SEARCH · Search NASA
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.
The present invention relates to genetically modified bacteria and methods of optimizing genetically modified bacteria for the production of a metabolite.
Nuclear power plants collect and store large volumes of heterogeneous data from various components and systems. With recent advances in machine learning (ML) techniques, these data can be leveraged to develop diagnostic and short-term forecasting models to better predict future equipment condition. Maintenance operations can then be planned in advance whenever degraded performance is predicted, thus resulting in fewer unplanned outages and the optimization of maintenance activities. This enables lower maintenance costs and improves the overall economics of nuclear power. This paper focuses on developing a short-term forecasting process that leverages a feature selection process to distill large volumes of heterogeneous data and predict specific equipment parameters. A variety of feature selection methods, including Shapley Additive Explanations (SHAP) and variance inflation factor (VIF), were used to select the optimal features as inputs for three ML methods: long short-term memory (LSTM) networks, support vector regression (SVR), and random forest (RF). Each combination of model and input features was used to predict a pump bearing temperature both 1 and 24 hours in advance, based on actual plant system data. The optimal inputs for the LSTM and SVR were selected using the SHAP values, while the optimal input for the RF consisted solely of the response variable itself. Each model produced similar 1-hour-ahead predictions, with root mean square errors (RMSEs) of roughly 0.006. For the 24-hour-ahead predictions, differences could be seen between LSTM, SVR, and RF, as reflected by model performances of 0.036 +- 0.014, 0.0026 +- 0, and 0.063 +- 0.004 RMSE, respectively. As big data and continuous online monitoring become more widely available, the proposed feature selection process can be used for many applications beyond the prediction of process parameters within nuclear infrastructure.
Abstract The vast size of chemical space necessitates computational approaches to automate and accelerate the design of molecular sequences to guide experimental efforts for drug discovery. Genetic algorithms provide a useful framework to incrementally generate molecules by applying mutations to known chemical structures. Recently, masked language models have been applied to automate the mutation process by leveraging large compound libraries to learn commonly occurring chemical sequences (i.e., using tokenization) and predict rearrangements (i.e., using mask prediction). Here, we consider how language models can be adapted to improve molecule generation for different optimization tasks. We use two different generation strategies for comparison, fixed and adaptive. The fixed strategy uses a pre-trained model to generate mutations; the adaptive strategy trains the language model on each new generation of molecules selected for target properties during optimization. Our results show that the adaptive strategy allows the language model to more closely fit the distribution of molecules in the population. Therefore, for enhanced fitness optimization, we suggest the use of the fixed strategy during an initial phase followed by the use of the adaptive strategy. We demonstrate the impact of adaptive training by searching for molecules that optimize both heuristic metrics, drug-likeness and synthesizability, as well as predicted protein binding affinity from a surrogate model. Our results show that the adaptive strategy provides a significant improvement in fitness optimization compared to the fixed pre-trained model, empowering the application of language models to molecular design tasks.
Computational optimization has revolutionized the field of stellarator design. To date, optimizations have focused primarily on optimization of neoclassical confinement and ideal MHD stability, although limited optimization of other parameters has also been performed. Here, we outline a select set of new concepts for stellarator optimization that, when taken as a group, present a significant step forward in the stellarator concept. One of the criticisms that has been leveled at existing methods of design is the complexity of the resultant field coils. Recently, a new coil optimization code—COILOPT++, which uses a spline instead of a Fourier representation of the coils,—was written and included in the STELLOPT suite of codes. The advantage of this method is that it allows the addition of real space constraints on the locations of the coils. The code has been tested by generating coil designs for optimized quasi-axisymmetric stellarator plasma configurations of different aspect ratios. As an initial exercise, a constraint that the windings be vertical was placed on large major radius half of the non-planar coils. Further constraints were also imposed that guaranteed that sector blanket modules could be removed from between the coils, enabling a sector maintenance scheme. Results of this exercise will be presented. New ideas on methods for the optimization of turbulent transport have garnered much attention since these methods have led to design concepts that are calculated to have reduced turbulent heat loss. We have explored possibilities for generating an experimental database to test whether the reduction in transport that is predicted is consistent with experimental observations. Thus, a series of equilibria that can be made in the now latent QUASAR experiment have been identified that will test the predicted transport scalings. Fast particle confinement studies aimed at developing a generalized optimization algorithm are also discussed. A new algorithm developed for the design of the scraper element on W7-X is presented along with ideas for automating the optimization approach.
The approach to carrying out multi-discipline aerospace design studies in the future, especially in massively parallel computing environments, comprises of choosing (1) suitable solvers to compute solutions to equations characterizing a discipline, and (2) efficient optimization methods. In addition, for aerodynamic optimization problems, (3) smart methodologies must be selected to modify the surface shape. In this research effort, a 'direct' optimization method is implemented on the Cray C-90 to improve aerodynamic design. It is coupled with an existing implicit Navier-Stokes solver, OVERFLOW, to compute flow solutions. The optimization method is chosen such that it can accomodate multi-discipline optimization in future computations. In the work , however, only single discipline aerodynamic optimization will be included.
The possibility of the transfer of a holographic structure that was originally registered on layers of dichromated gelatin (DCG) onto a substrate material of polymethylmethacrylate (PMMA) has been shown. The use of a selective destructive effect of short-wave UV radiation with a wavelength that is less than 270 nm is the basis of the mechanism of formation of secondary relief-phase holographic structure on the surface of PMMA. The optimization of processing modes and selection of developing compositions which are based on isopropanol and methylisobutylketone (MIBK) have been carried out, which made it possible to create reliefphase holographic gratings with high diffraction efficiency (DE) of about 25% and the maximum depth of the surface relief of the order of 1 μm on the substrates of PMMA.
The use of optimization methods as an effective design tool in the design of hybrid vehicle propulsion systems is demonstrated. Optimization techniques were used to select values for three design parameters (battery weight, heat engine power rating and power split between the two on-board energy sources) such that various measures of vehicle performance (acquisition cost, life cycle cost and petroleum consumption) were optimized. The apporach produced designs which were often significant improvements over hybrid designs already reported on in the literature. The principal conclusions are as follows. First, it was found that the strategy used to split the required power between the two on-board energy sources can have a significant effect on life cycle cost and petroleum consumption. Second, the optimization program should be constructed so that performance measures and design variables can be easily changed. Third, the vehicle simulation program has a significant effect on the computer run time of the overall optimization program; run time can be significantly reduced by proper design of the types of trips the vehicle takes in a one year period. Fourth, care must be taken in designing the cost and constraint expressions which are used in the optimization so that they are relatively smooth functions of the design variables. Fifth, proper handling of constraints on battery weight and heat engine rating, variables which must be large enough to meet power demands, is particularly important for the success of an optimization study. Finally, the principal conclusion is that optimization methods provide a practical tool for carrying out the design of a hybrid vehicle propulsion system.
Single-phase inverter or rectifier systems often make use of an active power decoupler (APD) to balance the mismatch between constant dc power and fluctuating ac power. This article deals with the comparison of continuous conduction mode (CCM) and critical conduction mode (CRM) operation-based design of a parallel boost-type APD for photovoltaic microinverter applications. From a design perspective, multiobjective analysis of efficiency, volume, and cost is explored within a decision space including planar inductors, gallium nitride based devices, film capacitors, switching frequency, and modulation (CCM vs. CRM). The theoretical study analyzes all possible design configurations within CCM and CRM and identifies Pareto-optimal designs, from which the selected CRM design can achieve reduced system volume and lower cost with the use of smaller inductor core, while operating with similar California Energy Commission efficiency drop as the selected CCM design. From a control perspective, a pulsewidth modulation based control strategy is proposed to implement closed-loop CRM modulation that does not rely on zero-crossing detection. Furthermore, closed-loop systems are designed for the optimal CCM and CRM realizations, and the final system characteristics are compared. Experimental results, obtained using two separate 40-V, 400-W hardware prototypes for CCM and CRM, are presented to verify the analyses.
Coal Combustion Residual (CCR) landfills in the U.S. can lead to elevated contaminant concentrations in groundwater and runoff, particularly of arsenic, boron, and selenium. Zeolites can be readily synthesized from materials contained within the coal combustion ash impoundments and can be cation-exchanged to enhance adsorption capacity, selectivity, and reactivity. However, the optimization of zeolites becomes a daunting task when accounting for the variety of Si:Al ratios, the species of extra-framework cations present, and the zeolite pore structure. Molecular simulations provide methods to study and guide the design of zeolites for the sorption of contaminants from aqueous solution. In this work, models that predict the sorption of arsenic, boron, and selenium from water using cation-exchanged zeolites. Because of the lack of experimental adsorption data for these species of contaminants, models were parametrized to reproduce the results of first-principles calculations and then used to predict sorption for zeolites for a dataset containing 6000 combinations of zeolites and sorption conditions. Machine learning was used to train a model to predict sorption for materials in this database based on the results of the molecular simulations. Next, a genetic algorithm was used to optimize zeolites for the removal of each contaminant from aqueous solution for individual impoundment sites based on contaminant concentrations reported by the Electric Power Research Institute.
A method for controlling automated driving operations of a vehicle includes determining vehicle origin and destination data, and generating a graphical representation of a road network with multiple candidate routes between the vehicle's origin and destination. Road-level data, including speed, turn angle, and/or gradient data, is received for each candidate route, and respective total energy uses are estimated for the vehicle to traverse across the candidate routes. Multiple candidate driving strategies, each having respective speed and acceleration profiles, are determined for the candidate route with the lowest estimated total energy use. An optimal candidate driving strategy is selected through a cost evaluation of the associated speed and acceleration profiles and forward movement simulations of the vehicle over a prediction horizon. Command signals are transmitted to the vehicle's steering and/or powertrain systems to execute control operations based on the optimal driving strategy and the candidate route with the lowest energy use.
Proposed Effort under Advanced Air Transport Technology (AATT): - Select advanced Al alloy with formability and additive manufacturability, high combined strength and toughness - Fabricate single-piece Integrally Stiffened Cylinders (~10 ft. dia. x 10 ft. long) with longitudinal stiffeners - Demonstrate joining technology for assembly of two cylinders into 20-ft. long fuselage structure - Optimize structural design incorporating tailored, selective reinforcement for durability and damage tolerance - Build-up window, door and circumferential ring frames with large-scale additive manufacturing - Structural test to validate manufacturing demonstration, structural models and analysis
The high performance computing (HPC) community is in a period of transition. The rise of AI/ML coupled with a changing landscape of resources deems portability a new metric of performance, and methods to move between on-premises and cloud environments and assess compatibility are paramount. Here we design and test a strategy for bridging the gap between traditional HPC and Kubernetes environments – first containerizing applications, providing automated orchestration to run studies, and packaging the setup with automated means to assess performance using low overhead eXtended Berkeley Packet Filter (eBPF) programs. We first assess different designs for eBPF collection, demonstrating a tradeoff between number of programs deployed on a node and overhead added. We develop 5 low overhead eBPF programs that combine with streaming ML models to assess CPU, futex, TCP, shared memory, and file access across four different builds of an HPC application for CPU and GPU. We use eBPF data to generate insights into the possible underlying etiology of scaling issues. We then assess compatibility of a well-known benchmark, HPCG, across matrices of micro-architectures and optimization levels (217 containers across 24 instance types and over 7500 runs). We provide to the community 30 applications to deploy in our automated setup and perform a scaling study from 4 to a maximum of 256 nodes for both CPU and GPU applications. Finally, we use our gained knowledge about performance to generate compatibility artifacts that are used by a newly developed Kubernetes controller to intelligently select instance type based on optimizing a figure of merit. Along with insights to scaling in this environment with a collection of applications and templates to work from, we provide an overall strategy for approaching HPC application deployment and image selection based on compatibility in cloud.
Medium‐voltage back‐to‐back (MVB2B) converters can connect two distribution systems and quantifiably transfer power between them. This function can enable the MVB2B converter to exchange distributed energy resource (DER)‐generated power between two systems and bring significant value to enhancing distribution system DER adoption. Our previous work analysed and demonstrated the value MVB2B converter can bring to DER integration. As continuous work, this paper presents a methodology that helps address the MVB2B converter sizing and location selection problem in distribution systems with high DER penetrations. The proposed methodology aims to address three critical problems for MVB2B converter implementation in the real world: (1) which distribution systems are better to be connected, (2) what converter size is appropriate for connecting the distribution systems, and (3) where the optimal connection points are in the systems for connecting the MVB2B converter. The proposed methodology has been demonstrated by case studies that include various scenarios involving distribution systems with different dominated load types and high photovoltaic penetrations. The results demonstrate that selecting the optimal converter size based on net revenue and time of return considerations leads to a balance between maximizing energy savings and minimizing financial payback periods. Furthermore, feeder pair selection based on load profile standard deviation effectively identifies systems that derive the greatest value from MVB2B integration. Finally, an optimized connection point selection approach using a voltage load sensitivity matrix ensures minimal system impact while facilitating efficient power exchange. These findings provide practical insights for the real‐world deployment of MVB2B converters to enhance DER hosting capacity and improve grid resilience.
We introduce an approach for finding individual excited states' energy stationary points in complete active space self-consistent field theory that is compatible with standard optimization methods and highly effective at overcoming difficulties due to root flipping and near-degeneracies. Inspired by both the maximum overlap method and recent progress in excited-state variational principles, our approach combines these ideas in order to track individual excited states throughout the orbital optimization process. In a series of tests involving root flipping, near-degeneracies, charge transfers, and double excitations, we show that this approach is more effective for state-specific optimization than either the naive selection of roots on the basis of energy ordering or a more direct generalization of the maximum overlap method. We offer evidence that this state-specific approach improves the performance of complete active space perturbation theory for vertical excitation energies. Furthermore, we find that the state-specific optimization can help avoid state-averaging-induced discontinuities on potential energy surfaces. With a simple implementation, a low cost, and compatibility with large active space methods, the approach is designed to be useful in a wide range of excited-state investigations.
We present the development of a photometrically selected massive galaxy catalog, targeting Luminous Red Galaxies (LRGs) and massive blue galaxies at redshifts of z⩾0.55. Massive galaxy candidates are selected using infrared/optical color–color cuts, with optical data from the Sloan Digital Sky Survey (SDSS) and infrared data from “unWISE” forced photometry derived from the Wide-field Infrared Survey Explorer (WISE). The selection method is based on previously developed techniques to select LRGs with z>0.5, and is optimized using receiver operating characteristic curves. The catalog contains 16,191,145 objects, selected over the full SDSS DR10 footprint. The redshift distribution of the resulting catalog is estimated using spectroscopic redshifts from the DEEP2 Galaxy Redshift Survey and photometric redshifts from COSMOS. Restframe U − B colors from DEEP2 are used to estimate LRG selection efficiency. Using DEEP2, the resulting catalog has an average redshift of z = 0.65, with a standard deviation of σ=2.0, and an average restframe of U−B=1.0, with a standard deviation of σ=0.27. Using COSMOS, the resulting catalog has an average redshift of z = 0.60, with a standard deviation of σ=1.8. We estimate 34% of the catalog to be blue galaxies with z⩾0.55. An estimated 9.6% of selected objects are blue sources with redshift z<0.55. Stellar contamination is estimated to be 1.8%.
This report describes the Phase 1 process and analysis used to select a refrigerant and thermodynamic cycle as the basis of a vapor compression heat pump requiring a high temperature lift, then to perform a preliminary design to implement the selected concept, including major component selection. Use of a vapor compression heat pump versus other types was based on prior work performed for the Electric Power Research Institute. A high lift heat pump is needed to enable a thermal control system to remove heat down to 275 K from a habitable volume when the external thermal environment is severe. For example, a long-term lunar base habitat will reject heat from a space radiator to a 325 K environment. The first step in the selection process was to perform an optimization trade study, quantifying the effect of radiator operating temperature and heat pump efficiency on total system mass; then, select the radiator operating temperature corresponding to the lowest system mass. Total system mass included radiators, all heat pump components, and the power supply system. The study showed that lunar night operation, with no temperature lift, dictated the radiator size. To operate otherwise would require a high mass penalty to store power. With the defined radiation surface, and heat pump performances assumed to be from 40 percent to 60 percent of the Carnot ideal, the optimum heat rejection temperature ranged from 387 K to 377 K, as a function of heat pump performance. Refrigerant and thermodynamic cycles were then selected to best meet the previously determined design conditions. The system was then adapted as a ground-based prototype lifting temperature to 360 K (versus 385 K for flight unit) and using readily available commercial-grade components. Over 40 refrigerants, separated into wet and dry compression behavioral types, were considered in the selection process. Refrigerants were initially screened for acceptable critical temperature. The acceptable refrigerants were analyzed in ideal single and two-stage thermodynamic cycles. Top candidates were analyzed assuming realistic component limits and system pressure drops, and were evaluated for other considerations such as safety, environmental impact, and commercial availability. A maximum coefficient of performance (COP) of 56 percent of the Carnot ideal was achievable for a three-stage CFC-11 cycle operating under the flight conditions above. The program was completed by defining a control scheme and by researching and selecting the major components, compressor and heat exchangers, that could be used to implement the thermodynamic cycle selected. Special attention was paid to using similar technologies for the SIRF and flight heat pumps resulting in the commercially available equivalent of the flight unit. A package concept was generated for the components selected and mass and volume estimated.
A computational simulation method is presented for Non-Deterministic Multidisciplinary Optimization of engine composite materials and structures. A hypothetical engine duct made with ceramic matrix composites (CMC) is evaluated probabilistically in the presence of combined thermo-mechanical loading. The structure is tailored by quantifying the uncertainties in all relevant design variables such as fabrication, material, and loading parameters. The probabilistic sensitivities are used to select critical design variables for optimization. In this paper, two approaches for non-deterministic optimization are presented. The non-deterministic minimization of combined failure stress criterion is carried out by: (1) performing probabilistic evaluation first and then optimization and (2) performing optimization first and then probabilistic evaluation. The first approach shows that the optimization feasible region can be bounded by a set of prescribed probability limits and that the optimization follows the cumulative distribution function between those limits. The second approach shows that the optimization feasible region is bounded by 0.50 and 0.999 probabilities.
This chapter reviews the basics of derivative-free optimization methods based on surrogate models and outlines how these methods can straightforwardly be applied to autonomously steering experimentation. It summarizes general solution approaches that use surrogate models and active learning. Surrogate modeling is often combined with active learning strategies, where in each iteration of the optimization algorithm, the surrogate model is used to identify which new inputs should be evaluated next and given the new input-output pair, the surrogate model is updated. Regardless of feasibility, the surrogate models for the constraints are updated in each iteration of the optimization algorithm, while the surrogate model for the objective function is only updated when a feasible point has been found. Similarly to the case of computationally cheap constraints, the surrogate models of the constraints should be incorporated into the definition of the auxiliary optimization problem that is solved to select new sample points.