Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “code 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.

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

At least 1,747 records · Page 97

Extension of SCALE/Sampler’s sensitivity index for the assessment of cross section, fission yield, and decay data uncertainties

Accurate prediction of nuclide compositions and neutron and gamma sources and spectra for fresh and spent nuclear fuel through computational modeling and simulations is the basis for safeguards instruments design, optimization, and calibration, and for validation of the measured responses. The accuracy of these simulations can be significantly impacted by uncertainties in input parameters to the applied computer code. Input parameters are, for example, dimensions, material compositions, and temperatures of the model. Additionally, important but sometimes neglected input parameters for simulations of nuclear systems are the nuclear data that include nuclear reaction cross sections, fission product yields, and decay data. It has been shown that uncertainties of calculated results are dominated by uncertainties of these nuclear data.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

TH-AB-207A-07: Radiation Dose Simulation for a Newly Proposed Dynamic Bowtie Filters for CT Using Fast Monte Carlo Methods

Purpose: Dynamic bowtie filter is an innovative design capable of modulating the X-ray and balancing the flux in the detectors, and it introduces a new way of patient-specific CT scan optimizations. This study demonstrates the feasibility of performing fast Monte Carlo dose calculation for a type of dynamic bowtie filter for cone-beam CT (Liu et al. 2014 9(7) PloS one) using MIC coprocessors. Methods: The dynamic bowtie filter in question consists of a highly attenuating bowtie component (HB) and a weakly attenuating bowtie (WB). The HB is filled with CeCl3 solution and its surface is defined by a transcendental equation. The WB is an elliptical cylinder filled with air and immersed in the HB. As the scanner rotates, the orientation of WB remains the same with the static patient. In our Monte Carlo simulation, the HB was approximated by 576 boxes. The phantom was a voxelized elliptical cylinder composed of PMMA and surrounded by air (44cm×44cm×40cm, 1000×1000×1 voxels). The dose to the PMMA phantom was tallied with 0.15% statistical uncertainty under 100 kVp source. Two Monte Carlo codes ARCHER and MCNP-6.1 were compared. Both used double-precision. Compiler flags that may trade accuracy for speed were avoided. Results: The wall time of the simulation was 25.4 seconds by ARCHER on a 5110P MIC, 40 seconds on a X5650 CPU, and 523 seconds by the multithreaded MCNP on the same CPU. The high performance of ARCHER is attributed to the parameterized geometry and vectorization of the program hotspots. Conclusion: The dynamic bowtie filter modeled in this study is able to effectively reduce the dynamic range of the detected signals for the photon-counting detectors. With appropriate software optimization methods, the accelerator-based (MIC and GPU) Monte Carlo dose engines have shown good performance and can contribute to patient-specific CT scan optimizations.

60 APPLIED LIFE SCIENCES↗

Non-Intrusive Inference Reduced Order Model for Fluids Using Deep Multistep Neural Network

In this effort we propose a data-driven learning framework for reduced order modeling of fluid dynamics. Designing accurate and efficient reduced order models for nonlinear fluid dynamic problems is challenging for many practical engineering applications. Classical projection-based model reduction methods generate reduced systems by projecting full-order differential operators into low-dimensional subspaces. However, these techniques usually lead to severe instabilities in the presence of highly nonlinear dynamics, which dramatically deteriorates the accuracy of the reduced-order models. In contrast, our new framework exploits linear multistep networks, based on implicit Adams–Moulton schemes, to construct the reduced system. The advantage is that the method optimally approximates the full order model in the low-dimensional space with a given supervised learning task. Moreover, our approach is non-intrusive, such that it can be applied to other complex nonlinear dynamical systems with sophisticated legacy codes. We demonstrate the performance of our method through the numerical simulation of a two-dimensional flow past a circular cylinder with Reynolds number Re = 100. The results reveal that the new data-driven model is significantly more accurate than standard projection-based approaches.

97 MATHEMATICS AND COMPUTING↗

A virtual trial framework for quantifying the detectability of masses in breast tomosynthesis projection data

Purpose: Digital breast tomosynthesis (DBT) is a promising breast cancer screening tool that has already begun making inroads into clinical practice. However, there is ongoing debate over how to quantitatively evaluate and optimize these systems, because different definitions of image quality can lead to different optimal design strategies. Powerful and accurate tools are desired to extend our understanding of DBT system optimization and validate published design principles. Methods: The authors developed a virtual trial framework for task-specific DBT assessment that uses digital phantoms, open-source x-ray transport codes, and a projection-space, spatial-domain observer model for quantitative system evaluation. The authors considered evaluation of reconstruction algorithms as a separate problem and focused on the information content in the raw, unfiltered projection images. Specifically, the authors investigated the effects of scan angle and number of angular projections on detectability of a small (3 mm diameter) signal embedded in randomly-varying anatomical backgrounds. Detectability was measured by the area under the receiver-operating characteristic curve (AUC). Experiments were repeated for three test cases where the detectability-limiting factor was anatomical variability, quantum noise, or electronic noise. The authors also juxtaposed the virtual trial framework with other published studies to illustrate its advantages and disadvantages. Results: The large number of variables in a virtual DBT study make it difficult to directly compare different authors' results, so each result must be interpreted within the context of the specific virtual trial framework. The following results apply to 25% density phantoms with 5.15 cm compressed thickness and 500 {mu}m{sup 3} voxels (larger 500 {mu}m{sup 2} detector pixels were used to avoid voxel-edge artifacts): 1. For raw, unfiltered projection images in the anatomical-variability-limited regime, AUC appeared to remain constant or increase slightly with scan angle. 2. In the same regime, when the authors fixed the scan angle, AUC increased asymptotically with the number of projections. The threshold number of projections for asymptotic AUC performance depended on the scan angle. In the quantum- and electronic-noise dominant regimes, AUC behaviors as a function of scan angle and number of projections sometimes differed from the anatomy-limited regime. For example, with a fixed scan angle, AUC generally decreased with the number of projections in the electronic-noise dominant regime. These results are intended to demonstrate the capabilities of the virtual trial framework, not to be used as optimization rules for DBT. Conclusions: The authors have demonstrated a novel simulation framework and tools for evaluating DBT systems in an objective, task-specific manner. This framework facilitates further investigation of image quality tradeoffs in DBT.

60 APPLIED LIFE SCIENCES↗

Study of Data-Driven Mesh-Model Optimization in System Thermal-Hydraulic Simulation

Quantification of nuclear power plant safety risk requires a systematic and yet practical approach to identification of accident scenarios, assessment of their likelihood and consequences. Instrumental to this goal is risk-informed safety margin characterization (RISMC) framework, whose realization requires computationally robust and affordable methods for sufficiently accurate simulation of complex multi-dimensional physical phenomena, such as turbulent and multi-phase flow. The CFD-like correlation-based codes with 3D simulation capability (e.g., GOTHIC) ensure computational efficiency using coarse mesh size and the sub-grid phenomena in the boundary layer that can be captured by adequate constitutive correlations. However, the error sources and user effects on the selection of mesh size and models lead to unpredictable simulation error, while rich High-Fidelity (HF) data from experiments and numerical simulation using validated code or DNS are not fully explored. It would be useful to have a “smart” data-driven multi-scale framework in which the low-resolution models can be “taught” to emulate high-resolution models. The objective of this work is to develop and evaluate a physical-based data-driven mesh-model optimization approach (Optimized Mesh/Model Information System, OMIS) to estimate the simulation error and give advice on the optimized selection of coarse mesh size and Low-Fidelity (LF) models for System Thermal Hydraulic (STH) simulation to achieve accuracy comparable to that of HF models. This approach takes advantages of computational efficiency of coarse-mesh simulation and application of Machine Learning (ML) algorithms.

97 MATHEMATICS AND COMPUTING↗

ITER Port Interspace Pressure Calculations

The ITER ('the way' in Latin) project in Cadarache, France, will operate the largest tokamak ever built to demonstrate the physical and technical feasibility of nuclear fusion as an energy source. The ITER vacuum vessel is equipped with 50 access ports. Each port has an opening in the bio-shield that communicates with a dedicated port cell. During tokamak operation, the bio-shield opening must be closed with a concrete plug to shield the radiation coming from the plasma. This port plug separates the port cell into a port interspace (between vacuum vessel closure lid and port plug) on the inner side and a port cell on the outer side. This paper presents calculations of pressures and temperatures in the ITER port interspace after a double-ended guillotine break (DEGB) of a pipe in the tokamak cooling water system (TCWS). It is assumed that the DEGB occurs during the worst possible conditions-during water baking operation, with water at 523 K (250 deg. C) and 4.4 MPa. These conditions are more severe than during normal tokamak operation with the water at 398 K (125 deg. C) and 2 MPa. Two computer codes are employed in these calculations: RELAP5-3D Version 4.2.1 to calculate blowdown releases from the pipe break, and MELCOR, Version 1.8.6 to calculate pressures and temperatures in the port interspace. A sensitivity study has been performed to optimize the maximum peak pressure in the port interspace volume of ITER. The optimum case was for a labyrinth area (FL131) of 0.5 m{sup 2} and a relief panel (FL135) kept open after the first cycle. This case keeps the pressure in the port interspace under 150 kPa and the temperatures under 385 K. (authors)

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Modeling filters for formation of mono-energetic neutron beams in the research reactor IRT MEPhI and optimization of radiation shielding for liquid-xenon detector

The paper considers formation of mono-energetic neutron beams at the entrance of experimental channels in research reactors for various applications. The problem includes the following steps: 1. Full-scale mathematical model of the research IRT MEPhI was developed for numerical evaluations of neutron spectra and neutron spatial distribution in the area of experimental channels. 2. Modeling of filters in the channel to shift neutron spectrum towards the required mono-energetic line was performed. 3. Some characteristics of neutron beams at the entrance of detector were evaluated. The filter materials were selected. The calculations were carried out with application of the computer code based on the high-precision Monte-Carlo code MCNP. As a result, mathematical model was created for the filter which is able to form mono-energetic (24 keV) neutron beam. The study was carried out within the frames of the research project on development of Russian emission detector with liquid noble gas to observe rare processes of neutrino scattering and particles of hypothetical dark matter in atomic nuclei. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Ground and excited state gradients with end-to-end differentiable semiempirical quantum chemistry

Accurate and efficient gradients of molecular energy with respect to nuclear degrees of freedom are essential for geometry optimization and molecular dynamics, including simulations that go beyond the Born–Oppenheimer regime. A common approach involves deriving analytical formulas for new electronic structure methods, which is often conceptually difficult and requires tedious coding. Here, we implement analytical, semi-numerical, and automatic differentiation (AD)-based gradient pathways for semiempirical Hamiltonian models in the PYSEQM software package, leveraging both graphics processing unit (GPU) and central processing unit (CPU) architectures. We further extend these capabilities to excited states calculated using the configuration interaction singles and time-dependent Hartree–Fock ansätze. We benchmark wall time, peak memory usage, and accuracy across three molecular families of varying chemical complexity, including systems of up to a thousand atoms. For ground-state simulations, analytical and AD gradients achieve near-identical GPU runtimes, while semi-numerical gradients are slower on GPU but remain competitive on CPU. For excited states, both analytical and custom AD approaches using implicit differentiation show similar performance and low memory requirements, whereas gradients with full AD are memory-limited. AD gradients match analytical ones in accuracy across all tested systems, aided by a quaternion-based diatomic frame rotation for two-center quantities that ensures smooth energy surfaces. Overall, automatic differentiation emerges as a practical alternative to analytical gradients in semiempirical quantum chemistry, offering high accuracy while allowing seamless integration in AI-driven workflows and popular packages, such as PyTorch and JAX. Our results provide actionable guidance for selecting optimal gradient strategies in large-scale ground- and excited-state molecular dynamics simulations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Enhanced production of 2,3‐butanediol in pyruvate decarboxylase‐deficient Saccharomyces cerevisiae through optimizing ratio of glucose/galactose

Abstract Galactose and glucose are two of the most abundant monomeric sugars in hydrolysates of marine biomasses. While Saccharomyces cerevisiae can ferment galactose, its uptake is tightly controlled in the presence of glucose by catabolite repression. It is desirable to construct engineered strains capable of simultaneous utilization of glucose and galactose for producing biofuels and chemicals from marine biomass. The MTH1 gene coding for transcription factor in glucose signaling was mutated in a pyruvate decarboxylase (Pdc)‐deficient S. cerevisiae expressing heterologous 2,3‐butanediol (2,3‐BD) biosynthetic genes. The engineered S. cerevisiae strain consumed glucose and galactose simultaneously and produced 2,3‐BD as a major product. Total sugar consumption rates increased with a low ratio of glucose/galactose, though, occurrence of the glucose depletion in a fed‐batch fermentation decreased 2,3‐BD production substantially. Through optimizing the profiles of sugar concentrations in a fed‐batch cultivation with the engineered strain, 99.1 ± 1.7 g/L 2,3‐BD was produced in 143 hours with a yield of 0.353 ± 0.022 g 2,3‐BD/g sugars. This result suggests that simultaneous and efficient utilization of glucose and galactose by the engineered yeast might be applicable to the economical production of not only 2,3‐BD, but also other biofuels and chemicals from marine biomass.

Choi, Eun‐Ji↗

High fidelity nuclear energy system optimization towards an environmentally benign, sustainable, and secure energy source.

A new high-fidelity integrated system method and analysis approach was developed and implemented for consistent and comprehensive evaluations of advanced fuel cycles leading to minimized Transuranic (TRU) inventories. The method has been implemented in a developed code system integrating capabilities of Monte Carlo N - Particle Extended (MCNPX) for high-fidelity fuel cycle component simulations. In this report, a Nuclear Energy System (NES) configuration was developed to take advantage of used fuel recycling and transmutation capabilities in waste management scenarios leading to minimized TRU waste inventories, long-term activities, and radiotoxicities. The reactor systems and fuel cycle components that make up the NES were selected for their ability to perform in tandem to produce clean, safe, and dependable energy in an environmentally conscious manner. The diversity in performance and spectral characteristics were used to enhance TRU waste elimination while efficiently utilizing uranium resources and providing an abundant energy source. A computational modeling approach was developed for integrating the individual models of the NES. A general approach was utilized allowing for the Integrated System Model (ISM) to be modified in order to provide simulation for other systems with similar attributes. By utilizing this approach, the ISM is capable of performing system evaluations under many different design parameter options. Additionally, the predictive capabilities of the ISM and its computational time efficiency allow for system sensitivity/uncertainty analysis and the implementation of optimization techniques.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Sandia National Laboratories Environmental Fluid Dynamics Code: Marine Hydrokinetic Module User’s Manual

This document describes the marine hydrokinetic (MHK) input file and subroutines for the Sandia National Laboratories Environmental Fluid Dynamics Code (SNL-EFDC), which is a combined hydrodynamic, sediment transport, and water quality model based on the Environmental Fluid Dynamics Code (EFDC) developed by John Hamrick, formerly sponsored by the U.S. Environmental Protection Agency, and now maintained by Tetra Tech, Inc. SNL-EFDC has been previously enhanced with the incorporation of the SEDZLJ sediment dynamics model developed by Ziegler, Lick, and Jones. SNL-EFDC has also been upgraded to more accurately simulate algae growth with specific application to optimizing biomass in an open-channel raceway for biofuels production. A detailed description of the input file containing data describing the MHK device/array is provided, along with a description of the MHK FORTRAN routine. Both a theoretical description of the MHK dynamics as incorporated into SNL-EFDC and an explanation of the source code are provided. This user manual is meant to be used in conjunction with the original EFDC and sediment dynamics SNL-EFDC manuals. Through this document, the authors provide information for users who wish to model the effects of an MHK device (or array of devices) on a flow system with EFDC and who also seek a clear understanding of the source code, which is available from staff in the Water Power Technologies Department at Sandia National Laboratories, Albuquerque, New Mexico.

97 MATHEMATICS AND COMPUTING↗

The Electrolyte Genome project: A big data approach in battery materials discovery

We present a high-throughput infrastructure for the automated calculation of molecular properties with a focus on battery electrolytes. The infrastructure is largely open-source and handles both practical aspects (input file generation, output file parsing, and information management) as well as more complex problems (structure matching, salt complex generation, and failure recovery). Using this infrastructure, we have computed the ionization potential (IP) and electron affinities (EA) of 4830 molecules relevant to battery electrolytes (encompassing almost 55,000 quantum mechanics calculations) at the B3LYP/6-31+G(*) level. We describe automated workflows for computing redox potential, dissociation constant, and salt-molecule binding complex structure generation. We present routines for automatic recovery from calculation errors, which brings the failure rate from 9.2% to 0.8% for the QChem DFT code. Automated algorithms to check duplication between two arbitrary molecules and structures are described. We present benchmark data on basis sets and functionals on the G2-97 test set; one finding is that a IP/EA calculation method that combines PBE geometry optimization and B3LYP energy evaluation requires less computational cost and yields nearly identical results as compared to a full B3LYP calculation, and could be suitable for the calculation of large molecules. Our data indicates that among the 8 functionals tested, XYGJ-OS and B3LYP are the two best functionals to predict IP/EA with an RMSE of 0.12 and 0.27 eV, respectively. Application of our automated workflow to a large set of quinoxaline derivative molecules shows that functional group effect and substitution position effect can be separated for IP/EA of quinoxaline derivatives, and the most sensitive position is different for IP and EA. Published by Elsevier B.V

Battery↗

SU-E-T-278: Realization of Dose Verification Tool for IMRT Plan Based On DPM

Purpose: To build a Monte Carlo dose verification tool for IMRT Plan by implementing a irradiation source model into DPM code. Extend the ability of DPM to calculate any incident angles and irregular-inhomogeneous fields. Methods: With the virtual source and the energy spectrum which unfolded from the accelerator measurement data,combined with optimized intensity maps to calculate the dose distribution of the irradiation irregular-inhomogeneous field. The irradiation source model of accelerator was substituted by a grid-based surface source. The contour and the intensity distribution of the surface source were optimized by ARTS (Accurate/Advanced Radiotherapy System) optimization module based on the tumor configuration. The weight of the emitter was decided by the grid intensity. The direction of the emitter was decided by the combination of the virtual source and the emitter emitting position. The photon energy spectrum unfolded from the accelerator measurement data was adjusted by compensating the contaminated electron source. For verification, measured data and realistic clinical IMRT plan were compared with DPM dose calculation. Results: The regular field was verified by comparing with the measured data. It was illustrated that the differences were acceptable (<2% inside the field, 2–3mm in the penumbra). The dose calculation of irregular field by DPM simulation was also compared with that of FSPB (Finite Size Pencil Beam) and the passing rate of gamma analysis was 95.1% for peripheral lung cancer. The regular field and the irregular rotational field were all within the range of permitting error. The computing time of regular fields were less than 2h, and the test of peripheral lung cancer was 160min. Through parallel processing, the adapted DPM could complete the calculation of IMRT plan within half an hour. Conclusion: The adapted parallelized DPM code with irradiation source model is faster than classic Monte Carlo codes. Its computational accuracy and speed satisfy the clinical requirement, and it is expectable to be a Monte Carlo dose verification tool for IMRT Plan. Strategic Priority Research Program of the China Academy of Science(XDA03040000); National Natural Science Foundation of China (81101132)

60 APPLIED LIFE SCIENCES↗

A code-to-code benchmark for magneto-convection in a horizontal duct

Liquid metals and magnetic fields are used in many technical applications such as metallurgy, crystal growth and nuclear fusion reactors. When an electrically conducting fluid moves in a magnetic environment, electric currents and electromagnetic forces are generated that affect velocity and pressure losses in the flow. These magnetohydrodynamic (MHD) interactions have to be investigated to optimize the engineering processes. The characteristics of MHD flows depend on the geometrical configuration, the strength of the applied magnetic field, the electrical properties of fluid and structural materials and the thermal conditions. In the so-called blankets for fusion reactors, where liquid metals are used to breed the plasma fuel component tritium and to extract the generated heat, magneto-convective flows play a crucial role in determining heat and mass transfer. Therefore, the availability of numerical codes to simulate this type of flow is mandatory and their validation is a necessary step to guarantee the reliability of the results. For that reason, a benchmark problem has been defined to simulate liquid metal flows in a horizontal rectangular duct heated from below and exposed to a non-uniform magnetic field. Results obtained by five research groups using different codes are compared.

benchmark↗

WARP

WARP, which can stand for ``Weaving All the Random Particles,'' is a three-dimensional (3D) continuous energy Monte Carlo neutron transport code developed at UC Berkeley to efficiently execute on NVIDIA graphics processing unit (GPU) platforms. WARP accelerates Monte Carlo simulations while preserving the benefits of using the Monte Carlo method, namely, that very few physical and geometrical simplifications are applied. WARP is able to calculate multiplication factors, neutron flux distributions (in both space and energy), and fission source distributions for time-independent neutron transport problems. It can run in both criticality or fixed source modes, but fixed source mode is currently not robust, optimized, or maintained in the newest version. WARP can transport neutrons in unrestricted arrangements of parallelepipeds, hexagonal prisms, cylinders, and spheres. The goal of developing WARP is to investigate algorithms that can grow into a full-featured, continuous energy, Monte Carlo neutron transport code that is accelerated by running on GPUs. The crux of the effort is to make Monte Carlo calculations faster while producing accurate results. Modern supercomputers are commonly being built with GPU coprocessor cards in their nodes to increase their computational efficiency and performance. GPUs execute efficiently on data-parallel problems, but most CPU codes, including those for Monte Carlo neutral particle transport, are predominantly task-parallel. WARP uses a data-parallel neutron transport algorithm to take advantage of the computing power GPUs offer.

Bergmann, RyanM. [University of California - Berke↗

WARP

WARP, which can stand for ``Weaving All the Random Particles,'' is a three-dimensional (3D) continuous energy Monte Carlo neutron transport code developed at UC Berkeley to efficiently execute on NVIDIA graphics processing unit (GPU) platforms. WARP accelerates Monte Carlo simulations while preserving the benefits of using the Monte Carlo method, namely, that very few physical and geometrical simplifications are applied. WARP is able to calculate multiplication factors, neutron flux distributions (in both space and energy), and fission source distributions for time-independent neutron transport problems. It can run in both criticality or fixed source modes, but fixed source mode is currently not robust, optimized, or maintained in the newest version. WARP can transport neutrons in unrestricted arrangements of parallelepipeds, hexagonal prisms, cylinders, and spheres. The goal of developing WARP is to investigate algorithms that can grow into a full-featured, continuous energy, Monte Carlo neutron transport code that is accelerated by running on GPUs. The crux of the effort is to make Monte Carlo calculations faster while producing accurate results. Modern supercomputers are commonly being built with GPU coprocessor cards in their nodes to increase their computational efficiency and performance. GPUs execute efficiently on data-parallel problems, but most CPU codes, including those for Monte Carlo neutral particle transport, are predominantly task-parallel. WARP uses a data-parallel neutron transport algorithm to take advantage of the computing power GPUs offer.

Bergmann, Ryan M.↗

A GPU accelerated mixed-precision Smoothed Particle Hydrodynamics framework with cell-based relative coordinates

Smoothed Particle Hydrodynamics (SPH) is essential for modeling complex large-deformation problems across various applications, requiring significant computational power. A major portion of SPH computation time is dedicated to the Nearest Neighboring Particle Search (NNPS) process. While advanced NNPS algorithms have been developed to enhance SPH efficiency, the potential efficiency gains from modern computation hardware remain underexplored. Here, this study investigates the impact of GPU parallel architecture, low-precision computing on GPUs, and GPU memory management on NNPS efficiency. Our approach employs a GPU-accelerated mixed-precision SPH framework, utilizing low precision float-point 16 (FP16) for NNPS while maintaining high precision for other components. To ensure FP16 accuracy in NNPS, we introduce a Relative Coordinated-based Link List (RCLL) algorithm, storing FP16 relative coordinates of particles within background cells. Our testing results show three significant speedup rounds for CPU-based NNPS algorithms. The first comes from parallel GPU computations, with up to a 1000x efficiency gain. The second is achieved through low-precision GPU computing, where the proposed FP16-based RCLL algorithm offers a 1.5x efficiency improvement over the FP64-based approach on GPUs. By optimizing GPU memory bandwidth utilization, the efficiency of the FP16 RCLL algorithm can be further boosted by 2.7x, as demonstrated in an example with 1 million particles. Our code is released at https://github.com/pnnl/lpNNPS4SPH.

97 MATHEMATICS AND COMPUTING↗

Analytical gradients of the state-average complete active space self-consistent field method with density fitting

An efficient implementation of the state-averaged complete active space self-consistent field (SA-CASSCF) gradients employing density fitting (DF) is presented. The DF allows a reduction both in scaling and prefactors of the different steps involved. The performance of the algorithm is demonstrated on a set of molecules ranging up to an iron-Heme b complex which with its 79 atoms and 811 basis functions is to our knowledge the largest SA-CASSCF gradient computed. For smaller systems where the conventional code could still be used as a reference, both the linear response calculation and the gradient formation showed a clear timing reduction and the overall cost of a geometry optimization is typically reduced by more than one order of magnitude while the accuracy loss is negligible.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗