Search NASA⌕ Search

SEARCH · Search NASA

Results for “online distributed 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 163 records · Page 9

Tula: Optimizing Time, Cost, and Generalization in Distributed Large-Batch Training

Distributed training increases the number of batches processed per iteration either by scaling-out (adding more nodes) or scaling-up (increasing the batch-size). However, the largest configuration does not necessarily yield the best performance. Horizontal scaling introduces additional communication overhead, while vertical scaling is constrained by computation cost and device memory limits. Thus, simply increasing the batch-size leads to diminishing returns: training time and cost decrease initially but eventually plateaus, creating a knee-point in the time/cost vs. batch-size pareto curve. The optimal batch-size therefore depends on the underlying model, data and available compute resources. Large batches also suffer from worse model quality due to the well-known “generalization gap”. In this paper, we present Tula, an online service that automatically optimizes time, cost, and convergence quality for large-batch training of convolutional models. It combines parallel-systems modeling with statistical performance prediction to identify the optimal batchsize. Tula predicts training time and cost within 7.5−14% error across multiple models, and achieves up to 20× overall speedup and improves test accuracy by ≈9% on average over standard large-batch training on various vision tasks, thus successfully mitigating the generalization gap and accelerating training at the same time.

Tyagi, Sahil [ORNL] (ORCID:0009000783144745)↗

Next Generation Logistics Systems for Delivering Optimal Biomass Feedstocks to Biorefining Industries in the Southeastern U.S.

The diverse portfolio of biomass sources that is available in the Southeastern U.S., including a significant supply of pine “residue”, represents a valuable strategic position for the region. Through blends formulated based on critical properties, this project will take full advantage of the range in biomass properties afforded by the portfolio to produce a consistent, high-performance feedstock for the industry, while lowering cost. Key developments being targeted to enable this potential include whole-tree transport to a state-of-the-art merchandising depot that will further access biomass from ongoing, forest industry operations. The approach will more effectively utilize the tree and distribute cost, while minimizing in-woods contamination of the woody biomass component. To implement this vision, information on the chemical composition and changes that are induced during multiple preprocessing steps (size reduction, moisture removal, densification, etc.) is needed. New NIR sensor technology will be developed for online monitoring of important biomass properties. The data will be incorporated into a statistical process control platform to improve process efficiency and meet required specifications. Advanced process models are being developed to inform the techno-economic and life-cycle assessment of the program’s impact. The new system will ultimately reduce operational risks from supply chain disruptions, and allow operation of larger-scale biorefineries.

09 BIOMASS FUELS↗

Atikokan Digital Twin: Machine learning in a biomass energy system

The Atikokan Generating Station, operated by Ontario Power Generation, has a 200 MW, biomass-fired tower boiler that operates on a dispatch schedule with a five-minute cycle. The boiler is generally operated in the range of 40–100 MW using two of five burner levels. In order to optimize boiler performance, we propose the implementation of a unique digital twin. Our digital twin abstraction couples Bayesian inference from science-based models and from observations (machine learning) with decision theory to predict operating-variable set points that optimize the physical asset (the boiler) in the presence of uncertainty (artificial intelligence). We focus this paper on the continuous Bayesian machine learning part of the Atikokan Digital Twin; we discuss decision theory in a companion paper. We identify and learn about 12 operational, model, and measured-output parameters and their uncertainties from high-fidelity, science-based simulations of the Atikokan boiler and from the observed measurements at the power plant. Since the goal of the Atikokan Digital Twin is to implement it online in real time, we require fast function evaluations for the quantities of interest extracted from the simulations in the Bayesian analysis. We use Gaussian process regression/interpolation to create accurate, robust surrogate models. We define the Bayesian priors and likelihood function and solve for the posterior distributions of the 12 parameters. Here we then propagate these distributions (i.e., parameters with uncertainty) into the predicted distributions of 790 quantities of interest to learn about the relative importance of various sources of error including experimental, model, and operating-parameter errors.

09 BIOMASS FUELS↗

Optimal Bayesian supervised domain adaptation for RNA sequencing data

Abstract Motivation When learning to subtype complex disease based on next-generation sequencing data, the amount of available data is often limited. Recent works have tried to leverage data from other domains to design better predictors in the target domain of interest with varying degrees of success. But they are either limited to the cases requiring the outcome label correspondence across domains or cannot leverage the label information at all. Moreover, the existing methods cannot usually benefit from other information available a priori such as gene interaction networks. Results In this article, we develop a generative optimal Bayesian supervised domain adaptation (OBSDA) model that can integrate RNA sequencing (RNA-Seq) data from different domains along with their labels for improving prediction accuracy in the target domain. Our model can be applied in cases where different domains share the same labels or have different ones. OBSDA is based on a hierarchical Bayesian negative binomial model with parameter factorization, for which the optimal predictor can be derived by marginalization of likelihood over the posterior of the parameters. We first provide an efficient Gibbs sampler for parameter inference in OBSDA. Then, we leverage the gene-gene network prior information and construct an informed and flexible variational family to infer the posterior distributions of model parameters. Comprehensive experiments on real-world RNA-Seq data demonstrate the superior performance of OBSDA, in terms of accuracy in identifying cancer subtypes by utilizing data from different domains. Moreover, we show that by taking advantage of the prior network information we can further improve the performance. Availability and implementation The source code for implementations of OBSDA and SI-OBSDA are available at the following link. https://github.com/SHBLK/BSDA. Supplementary information Supplementary data are available at Bioinformatics online.

Biochemistry & Molecular Biology↗

Grid Optimization Competition Challenge 3 Problem Formulation

This report contains the problem formulation for the Grid Optimization (GO) Competition Challenge 3. The Grid Optimization Competition is run by a team of researchers from a number of organizations, including the sponsor Advanced Research Projects Agency - Energy (ARPA-E), lead organization Pacific Northwest National Laboratory (PNNL), and technical contributors from Los Alamos National Laboratory (LANL), National Renewable Energy Laboratory (NREL), Texas A&M University (TAMU), Georgia Institute of Technology (GT), University of Wisconsin (UW), and others. The GO Competition poses challenge problems in the field of power grid management, invites entrants to develop solvers for these problems, invokes the solvers on a set of problem instances using common hardware, ranks the solvers according to their performance, and awards prizes according to the rankings. The overall goal of the GO Competition is to spur innovative research on high impact and computationally challenging problems in power grid management from initial development through commercial deployment. Complete information about the GO Competition can be found online at [2]. The webpage covers previous Challenges, rules, timeline, registration information, data formats, scoring methods, computational platform information, information on supported solvers and languages, sponsor information, frequently asked questions, administrator contact information, publicly available problem instances, computer code for reading and evaluating problem and solution data, a sandbox for testing solvers, and a solver submission interface, results, and publications.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Optimizing the Accelerated Recursive Doubling Algorithm for Block Tridiagonal Systems of Equations

The need to solve block tridiagonal systems with hundreds or thousands of right-hand sides for the same block tridiagonal matrix is common in a variety of disciplines. To meet this need, the Accelerated Recursive Doubling Algorithm was developed. After a right-hand side independent phase, the algorithm allows for the quick, online calculation of solutions for different right-hand sides. In this work, we present methods to optimize the Accelerated Recursive Doubling Algorithm in memory usage and computation time in a hybrid parallelization model. The right-hand side independent phase of the naïve implementation takes ≥ 11/3 the amount of memory required to store the tridiagonal matrix, while our implementation reduces the fraction to ≈ 5/3 . The right-hand side dependent phase of the naïve implementation takes ≥ 6 times the amount of memory required to store the right-hand side, while our implementation reduces the fraction to ≈ 3. The computation time for the independent phase is reduced to ≈ 2/3 times that of the naïve implementation, while the computation time for the dependent phase is reduced to ≈ 5/9 . With increasing numbers of shared-memory threads q on every distributed processing element, we have O(q) theoretical speedup.

97 MATHEMATICS AND COMPUTING↗

Sensor Placement Optimization using Chama

Continuous or regularly scheduled monitoring has the potential to quickly identify changes in the environment. However, even with low - cost sensors, only a limited number of sensors can be deployed. The physical placement of these sensors, along with the sensor technology and operating conditions, can have a large impact on the performance of a monitoring strategy. Chama is an open source Python package which includes mixed - integer, stochastic programming formulations to determine sensor locations and technology that maximize monitoring effectiveness. The methods in Chama are general and can be applied to a wide range of applications. Chama is currently being used to design sensor networks to monitor airborne pollutants and to monitor water quality in water distribution systems. The following documentation includes installation instructions and examples, description of software features, and software license. The software is intended to be used by regulatory agencies, industry, and the research community. It is assumed that the reader is familiar with the Python Programming Language. References are included for addit ional background on software components. Online documentation, hosted at http://chama.readthedocs.io/, will be updated as new features are added. The online version includes API documentation .

47 OTHER INSTRUMENTATION↗

A Selectivity based approach to Continuous Pattern Detection in Streaming Graphs

Cyber security is one of the most significant technical challenges in current times. Detecting adversarial activities, prevention of theft of intellectual properties and customer data is a high priority for corporations and government agencies around the world. Cyber defenders need to analyze massive-scale, high-resolution network flows to identify, categorize, and mitigate attacks involving networks spanning institutional and national boundaries. Many of the cyber attacks can be described as subgraph patterns, with prominent examples being insider infiltrations (path queries), denial of service (parallel paths) and malicious spreads (tree queries). This motivates us to explore subgraph matching on streaming graphs in a continuous setting. The novelty of our work lies in using the subgraph distributional statistics collected from the streaming graph to determine the query processing strategy. We introduce a ``Lazy Search" algorithm where the search strategy is decided on a vertex-to-vertex basis depending on the likelihood of a match in the vertex neighborhood. We also propose a metric named ``Relative Selectivity" that is used to select between different query processing strategies. Our experiments performed on real online news, network traffic stream and a synthetic social network benchmark demonstrate 10-100x speedups over non-incremental, selectivity agnostic approaches.

dynamic graphs↗

SU-F-J-66: Anatomy Deformation Based Comparison Between One-Step and Two-Step Optimization for Online ART

Purpose: This study investigated that how the quality of adapted plan was affected by inter-fractional anatomy deformation by using one-step and two-step optimization for on line adaptive radiotherapy (ART) procedure. Methods: 10 lung carcinoma patients were chosen randomly to produce IMRT plan by one-step and two-step algorithms respectively, and the prescribed dose was set as 60 Gy on the planning target volume (PTV) for all patients. To simulate inter-fractional target deformation, four specific cases were created by systematic anatomy variation; including target superior shift 0.5 cm, 0.3cm contraction, 0.3 cm expansion and 45-degree rotation. Based on these four anatomy deformation, adapted plan, regenerated plan and non-adapted plan were created to evaluate quality of adaptation. Adapted plans were generated automatically by using one-step and two-step algorithms respectively to optimize original plans, and regenerated plans were manually created by experience physicists. Non-adapted plans were produced by recalculating the dose distribution based on corresponding original plans. The deviations among these three plans were statistically analyzed by paired T-test. Results: In PTV superior shift case, adapted plans had significantly better PTV coverage by using two-step algorithm compared with one-step one, and meanwhile there was a significant difference of V95 by comparison with adapted and non-adapted plans (p=0.0025). In target contraction deformation, with almost same PTV coverage, the total lung received lower dose using one-step algorithm than two-step algorithm (p=0.0143,0.0126 for V20, Dmean respectively). In other two deformation cases, there were no significant differences observed by both two optimized algorithms. Conclusion: In geometry deformation such as target contraction, with comparable PTV coverage, one-step algorithm gave better OAR sparing than two-step algorithm. Reversely, the adaptation by using two-step algorithm had higher efficiency and accuracy as target occurred position displacement. We want to thank Dr. Lei Xing and Dr. Yong Yang in the Stanford University School of Medicine for this work. This work was jointly supported by NSFC (61471226), Natural Science Foundation for Distinguished Young Scholars of Shandong Province (JQ201516), and China Postdoctoral Science Foundation (2015T80739, 2014M551949).

60 APPLIED LIFE SCIENCES↗

Forward Modeling of Large-scale Structure: An Open-source Approach with Halotools

Here, we present the first stable release of Halotools (v0.2), a community-driven Python package designed to build and test models of the galaxy-halo connection. Halotools provides a modular platform for creating mock universes of galaxies starting from a catalog of dark matter halos obtained from a cosmological simulation. The package supports many of the common forms used to describe galaxy-halo models: the halo occupation distribution (HOD), the conditional luminosity function (CLF), abundance matching, and alternatives to these models that include effects such as environmental quenching or variable galaxy assembly bias. Satellite galaxies can be modeled to live in subhalos, or to follow custom number density profiles within their halos, including spatial and/or velocity bias with respect to the dark matter profile. Here, the package has an optimized toolkit to make mock observations on a synthetic galaxy population, including galaxy clustering, galaxy-galaxy lensing, galaxy group identification, RSD multipoles, void statistics, pairwise velocities and others, allowing direct comparison to observations. Halotools is object-oriented, enabling complex models to be built from a set of simple, interchangeable components, including those of your own creation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

LLGOMAX: Enhancing Industry-Standard Tools for AC Optimal Unit Commitment (CRADA Final Report)

This was a collaborative effort between Lawrence Livermore National Security, LLC ("LLNS"), as manager and operator of Lawrence Livermore National Laboratory ("LLNL") and ECCO International Inc. ("Participant"). The team designed and implemented a new approach for the Security-Constrained Alternating-Current Unit Commitment (SCACUC) problem. The SCACUC problem is a mathematical optimization problem that decides which generating units should be online and how much power should be produced (and consumed) at each point of the power grid. The decisions are made so as to minimize the total cost of supplying electricity while respecting technical constraints of the power grid, both under normal and emergency conditions. The team developed a solution for SCACUC as specified in the ARPA-E Grid Optimization Competition (GOC) Challenge 3, which put forth a forward-looking version of the problem, which many features not present in today’s electricity market specifications for SCACUC.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Accelerating Asymptotically Exact MCMC for Computationally Intensive Models via Local Approximations

We construct a new framework for accelerating Markov chain Monte Carlo in posterior sampling problems where standard methods are limited by the computational cost of the likelihood, or of numerical models embedded therein. Our approach introduces local approximations of these models into the Metropolis–Hastings kernel, borrowing ideas from deterministic approximation theory, optimization, and experimental design. Previous efforts at integrating approximate models into inference typically sacrifice either the sampler’s exactness or efficiency; our work seeks to address these limitations by exploiting useful convergence characteristics of local approximations. We prove the ergodicity of our approximate Markov chain, showing that it samples asymptotically from the exact posterior distribution of interest. We describe variations of the algorithm that employ either local polynomial approximations or local Gaussian process regressors. Our theoretical results reinforce the key observation underlying this article: when the likelihood has some local regularity, the number of model evaluations per Markov chain Monte Carlo (MCMC) step can be greatly reduced without biasing the Monte Carlo average. Numerical experiments demonstrate multiple order-of-magnitude reductions in the number of forward model evaluations used in representative ordinary differential equation (ODE) and partial differential equation (PDE) inference problems, with both synthetic and real data. Supplementary materials for this article are available online.

97 MATHEMATICS AND COMPUTING↗

Lossy Compression: An Online Multi-Stage Technology for High-Fidelity Synchro- Waveform Measurements

Effective real-time monitoring and analysis of distributed grids necessitate the use of synchro-waveform measurements, which capture almost all high-frequency disturbances and transient phenomena. However, due to limitations in high-speed measurements and network bandwidth, it is challenging to transfer all high-fidelity synchro-waveforms losslessly and successfully. To cope with these challenges, a hybrid-based online multi-stage compression algorithm is proposed to significantly improve the compression efficiency for synchro-waveform measurements. Initially, the multiple discrete Wavelet transformation is deployed to deconstruct the waveform components. The delta encoding is further developed to decrease the magnitude. In conjunction with the Lempel-Ziv-Markov chain, the hybrid compression algorithm is implemented to achieve real-time compression for the synchro-waveform measurements. Moreover, an innovative error index that synergizes the time and frequency domain error and correlation is formulated to evaluate the waveform distortion. By integrating compression ratio, suitable parameters can be optimally selected. Finally, the simulation, laboratory experiments, as well as field tests across a spectrum of sampling frequencies and time intervals are conducted to substantiate the efficacy of the proposed method. Here, the outcomes demonstrated that a compression ratio of approximately 15.5 and 17.83 can be reached for 0.5 s and 1 s data under both offline and online scenarios, which equates to a substantial 93.5% to 94.39% reduction in data storage requirements.

High-fidelity synchro-waveform measurements↗

Multi-time Scale Coordination of Distributed Energy Resources in Isolated Power Systems

In isolated power systems, including microgrids, distributed assets, such as renewable energy resources (e.g. wind, solar) and energy storage, can be actively coordinated to reduce dependency on fossil fuel generation. The key challenge of such coordination arises from significant uncertainty and variability occurring at small time scales associated with increased penetration of renewables. Specifically, the problem is with ensuring economic and efficient utilization of DERs, while also meeting operational objectives such as adequate frequency performance. One possible solution is to reduce the time step at which tertiary controls are implemented and to ensure feedback and look-ahead capability are incorporated to handle variability and uncertainty. However, reducing the time step of tertiary controls necessitates investigating time-scale coupling with primary controls so as not to exacerbate system stability issues. In this paper, an optimal coordination (OC) strategy, which considers multiple time-scales, is proposed for isolated microgrid systems with a mix of DERs. This coordination strategy is based on an online moving horizon optimization approach. The effectiveness of the strategy was evaluated in terms of economics, technical performance, and computation time by varying key parameters that significantly impact performance. The illustrative example with realistic scenarios on a simulated isolated microgrid test system suggests that the proposed approach is generalizable towards designing multi-time scale optimal coordination strategies for isolated power systems.

Mayhorn, Ebony↗

Toward Statistical Real-Time Power Fault Detection

We propose statistical fault detection methodology based on high-frequency data streams that are becoming available in modern power grids. Our approach can be treated as an online (sequential) change point monitoring methodology. However, due to the mostly unexplored and very nonstandard structure of high-frequency power grid streaming data, substantial new statistical development is required to make this methodology practically applicable. The paper includes development of scalar detectors based on multichannel data streams, determination of data-driven alarm thresholds and investigation of the performance and robustness of the new tools. Due to a reasonably large database of faults, we can calculate frequencies of false and correct fault signals, and recommend implementations that optimize these empirical success rates.

bolted faults↗

Scalable Deep-Learning-Accelerated Topology Optimization for Additively Manufactured Materials

Topology optimization (TO) is a popular and powerful computational approach for designing novel structures, materials, and devices. Two computational challenges have limited the applicability of TO to a variety of industrial applications. First, a TO problem often involves a large number of design variables to guarantee sufficient expressive power. Second, many TO problems require a large number of expensive physical model simulations, and those simulations cannot be parallelized. To address these issues, we propose a general scalable deep-learning (DL) based TO framework, referred to as SDL-TO, which utilizes parallel schemes in high performance computing (HPC) to accelerate the TO process for designing additively manufactured (AM) materials. Unlike the existing studies of DL for TO, our framework accelerates TO by learning the iterative history data and simultaneously training on the mapping between the given design and its gradient. The surrogate gradient is learned by utilizing parallel computing on multiple CPUs incorporated with a distributed DL training on multiple GPUs. The learned TO gradient enables a fast online update scheme instead of an expensive update based on the physical simulator or solver. Using a local sampling strategy, we achieve to reduce the intrinsic high dimensionality of the design space and improve the training accuracy and the scalability of the SDL-TO framework. The method is demonstrated by benchmark examples and AM materials design for heat conduction. The proposed SDL-TO framework shows competitive performance compared to the baseline methods but significantly reduces the computational cost by a speed up of around 8.6x over the standard TO implementation.

Bi, Sirui↗

A Nonlinear Direct Power Controller for a Three-Phase Grid-Connected Inverter with Online Parameter Update for PV Application

This work proposes a nonlinear direct power controller for a three-phase grid-connected inverter for photovoltaic application. Grid-connected converters for solar applications require advanced control architectures to ensure optimal operation and extraction of maximum power from the photovoltaic arrays. Therefore, to ensure this, a nonlinear power controller architecture has been proposed to extract the maximum power from the solar arrays and to achieve the most optimal operation. To further improve the performance, a parameter update scheme has been utilized which is based on model reference adaptive system. The latter helps to accomplish not only good dynamic response but also excellent tracking performance. The overall control come estimation architecture is verified based on computer simulations via MATLAB/Simulink and PLECS domain and various important case study results are presented.

direct power control↗

Exploration and Analysis of Legacy Data at the Hanford Site Using PHOENIX - 19044

Radioactive contaminants pose risks to human health and the environment. These risks are influenced by factors such as disposal/release and transport mechanisms, location in the environment, potential exposure receptors, and the presence of comingled contaminants. Understanding the behavior, transport, and distribution of radionuclides in the environment is essential to facilitating risk-informed decision-making related to remedial design, action, and monitoring. These decisions are based on significant amounts of complex data that must be accessed, integrated, and analyzed to effectively evaluate data and formulate appropriate conclusions and recommendations. Often, data volume, complexity, and legacy access issues pose severe challenges for analysts, regulators, and stakeholders interested in evaluating scientific data to support decision-making. To address these challenges at the U.S. Department of Energy Hanford Site, Pacific Northwest National Laboratory (PNNL) has developed PHOENIX (PNNL-Hanford Online Environmental Information exchange) to facilitate risk-informed decision-making by streamlining investigative and analytical processes through the integration and visualization of previously isolated datasets created under distinct regulatory or operational regimes, each serving a specific purpose. With the large volume of data dispersed across many Hanford environmental databases, finding the correct sources of relevant data, accessing them, and relating them in a meaningful way can prove challenging and time consuming. By analyzing Hanford databases, performing substantial optimization using modern data processing methods and developing extensive queries on the refactored data, PHOENIX enables users to access, integrate, display, visualize, and retrieve data that would have been historically unfeasible to examine. The data and information accessible through PHOENIX help inform remediation-related assessments and evaluations, predict and quantify the risks that radioactive and chemical contaminants pose to human health and the environment, and communicate this information to decision makers responsible for risk-based remediation actions. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗