest performing algorithm for the time-consuming convolution layers of convolutional neural networks. ARM CPUs cover a wide range designs and specifications, from embedded devices to HPC-grade CPUs. The leads to the dilemma of how to consistently optimize Winograd-based convolution solvers for convolution layers of different shapes. FastConv addresses this problem by using templates to auto-generate multiple shapes of tuned kernels variants suitable for skinny tall matrices. As a performance portable library, FastConv transparently searches for the best combination of kernel shapes, cache tiles, scheduling of loop orders, packing strategies, access patterns, and online/offline computations. Auto-tuning is used to search the parameter configuration space for the best performance for a given target architecture and problem size. The experiments with layer-wise evaluation on the VGG--16 model confirms a 1.25x performance gains is got by tuning the Winograd library. Integrated comparison results shows 1.02x to 1.40x, 1.14x to 2.17x, and 1.22x and 2.48x speedup is achieved over NNPACK, Arm NN, and FeatherCNN on the Kunpeng 920 beside few cases. Furthermore, problem size performance portability experiments with various convolution shapes shows that FastConv achieves 1.2x to 1.7x speedup and 2x to 22x speedup over NNPACK and ARM NN inference engine using Winograd on Kunpeng 920 . CPU performance portability evaluation on the VGG--16 show an average speedup over NNPACK of 1.42x, 1.21x, 1.26x, 1.37x, 2.26x, and 11.02x is observed on Kunpeng 920, Snapdragon 835, 855, 888, Apple M1, and AWS Graviton2, respectively.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Electrolyte and Cutoff Potential Effects on Cycle Life of Li4Ti5O12/LiNi0.9Mn0.1O2 Batteries for Behind-the-Meter Storage Applications

Behind-the-Meter Storage (BTMS) is a stationary battery energy storage system that is connected to the electrical distribution system on the customer's side of the utility's service meter. BTMS systems are used to store electrical energy from the grid as well as inconstant, renewable energy, such as local solar and wind generation. A successful BTMS system will allow the customer to pair their energy generation and storage to optimize electrical consumption from the grid, improving reliability and minimizing cost. For BTMS applications, batteries must be designed and optimized with different set of criteria from other leading segments of the Li-ion battery market, like transportation, due the system being stationary and proximal to the residential or commercial building it's benefitting. BTMS applications prioritize safety, cost (low/no-critical materials), reliability (20-year calendar life), and durability (10,000 cycle life), while having the ability to (minimally) compromise energy density and rate capability. Lithium titanate (Li4Ti5O12-, LTO) is a promising anode candidate for BTMS applications due to its high safety and capacity retention, while maintaining a reasonable 160 mAhg-1 reversable capacity and composition of relatively abundant materials. (1) Specifically, LTO has a high working voltage which helps to prevent Li dendrite formation, improving safety. Furthermore, LTO also has negligible lithiation-based volume change, leading to less mechanical pulverization, or loss of active material, upon cycling. For the cathode, materials with little or no Co are of high interest due to the high cost and low abundance of Co. LiMn2O4 (LMO) has been paired with LTO for BTMS applications in the past due to its safety, low cost (abundancy), and reasonably high operating voltage. (2-4) However, the low capacity of LMO limits energy density and specific energy. While not the highest priority for BTMS applications, increasing energy density will enable deployment in space constrained BTMS applications and decrease total cost. LiNi0.9Mn0.1O2 (LN-MO) is a recently developed material with promise due to its high operating voltage and relatively low price. (5) However, Ni-rich layered oxides, including LNMO, tend to struggle with capacity retention during high-voltage cycling due to mechanical pulverization, irreversible phase transitions, and unstable solid-electrolyte interphase. The study presented here focuses on building an understanding of how electrolyte solvent and varied cutoff potentials will impact the cycle life of LTO/LN-MO cells. Specifically, a comparison is provided between ethylene carbonate (EC), ethyl methyl carbonate (EMC), fluoroethylene carbonate (FEC), and Gen2 electrolyte solvents with 1M Lithium hexafluorophosphate (LiPF6) salt, cycling to two upper termination potentials, 2.6V and 2.7V. Electrochemical testing and diagnostics (e.g., differential capacity analysis, area specific impedance, constant voltage hold, and rate capability) and post-mortem characterization will be used to understand the aging behavior and failure mechanisms of the 8 cell combinations (four electrolytes and two voltage cutoffs). Cells with FEC electrolyte showed a lower initial capacity compared to cells with Gen2, EMC, and EC cycling at both voltages; however, the cells with FEC showed consistent trends in capacity retention with 2.6V and 2.7V termination potentials, while the cells with the other electrolytes showed much higher rates of capacity loss when cycling to the higher voltage. These results indicate that FEC may play a role in improving durability of high-voltage, Ni-rich electrode systems for use in high-cycle applications, such as BTMS.

electrolyte↗

DOE OSTI2023

A Hardware and Software Co-design Framework for Energy Efficient Neuromorphic Systems

Neuromorphic systems can be realized by a variety of algorithms and architectures. A common understanding is that spiking neuromorphic designs, which encode information into spatio-temporal spiking events, are both a biologically-accurate and efficient way of processing information. However, representing the information through timing relationships induces sophisticated circuit designs in traditional CMOS-based implementations. In recent years, high-capacity resistive memory (RRAM, aka, memristor) has demonstrated great potential in mimicking synaptic behaviors. Several RRAM-based spiking neuromorphic designs exist, most of which focus on rate coding schemes. These designs simplify circuit implementations of neuron models and explore challenges such as unsatisfactory speed, resolution, and performance. As an alternative, we will explore temporal coding spiking neuromorphic systems that encode information as the relative timing of neuron activations (spikes), which have been proven to be more adaptive and energy-efficient. Developing a neuromorphic system for spiking neural network (SNN) inference and online training, however, faces some major technical challenges: (1) It lacks circuit implementation support for temporal-coding SNN to achieve satisfying power efficiency and accuracy; (2) Although existing research works have investigated memristive synapse and neuron designs for spike-timing-dependent plasticity, the non-ideal conditions in implementation, such as device variations and signal degradation, degrade online learning accuracy of large scale systems; and (3) Non-optimized, inter-layer data traffic in SNNs, leads to unnecessary data communication costs. In this project, we plan to address these challenges by a hardware and software co-design framework that incorporates solutions at the circuit, architecture, and algorithm levels. At the circuit-level, we will elaborate on the in-situ SNN processing element designs for supporting both inference and online training modes. Variation-aware schemes will be studied to improve reliability. At the architecture level, we propose a pipelined, asynchronous architecture to retain the timing resolution of spikes. At the algorithm level, we will investigate an innovative SNN training algorithm for enabling activation sparsification and reducing unnecessary data communication costs. This neuromorphic system will provide an effective solution to real-life energy-constrained applications and significantly contribute to the exploration of next-generation high-performance computing systems under the DOE context.

97 MATHEMATICS AND COMPUTING↗