Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “probabilistic modeling”

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 1,045 records · Page 58

Basic Diagnosis and Prediction of Persistent Contrail Occurrence using High-resolution Numerical Weather Analyses/Forecasts and Logistic Regression. Part II: Evaluation of Sample Models

Previous studies have shown that probabilistic forecasting may be a useful method for predicting persistent contrail formation. A probabilistic forecast to accurately predict contrail formation over the contiguous United States (CONUS) is created by using meteorological data based on hourly meteorological analyses from the Advanced Regional Prediction System (ARPS) and from the Rapid Update Cycle (RUC) as well as GOES water vapor channel measurements, combined with surface and satellite observations of contrails. Two groups of logistic models were created. The first group of models (SURFACE models) is based on surface-based contrail observations supplemented with satellite observations of contrail occurrence. The second group of models (OUTBREAK models) is derived from a selected subgroup of satellite-based observations of widespread persistent contrails. The mean accuracies for both the SURFACE and OUTBREAK models typically exceeded 75 percent when based on the RUC or ARPS analysis data, but decreased when the logistic models were derived from ARPS forecast data.

Duda, David P.↗

Prototypic Enhanced Risk Monitor Framework and Evaluation - Advanced Reactor Technology Milestone: M3AT-15PN2301054

This research report summaries the development and evaluation of a prototypic enhanced risk monitor (ERM) methodology (framework) that includes alternative risk metrics and uncertainty analysis. This updated ERM methodology accounts for uncertainty in the equipment condition assessment (ECA), the prognostic result, and the probabilistic risk assessment (PRA) model. It is anticipated that the ability to characterize uncertainty in the estimated risk and update the risk estimates in real time based on equipment condition assessment (ECA) will provide a mechanism for optimizing plant performance while staying within specified safety margins. These results (based on impacting active component O&M using real-time equipment condition information) are a step towards ERMs that, if integrated with AR supervisory plant control systems, can help control O&M costs and improve affordability of advanced reactors.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

R&D Roadmap for Enhanced Resilient Plant Systems, Metrics, Scenarios, Risk Analysis, and Modeling and Simulation

The nuclear industry and the U.S. Department of Energy are jointly developing accident tolerant fuel (ATF) which has the potential to greatly improve the safety and economic performance of the current fleet of NPPs. The combination of ATF with the optimal use of FLEX equipment (deployed as an industry response to the Fukushima Daiichi Nuclear Power Plant (NPP) accident in 2012), enhancements on plant components and systems, incorporation of augmented cooling systems or new passive cooling systems, and improvements in fuel cycle efficiency, have the potential to render the NPPs more efficient and resilient to events (both internally and externally initiated) that have the potential to challenge the integrity of the fuel or plant containment. In this report, combinations of these plant improvements are called Enhanced Resilient Plant Systems. In order to accelerate the development and deployment of the components that comprise Enhanced Resilient Plant Systems, a well-integrated toolkit that is fuel/clad-, fuel cycle-, systems-, and scenario-centric is required to provide a ready platform for the analysis of the advanced ATF fuel and cladding materials that are being developed, as well as plant systems modifications to achieve enhanced resiliency. Since the deployment of these enhancements will comprise a risk/benefit tradeoff of their impacts on plant safety and economics, the envisioned toolkit will need to be capable to characterizing the multidimensional aspects associated with these evaluations and provide decision-makers with information that can optimize the technological, economic, and safety benefits that are to be obtained. The Risk-Informed Systems Analysis (RISA) Pathway of the DOE Light Water Reactor Sustainability (LWRS) Program initiated several research tasks to develop analytical capabilities to support the industry in the transition to the ATF and Enhanced Resilient Plant Systems. The general idea behind the initiative is the development of an integrated evaluation approach which combines the plant probabilistic risk assessment (PRA) models with Multi-Physics Best Estimate Plus Uncertainty (MP-BEPU) analyses in a seamless fashion. The integrated evaluation framework aims at enabling plant system configuration variations to be studied with speed and precision, including detailed risk and benefit assessments of introducing ATF and Enhanced Resilient Plant Systems into current LWR plants to achieve both safety and operational performance enhancements. The focus of this report is to present an integrated research and development (R&D) roadmap to identify and perform high value evaluations of proposed ATF and Enhanced Resilient Plant Systems concepts to identify both the technical (e.g. benefits to risk, safety, and operational margins) and the economic (i.e. business and cost) elements associated with industry adoption of the technology. The integrated evaluation approach will be developed to support the development, testing, qualification, licensing, and deployment of ATF and Enhanced Resilient Plant Systems technology that is capable of achieving substantial safety and economic improvements as well as timely widespread adoption by the US nuclear industry.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Logistics Modeling for Lunar Exploration Systems

The extensive logistics required to support extended crewed operations in space make effective modeling of logistics requirements and deployment critical to predicting the behavior of human lunar exploration systems. This paper discusses the software that has been developed as part of the Campaign Manifest Analysis Tool in support of strategic analysis activities under the Constellation Architecture Team - Lunar. The described logistics module enables definition of logistics requirements across multiple surface locations and allows for the transfer of logistics between those locations. A key feature of the module is the loading algorithm that is used to efficiently load logistics by type into carriers and then onto landers. Attention is given to the capabilities and limitations of this loading algorithm, particularly with regard to surface transfers. These capabilities are described within the context of the object-oriented software implementation, with details provided on the applicability of using this approach to model other human exploration scenarios. Some challenges of incorporating probabilistics into this type of logistics analysis model are discussed at a high level.

Andraschko, Mark R.↗

MODIFICATION OF THE SPAR-H METHOD TO SUPPORT HRA FOR LEVEL 2 PSA

Currently, available Human Reliability Analysis (HRA) methods were generally developed to support Level 1 Probabilistic Safety Analysis (PSA) models. There has been an increased emphasis placed on Level 2 PSA in recent years; however, the currently used HRA methods are not ideal for this application, including the SPAR-H method. Challenges that will likely be present during a severe accident such as degraded or hazardous operating conditions, shift in control from the main control room to the technical support center, unavailability of instrumentation, and others are not routinely considered for Level 1 HRA analysis. These factors combine to create a much more uncertain condition to be accounted for in the HRA analysis. While the SPAR-H shaping factors were established to support Level 1 HRA, previous studies have shown it may be used for Level 2 HRA analysis as well. The Canadian Nuclear Safety Commission (CNSC) and Idaho National Laboratory (INL) in a joint project are investigating modifications to the SPAR-H method to create more consistency in applying the performance shaping factors used in the method for Level 2 analysis.

99 GENERAL AND MISCELLANEOUS↗

Order priors for Bayesian network discovery with an application to malware phylogeny

Here, Bayesian networks have been used extensively to model and discover dependency relationships among sets of random variables. We learn Bayesian network structure with a combination of human knowledge about the partial ordering of variables and statistical inference of conditional dependencies from observed data. Our approach leverages complementary information from human knowledge and inference from observed data to produce networks that reflect human beliefs about the system as well as to fit the observed data. Applying prior beliefs about partial orderings of variables is an approach distinctly different from existing methods that incorporate prior beliefs about direct dependencies (or edges) in a Bayesian network. We provide an efficient implementation of the partial-order prior in a Bayesian structure discovery learning algorithm, as well as an edge prior, showing that both priors meet the local modularity requirement necessary for an efficient Bayesian discovery algorithm. In benchmark studies, the partial-order prior improves the accuracy of Bayesian network structure learning as well as the edge prior, even though order priors are more general. Our primary motivation is in characterizing the evolution of families of malware to aid cyber security analysts. For the problem of malware phylogeny discovery, we find that our algorithm, compared to existing malware phylogeny algorithms, more accurately discovers true dependencies that are missed by other algorithms.

97 MATHEMATICS AND COMPUTING↗

Distributions of the Mean for Lognormally Distributed Data - 19427

In environmental sciences, a distribution of the mean is often needed for right-skewed data. For example, risk assessments are often based on distributions of the mean to obtain upper confidence limits (UCLs) for analyte concentrations, and, in probabilistic performance assessment (PPA) models, parameters are given distributions rather than deterministic values, and they are usually given distributions of their mean rather than the population distributions to account for the large spatial and temporal domains to which they are applied. How this distribution of the mean is developed depends on the characteristics of the parameter's full distribution and how the parameter relates to its use. In this presentation, I will discuss the implications of a right skew in distribution development. Parameters may vary over orders of magnitude and/or may be right-skewed. The material property of saturated hydraulic conductivity, Ks, is a prime example of such a parameter. Right-skewed parameters are often assumed to follow lognormal distributions. If X follows a lognormal distribution, then log(X) follows a normal distribution. For some right-skewed parameters, practitioners go back and forth between the two forms of the data depending on their needs. For example, it may be preferred to use the untransformed version of X in a model or equation, but it is easier to visualize the parameter values on a logarithmic scale. However, distribution development for the average value of the parameter will differ depending on whether X or log(X) is modeled. This difference is due to Jensen's Inequality, a theorem which states that the expected value of a concave function of a variable is less than or equal to the same function of the expected value of the variable; a logarithm is a concave function. In other words, the mean of log(X) is always less than or equal to the log of the mean of X and therefore the mean estimated from a distribution based on log(X) is less than or equal to the mean estimated from a distribution based on X. If a parameter is naturally considered on a log scale by subject matter experts, then distributions developed for the mean of log(X) may be more consistent with the use of the parameters in the literature than distributions developed for X. The statistical decisions of distribution development will be discussed within the context of right-skewed, lognormal data. These issues include method-of-moments versus maximum likelihood estimation of the distribution parameters and the use of bootstrapping to aid in the visualization of a distribution of the mean. The details surrounding Jensen's Inequality will be explained and visualized within the context of distribution development of a right-skewed parameter such as saturated hydraulic conductivity. (authors)

54 ENVIRONMENTAL SCIENCES↗

Classical versus quantum models in machine learning: insights from a finance application

Although several models have been proposed towards assisting machine learning (ML) tasks with quantum computers, a direct comparison of the expressive power and efficiency of classical versus quantum models for datasets originating from real-world applications is one of the key milestones towards a quantum ready era. Here, we take a first step towards addressing this challenge by performing a comparison of the widely used classical ML models known as restricted Boltzmann machines (RBMs), against a recently proposed quantum model, now known as quantum circuit Born machines (QCBMs). Both models address the same hard tasks in unsupervised generative modeling, with QCBMs exploiting the probabilistic nature of quantum mechanics and a candidate for near-term quantum computers, as experimentally demonstrated in three different quantum hardware architectures to date. To address the question of the performance of the quantum model on real-world classical data sets, we construct scenarios from a probabilistic version out of the well-known portfolio optimization problem in finance, by using time-series pricing data from asset subsets of the S&P500 stock market index. It is remarkable to find that, under the same number of resources in terms of parameters for both classical and quantum models, the quantum models seem to have superior performance on typical instances when compared with the canonical training of the RBMs. Our simulations are grounded on a hardware efficient realization of the QCBMs on ion-trap quantum computers, by using their native gate sets, and therefore readily implementable in near-term quantum devices.

97 MATHEMATICS AND COMPUTING↗

Explainable Bayesian Neural Network for Probabilistic Transient Stability Analysis Considering Wind Energy

While several data-driven models have been developed for transient stability assessment, how to consider the uncertainties from load and renewable generations and provide interpretation of data-driven assessment results are still open. This paper proposes an explainable Bayesian Neural Network (BNN) for probabilistic transient stability assessment (TSA). By extracting the uncertainties from loads and wind farms, the BNN model can make a reliable prediction and quantify the prediction uncertainties. We also develop the Gradient Shap algorithm to make the global and local explanations for the probabilistic TSA model, a significant advantage over existing black-box data-driven methods. Numerical results on the modified IEEE 39-bus system show that the proposed method outperforms the existing methods in terms of prediction accuracy and uncertainty quantification capabilities. The explainability of the proposed method allows system operators to design preventive controls for enhancing system stability.

Bayesian Neural Network↗

TOWARD A NEW KIND OF ASTEROSEISMIC GRID FITTING

Recent developments in instrumentation (e.g., in particular the Kepler and CoRoT satellites) provide a new opportunity to improve the models of stellar pulsations. Surface layers, rotation, and magnetic fields imprint erratic frequency shifts, trends, and other non-random behavior in the frequency spectra. As our observational uncertainties become smaller, these are increasingly important and difficult to deal with using standard fitting techniques. To improve the models, new ways to compare their predictions with observations need to be conceived. In this paper, we present a completely probabilistic (Bayesian) approach to asteroseismic model fitting. It allows for varying degrees of prior mode identification, corrections for the discrete nature of the grid, and most importantly implements a treatment of systematic errors, such as the 'surface effects'. It removes the need to apply semi-empirical corrections to the observations prior to fitting them to the models and results in a consistent set of probabilities with which the model physics can be probed and compared. As an example, we show a detailed asteroseismic analysis of the Sun. We find a most probable solar age, including a 35 {+-} 5 million year pre-main-sequence phase, of 4.591 billion years, and initial element mass fractions of X{sub 0} = 0.72, Y{sub 0} = 0.264, Z{sub 0} = 0.016, consistent with recent asteroseismic and non-asteroseismic studies.

79 ASTRONOMY AND ASTROPHYSICS↗

A Multi-Fidelity Gaussian Process Regression Method for Probabilistic Wind Farm Power Curve Estimation

Accurate estimation of the power curve for wind turbines or wind farms is crucial to ensure their efficient operation and management. However, conventional methods for power curve estimation rely either on expensive and infrequent measurements or on low-quality numerical simulations. Moreover, the majority of previous studies on power curve estimation for wind turbines or wind farms focused on deterministic estimation, which provides a point estimate of the relationship between wind speed and power generation. Nevertheless, the deterministic approach fails to consider the inherent uncertainty associated with wind energy production resulting from varying turbine characteristics. This can lead to inaccurate power generation estimation and suboptimal decisions regarding energy management. In this paper, a kernel density estimation (KDE) based Multi-Fidelity Gaussian Process Regression (MFGPR) model is proposed to fuse theoretical power curve data and the ground true measurements to create a mapping of wind speed and wind power. By conducting a case study on an actual wind farm in China, the efficacy of the proposed MFGPR model was demonstrated in characterizing the variability of wind power. The probabilistic MFGPR model was also able to generate confidence intervals that encompassed the measured power, thereby improving the accuracy and confidence in wind power estimation or wind resource assessment. Overall, the proposed MFGPR model offers a reliable approach to integrate high-fidelity ground measurements and theoretical power curve data, resulting in precise wind resource assessment and power estimation.

Gaussian process regression↗

Application of Probabilistic Analysis to Aircraft Impact Dynamics

Full-scale aircraft crash simulations performed with nonlinear, transient dynamic, finite element codes can incorporate structural complexities such as: geometrically accurate models; human occupant models; and advanced material models to include nonlinear stressstrain behaviors, laminated composites, and material failure. Validation of these crash simulations is difficult due to a lack of sufficient information to adequately determine the uncertainty in the experimental data and the appropriateness of modeling assumptions. This paper evaluates probabilistic approaches to quantify the uncertainty in the simulated responses. Several criteria are used to determine that a response surface method is the most appropriate probabilistic approach. The work is extended to compare optimization results with and without probabilistic constraints.

Lyle, Karen H.↗

Probabilistic impact of electricity tariffs on distribution grids considering adoption of solar and storage technologies

This study models the role of electricity tariffs on the long-term adoption of photovoltaic and storage technologies as well as the consequent impact on the distribution grid. An adoption model that captures the economic rationality of tariff-driven investments and considers the stochastic nature of individual consumers’ decisions is proposed. This model is then combined with a probabilistic load flow to evaluate the long-term impacts of the adoption on the voltage profiles of the distribution grid. To illustrate the methodology, different components of the electricity tariffs, including solar compensation mechanisms and time differentiation of Time-of-Use (ToU) rates, are evaluated, using a case study involving a section of a medium-voltage network with 118 nodes.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Towards violation of Born's rule: description of a simple experiment

Recently a new model with hidden variables of the wave type was elaborated, so called prequantum classical statistical field theory (PCSFT). Roughly speaking PCSFT is a classical signal theory applied to a special class of signals - 'quantum systems'. PCSFT reproduces successfully all probabilistic predictions of QM, including correlations for entangled systems. This model peacefully coexists with all known no-go theorems, including Bell's theorem. In our approach QM is an approximate model. All probabilistic predictions of QM are only (quite good) approximations of 'real physical averages'. The latter are averages with respect to fluctuations of prequantum fields. In particular, Born's rule is only an approximate rule. More precise experiments should demonstrate its violation. We present a simple experiment which has to produce statistical data violating Born's rule. Since the PCSFT-presentation of this experiment may be difficult for experimenters, we reformulate consequences of PCSFT in terms of the conventional wave function. In general, deviation from Born's rule is rather small. We found an experiment amplifying this deviation. We start with a toy example. Then we present a more realistic example based on Gaussian states with very small dispersion.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Defect Detection of Industrial Radiography Images of Ammonia Pipes by a Sparse Coding Model

Pipeline transportation systems for liquid anhydrous ammonia require periodic inspections for pipe defects. Defects such as crack-like flaws, including those due to stress corrosion and fatigue crack growth and fatigue of welded joints, can be tested by in situ radiography testing (RT). It is required that the detection technique should reveal the defect region without any significant distortion. Imaging noise such as that due to radiation scattering reduces contrast over the whole or part of the image, causing a reduction in image quality. Various image processing methods can be utilized to improve the outcome of RT of pipes. The sparse coding model is a powerful contrast improvement algorithm, but it tends to eliminate imaging details, passing them off as noise. We present a novel implementation of the sparse coding model based on a probabilistic interpretation to improve defect detection in ammonia pipelines radiography images. The performance of this procedure was evaluated in radiographic images of defect and corrosion regions for different pipe types. For ammonia pipes, the technique was found to offer a significant improvement in defect detection while preserving imaging details.

42 ENGINEERING↗

Cost-Effectiveness Analysis of Preoperative Versus Postoperative Radiation Therapy in Extremity Soft Tissue Sarcoma

Purpose: Surgery combined with radiation therapy (RT) is the cornerstone of multidisciplinary management of extremity soft tissue sarcoma (STS). Although RT can be given in either the preoperative or the postoperative setting with similar local recurrence and survival outcomes, the side effect profiles, costs, and long-term functional outcomes are different. The aim of this study was to use decision analysis to determine optimal sequencing of RT with surgery in patients with extremity STS. Methods and Materials: A cost-effectiveness analysis was conducted using a state transition Markov model, with quality-adjusted life years (QALYs) as the primary outcome. A time horizon of 5 years, a cycle length of 3 months, and a willingness-to-pay threshold of $50,000/QALY was used. One-way deterministic sensitivity analyses were performed to determine the thresholds at which each strategy would be preferred. The robustness of the model was assessed by probabilistic sensitivity analysis. Results: Preoperative RT is a more cost-effective strategy ($26,633/3.00 QALYs) than postoperative RT ($28,028/2.86 QALYs) in our base case scenario. Preoperative RT is the superior strategy with either 3-dimensional conformal RT or intensity-modulated RT. One-way sensitivity analyses identified the relative risk of chronic adverse events as having the greatest influence on the preferred timing of RT. The likelihood of preoperative RT being the preferred strategy was 82% on probabilistic sensitivity analysis. Conclusions: Preoperative RT is more cost effective than postoperative RT in the management of resectable extremity STS, primarily because of the higher incidence of chronic adverse events with RT in the postoperative setting.

61 RADIATION PROTECTION AND DOSIMETRY↗

Uncertainty quantification for Bayesian active learning in rupture life prediction of ferritic steels

Abstract Three probabilistic methodologies are developed for predicting the long-term creep rupture life of 9–12 wt%Cr ferritic-martensitic steels using their chemical and processing parameters. The framework developed in this research strives to simultaneously make efficient inference along with associated risk, i.e., the uncertainty of estimation. The study highlights the limitations of applying probabilistic machine learning to model creep life and provides suggestions as to how this might be alleviated to make an efficient and accurate model with the evaluation of epistemic uncertainty of each prediction. Based on extensive experimentation, Gaussian Process Regression yielded more accurate inference ( $$Pearson\;correlation\;coefficent> 0.95$$ P e a r s o n c o r r e l a t i o n c o e f f i c e n t > 0.95 for the holdout test set) in addition to meaningful uncertainty estimate (i.e., coverage ranges from 94 to 98% for the test set) as compared to quantile regression and natural gradient boosting algorithm. Furthermore, the possibility of an active learning framework to iteratively explore the material space intelligently was demonstrated by simulating the experimental data collection process. This framework can be subsequently deployed to improve model performance or to explore new alloy domains with minimal experimental effort.

20 FOSSIL-FUELED POWER PLANTS↗

A Bayesian Approach to Real-Time Dynamic Parameter Estimation Using Phasor Measurement Unit Measurement

Here, we develop a polynomial-chaos-expansion (PCE)-based approach for decentralized dynamic parameter estimation through Bayesian inference. Using this approach, the non-Gaussian distribution of the inverted parameters is obtained. More specifically, we first represent the decentralized generator model with the PCE-based surrogate. This surrogate allows us to efficiently evaluate the time-consuming dynamic solver at parameter values through Metropolis-Hastings (M-H)-based Markov chain Monte Carlo (MCMC). Then, we propose a two-stage hybrid Markov chain Monte Carlo (MCMC) to recover a posteriori distribution of the decentralized generator model parameters. In the first stage, we use the gradient-enhanced Langevin MCMC algorithm to characterize an intermediate posterior parameter distribution. This algorithm is computationally scalable to the high-dimensional parameter space. Based on the intermediate posterior distribution, during the second stage, we use the adaptive MCMC algorithm to fine-tune the strong correlations between the parameters. Finally, the fully recovered a posterior distribution is obtained in the end. The simulation results show that the proposed PCE-based hybrid MCMC algorithm can accurately and efficiently estimate the high-dimensional generator dynamic model parameters with full probabilistic distribution provided.

42 ENGINEERING↗