Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “statistical model”

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 1,387 records · Page 77
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Machine Learning for Improved Availability of the SNS Klystron High Voltage Converter Modulators

Beam availability has increased at the SNS, however, the targeted availability is greater than 95 %, while the SNS has failed to meet lower targets in the past. The HVCM used to power the linac klystrons have been one source of lost beam time and was chosen to explore using AI/ML techniques to improve reliability. Among the possibilities being explored are automating the tuning of HVCMs and predicting component failures such as capacitor aging, rectifier assemblies containing hundreds of diodes, and insulating oil degradation. The methodology pursued includes data cleaning, de-noising, post-analysis data labeling, and machine learning model development. We explore using Long Short-Term Memory and autoencoders for anomaly detection and prognostication used to schedule maintenance. We evaluate the use of model regularizers and constraints to improve the performance of the model and investigate methods to estimate the uncertainty of the models to provide a robust prediction with statistical interoperability. This paper describes the operational experience and known failures of the HVCMs and the proposed ML methodology and the preliminary results of training the AI/ML algorithms.

Pappas, Chris↗
DOE OSTI2026

Mortality among workers at the Rocky Flats Plant, 1951–2017

The Rocky Flats (RFs) Plant operated from 1951–1989 as part of the U.S. Department of Energy (DOE) nuclear complex. Its primary mission was weapons component fabrication, whereby workers were potentially exposed to radioactive and non-radioactive hazards. RF worker mortality was compared to the general population, and dose-response relationships between mortality and radiation organ doses were examined. RF workers first employed between 1951 and 1979 for ⩾30 d were identified (n = 9397). Vital status was determined using national and state death records up to 2017. Organ doses from external photons and neutrons irritation and internalised plutonium (Pu), americium (Am), and uranium (U) were modelled as cumulative lagged total doses per year. Beryllium exposure was evaluated as an effect modifier using data from the DOE Nationwide Beryllium Medical Program. Statistical analyses included standardised mortality ratios (SMRs), Cox proportional hazard models, and excess relative risk (ERR) models. Approximately 53.2% of workers were deceased by the end of the study. Nearly 90% were monitored for radiation exposure, with a mean weighted absorbed dose of 59.0 mGy for the lungs. Nearly 45% of workers had intakes of alpha-particle emitting radionuclides, and 46.7% were monitored for neutrons. Leading causes of death included ischemic heart disease (n = 999) and lung cancer (n = 361). The highest SMRs were observed for berylliosis (SMR: 176.9; 95% CI: 76.2, 348.7; n < 10) and asbestosis (SMR: 4.65; 95% CI: 2.23, 8.55; n = 10). Dose-response analyses showed no statistical increase in risk from low-dose radiation including lung cancer (ERR per 100 mGy: −0.02; 95% CI: −0.11, 0.08; n = 361) and Parkinson’s disease (ERR per 100 mGy: 0.13; 95% CI: −0.26, 0.31; n = 57). Approximately 45% of workers were monitored for beryllium, with a weak non-significant indication of effect modification for lung cancer risk. The RF cohort showed no evidence of a statistically significant increase in mortality from occupational radiation exposure. However, this study was limited by low statistical power, which inhibits the ability to detect effects. Future pooling of Million Person Study (MPS) cohorts will provide further insights, particularly regarding Pu as a carcinogen.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2020

FAST.Farm Development and Validation of Structural Load Prediction Against Large Eddy Simulations

FAST.Farm is a mid-fidelity engineering tool developed by the National Renewable Energy Laboratory targeted at accurately and efficiently predicting wind turbine power production and structural loading in wind farm settings, including wake interactions between turbines. FAST.Farm is based on several principles of the dynamic wake meandering (DWM) model, but also addresses limitations of previous DWM implementations. Previous FAST.Farm studies have shown the similarities and differences between FAST.Farm and large eddy simulations for rigid turbine cases. The objective of this work is to quantify the ability of FAST.Farm to accurately predict turbine structural response in a small wind farm. This is done by comparing FAST.Farm structural response to SOWFA-OpenFAST results for three laterally-aligned turbines. The purpose of this study is to characterize the similarities and differences between FAST.Farm and a higher fidelity model for predicting turbine structural response in a wind farm for differing atmospheric inflows. Strong statistical agreement was found between FAST.Farm and SOWFA-OpenFAST structural response for the non-waked upstream turbine, and good agreement was found for the downstream turbines for most structural quantities. Higher differences were seen for downstream turbines with low ambient turbulence intensity or yawed turbines, suggesting areas for FAST.Farm wake dynamics modeling improvements. For all cases and turbines, small statistical differences were seen between blade deflections and bending moments, with larger differences for tower-top and tower-base bending moments. Overall, the results establish confidence for applying FAST.Farm to wind farm power and loads analyses and identify areas where further model validations and model improvements should be targeted.

49 EE - Wind and Water Power Program - Wind (EE-4W↗
DOE OSTI2022

Toward Better Understanding of Microphysical Processes and Resulting Precipitation Physics: A Merger Of Observations and Cloud Models (Final Report)

A large database of global disdrometer observations and a diverse set of simulations from the Regional Atmospheric Modeling System (RAMS) were compared using principal component analysis (PCA) in order to better understand warm and ice-based precipitation processes. The analysis demonstrated that six distinct precipitation groups (PGs) with common characteristics were revealed in both the observations and model simulations. These PGs were defined by similar co-variability of rain parameters. However, the model showed large concentrations of drops with a given size compared to the observations. Detailed investigations determined the parameterization of drop breakup was forcing the drops to an equilibrium size with a stronger constraint than was indicated by the observations. A series of sensitivity studies permutating the rain shape parameter in the RAMS 2-moment bulk scheme demonstrated how the assumed shape of the rain distribution influences the microphysics as well as precipitation characteristics. Results were compared to the same simulations performed with a bin microphysics scheme where the rain shape parameter is freely evolving. In the bin scheme, a wide range of shape parameters were found, with horizontal and vertical variability, in addition to being a function of storm lifetime. The results of this study highlight the limitations of a fixed assumed rain distribution in 2-moment microphysics schemes. Model simulations were used to probe the microphysical origins of the six PGs. Rain budgets from a supercell case showed that several of the groups hypothesized to be associated with strong ice microphysical processes had significant contributions from melting hail, but the complexity of ice and warm-rain processes in the supercell precluded definitive determination of microphysical origins for many of the PGs. A case study from the Mid-latitude Continental Convection and Clouds Experiment (MC3E) compared the model PGs to disdrometer observations, and demonstrated the spatial and temporal variability of the PGs which is not possible from disdrometer point measurements alone. The analyses performed during this project demonstrated how statistical analysis can provide a robust method for comparing model simulations and observations, which ultimately resulted identification of limitations in some current microphysics parameterizations as well as insights into precipitation variability which ultimately benefits both observations and modeling efforts.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Systematics in asteroseismic modelling: application of a correlated noise model for oscillation frequencies

ABSTRACT The detailed modelling of stellar oscillations is a powerful approach to characterizing stars. However, poor treatment of systematics in theoretical models leads to misinterpretations of stars. Here, we propose a more principled statistical treatment for the systematics to be applied to fitting individual mode frequencies with a typical stellar model grid. We introduce a correlated noise model based on a Gaussian process (GP) kernel to describe the systematics given that mode frequency systematics are expected to be highly correlated. We show that tuning the GP kernel can reproduce general features of frequency variations for changing model input physics and fundamental parameters. Fits with the correlated noise model better recover stellar parameters than traditional methods that either ignore the systematics or treat them as uncorrelated noise.

Li, Tanda (ORCID:0000000163962563)↗
DOE OSTI2023

Towards a Comparative Assessment of Data-Driven Process Models in Health Information Technology

Process mining for conformance analysis focuses on comparing a reference process model against a data-driven process model that is generated via log files from information technology systems. While this approach is helpful when there is an existing process model in an organization, it leaves the question of what to do in the absence of a complete reference process model unanswered. In this paper, we present a comparative assessment approach that combines process mining, process mapping for dimensionality reduction, and statistical analysis. Our goal is to find similarities and dissimilarities in data-driven process models among U.S. Veterans Health Administration (VHA) facilities to assess process conformance among different healthcare facilities, which can help assess the standardization of care. We illustrate our approach by applying it to two clinical radiology order process models generated by two similar facilities. Our results demonstrate statistical similarities in the standardization of care among those two facilities.

Klasky, Hilda↗
DOE OSTI2022

Lifetime measurements of states of S 35 , S 36 , S 37 , and S 38 using the AGATA γ -ray tracking spectrometer

Lifetimes or lifetime limits of a small number of excited states of the sulfur isotopes with mass numbers A = 35 , 36, 37, and 38 have been measured using the differential recoil-distance method. The isotopes of sulfur were populated in binary grazing reactions initiated by a beam of 36 S ions of energy 225 MeV incident on a thin 208 Pb target which was mounted in the Cologne plunger apparatus. The combination of the PRISMA magnetic spectrometer and an early implementation of the AGATA γ-ray tracking array was used to detect γ rays in coincidence with projectile-like nuclear species. Lifetime measurements of populated states were measured within the range from about 1 to 100 ps. The number of states for which lifetime measurements or lifetime limits were possible was limited by statistics. For 35 S, the lifetime was determined for the first 1/2 + state at 1572 keV; the result is compared with a previous published lifetime value. The lifetime of the 3 - state of 36 S at 4193 keV was determined and compared with earlier measurements. No previous lifetime information exists for the (6 + ) state at 6690 keV; a lifetime measurement with large associated error was made in the present work. For 37 S, the states for which lifetime limits were established were those at 646 keV with J π = 3/2 - and at 2776 keV with J π = 11/2 - ; there are no previously published lifetime values for excited states of 37 S. Finally, a lifetime limit was established for the J π =(6 + ) state of 38 S at 3675 keV; no lifetime information exists for this state in the literature. Measured lifetime values were compared with the results of state-of-the-art shell-model calculations based on the PSDPF, SDPF-U, and FSU effective interactions. In addition, nuclear magnetic-dipole and electric-quadrupole moments, branching ratios, mixing ratios, and electromagnetic transition rates, where available, have been compared with shell-model values. The current work suffers from poor statistics; nevertheless, lifetime values and limits have been possible, allowing a useful discussion of the ability of state-of-the-art shell-model calculations to reproduce the experimental results.

electromagnetic transitions↗
DOE OSTI2024

The Aemulus Project. VI. Emulation of Beyond-standard Galaxy Clustering Statistics to Improve Cosmological Constraints

Abstract There is untapped cosmological information in galaxy redshift surveys in the nonlinear regime. In this work, we use the Aemulus suite of cosmological N -body simulations to construct Gaussian process emulators of galaxy clustering statistics at small scales (0.1–50 h −1 Mpc) in order to constrain cosmological and galaxy bias parameters. In addition to standard statistics—the projected correlation function w p ( r p ), the redshift-space monopole of the correlation function ξ 0 ( s ), and the quadrupole ξ 2 ( s )—we emulate statistics that include information about the local environment, namely the underdensity probability function P U ( s ) and the density-marked correlation function M ( s ). This extends the model of Aemulus III for redshift-space distortions by including new statistics sensitive to galaxy assembly bias. In recovery tests, we find that the beyond-standard statistics significantly increase the constraining power on cosmological parameters of interest: including P U ( s ) and M ( s ) improves the precision of our constraints on Ω m by 27%, σ 8 by 19%, and the growth of structure parameter, f σ 8 , by 12% compared to standard statistics. We additionally find that scales below ∼6 h −1 Mpc contain as much information as larger scales. The density-sensitive statistics also contribute to constraining halo occupation distribution parameters and a flexible environment-dependent assembly bias model, which is important for extracting the small-scale cosmological information as well as understanding the galaxy–halo connection. This analysis demonstrates the potential of emulating beyond-standard clustering statistics at small scales to constrain the growth of structure as a test of cosmic acceleration.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Model comparisons for two-temperature plasma equations of state

When a plasma is generated in the laboratory, energy is often deposited preferentially into either the electrons or the ions, giving rise to a quasiequilibrium state in which the two species, electrons and ions, are well described by two effective temperatures, T e and T i . Accurate hydrodynamic modeling of such a two-temperature plasma requires an equation of state that captures the relevant many-body physics without assuming a strict local thermodynamic equilibrium. Several models have been proposed within the literature, which extend conventional statistical approaches, each employing a different combination of assumptions for modifying the equilibrium equations. Here, in this work, we compare the predictions for several models, presenting derivations of the internal energy and pressure for each microscopic model within a unified framework so that the assumptions of each model may be more easily compared to one another. We find that for sufficiently weak coupling, all models agree with one another. However, as the coupling strength is increased, the disagreement between the models becomes more pronounced. Moreover, the relative sizes of the corrections predicted by each model depend on which species has the higher temperature, T e >T i vs T e < T i .

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Laser Powder Bed Fusion Microstructure Surrogate Model

SAND2025-11467O The Laser Powder Bed Fusion (LPBF) Microstructure Surrogate Model is a machine-learning-based tool. It predicts statistics of microstructures that are produced by the LPBF additive manufacturing process. It includes a series of codes for training, testing, and analyzing the model as well as utility scripts for handling data. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Moser, Daniel [Sandia National Lab. (SNL-CA), Live↗
DOE OSTI2023

Discovering Active Subspaces for High-Dimensional Computer Models

Dimension reduction techniques have long been an important topic in statistics, and active subspaces (AS) have received much attention this past decade in the computer experiments literature. The most common approach towards estimating the AS is to use Monte Carlo with numerical gradient evaluation. While sensible in some settings, this approach has obvious drawbacks. Recent research has demonstrated that active subspace calculations can be obtained in closed form, conditional on a Gaussian process (GP) surrogate, which can be limiting in high-dimensional settings for computational reasons. In this paper, we produce the relevant calculations for a more general case when the model of interest is a linear combination of tensor products. These general equations can be applied to the GP, recovering previous results as a special case, or applied to the models constructed by other regression techniques including multivariate adaptive regression splines (MARS). Furthermore, using a MARS surrogate has many advantages including improved scaling, better estimation of active subspaces in high dimensions and the ability to handle a large number of prior distributions in closed form. In one real-world example, we obtain the active subspace of a radiation-transport code with 240 inputs and 9,372 model runs in under half an hour.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Monitoring and prediction of porosity in laser powder bed fusion using physics-informed meltpool signatures and machine learning

In this work we accomplished the monitoring and prediction of porosity in laser powder bed fusion (LPBF) additive manufacturing process. This objective was realized by extracting physics-informed meltpool signatures from an in-situ dual-wavelength imaging pyrometer, and subsequently, analyzing these signatures via computationally tractable machine learning approaches. Porosity in LPBF occurs despite extensive optimization of processing conditions due to stochastic causes. Hence, it is essential to continually monitor the process with in-situ sensors for detecting and mitigating incipient pore formation. In this work a tall cuboid-shaped part (10 mm × 10 mm × 137 mm, material ATI 718Plus) was built with controlled porosity by varying laser power and scanning speed. This test caused various types of porosity, such as lack-of-fusion and keyhole formation, with varying degrees of severity in the part. The meltpool was continuously monitored using a dual-wavelength imaging pyrometer installed in the machine. Physically intuitive process signatures, such as meltpool length, temperature distribution, and ejecta (spatter) characteristics, were extracted from the meltpool images. Subsequently, relatively simple machine learning models, e.g., K-Nearest Neighbors, were trained to predict both the severity and type of porosity as a function of these physics-informed meltpool signatures. These models resulted in a prediction accuracy exceeding 95% (statistical F1-score). The same analysis was carried out with a complex, black-box deep learning convolutional neural network which directly used the meltpool images instead of physics-informed features. The convolutional neural network produced a comparable F1-score in the range of 89–97%. Finally, these results demonstrate that using pragmatic, physics-informed meltpool signatures within a simple machine learning model is as effective for flaw prediction in LPBF as using a complex and computationally demanding black-box deep learning model.

36 MATERIALS SCIENCE↗
DOE OSTI2023

A framework to measure the properties of intergalactic metal systems with two-point flux statistics

ABSTRACT The abundance, temperature, and clustering of metals in the intergalactic medium are important parameters for understanding their cosmic evolution and quantifying their impact on cosmological analysis with the Ly α forest. The properties of these systems are typically measured from individual quasar spectra redward of the quasar’s Ly α emission line, yet that approach may provide biased results due to selection effects. We present an alternative approach to measure these properties in an unbiased manner with the two-point statistics commonly employed to quantify large-scale structure. Our model treats the observed flux of a large sample of quasar spectra as a continuous field and describes the one-dimensional, two-point statistics of this field with three parameters per ion: the abundance (column density distribution), temperature (Doppler parameter), and clustering (cloud–cloud correlation function). We demonstrate this approach on multiple ions (e.g. ${\rm C\, \small {\rm IV}}$ , ${\rm Si\, \small {\rm IV}}$ , and ${\rm Mg\, \small {\rm II}}$ ) with early data from the Dark Energy Spectroscopic Instrument (DESI) and high-resolution spectra from the literature. Our initial results show some evidence that the ${\rm C\, \small {\rm IV}}$ abundance is higher than previous measurements and evidence for abundance evolution over time. The first full year of DESI observations will have over an order of magnitude more quasar spectra than this study. In a future paper, we will use those data to measure the growth of clustering and its impact on the Ly α forest, as well as test other DESI analysis infrastructure such as the pipeline noise estimates and the resolution matrix.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Quantifying modeling uncertainties when combining multiple gravitational-wave detections from binary neutron star sources

Here, with the increasing sensitivity of gravitational-wave detectors, we expect to observe multiple binary neutron-star systems through gravitational waves in the near future. The combined analysis of these gravitational-wave signals offers the possibility to constrain the neutron-star radius and the equation of state of dense nuclear matter with unprecedented accuracy. However, it is crucial to ensure that uncertainties inherent in the gravitational-wave models will not lead to systematic biases when information from multiple detections is combined. To quantify waveform systematics, we perform an extensive simulation campaign of binary neutron-star sources and analyze them with a set of four different waveform models. For our analysis with 38 simulations, we find that statistical uncertainties in the neutron-star radius decrease to ± 250 m (2% at 90% credible interval) but that systematic differences between currently employed waveform models can be twice as large. Hence, it will be essential to ensure that systematic biases will not become dominant in inferences of the neutron-star equation of state when capitalizing on future developments.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2020

Differentially Private Generation of Social Networks via Exponential Random Graph Models

Many social networks contain sensitive relational information. One approach to protect the sensitive relational information while offering flexibility for social network research and analysis is to release synthetic social networks at a pre-specified privacy risk level, given the original observed network. In this work, we propose the DP-ERGM procedure that synthesizes networks that satisfy the differential privacy (DP) via the exponential random graph model (EGRM). We apply DP-ERGM to a college student friendship network and compare its original network information preservation in the generated private networks with two other approaches: differentially private DyadWise Randomized Response (DWRR) and Sanitization of the Conditional probability of Edge given Attribute classes (SCEA). The results suggest that DP-EGRM preserves the original information significantly better than DWRR and SCEA in both network statistics and inferences from ERGMs and latent space models. In addition, DP-ERGM satisfies the node DP, a stronger notion of privacy than the edge DP that DWRR and SCEA satisfy.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Chemometrics and Experimental Design for the Quantification of Nitrate Salts in Nitric Acid: Near-Infrared Spectroscopy Absorption Analysis

Implementing remote, real-time spectroscopic monitoring of radiochemical processing streams in hot cell environments requires efficiency and simplicity. The success of optical spectroscopy for the quantification of species in chemical systems highly depends on representative training sets and suitable validation sets. Selecting a training set (i.e., calibration standards) to build multivariate regression models is both time- and resource-consuming using standard one-factor-at-a-time approaches. This study describes the use of experimental design to generate spectral training sets and a validation set for the quantification of sodium nitrate (0–1 M) and nitric acid (0.1–10 M) using the near-infrared water band centered at 1440 nm. Partial least squares regression models were built from training sets generated by both D- and I-optimal experimental designs and a one-factor-at-a-time approach. The prediction performance of each model was evaluated by comparing the bias and standard error of prediction for statistical significance. D- and I-optimal designs reduced the number of samples required to build regression models compared with one-factor-at-a-time while also improving performance. Models must be confirmed against a validation sample set when minimizing the number of samples in the training set. The D-optimal design performed the best when considering both performance and efficiency by improving predictive capability and reducing number of samples in the training set by 64% compared with the one-factor-at-a-time approach. The experimental design approach objectively selects calibration and validation spectral data sets based on statistical criterion to optimize performance and minimize resources.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗
DOE OSTI2026

Statistical Uncertainty of Inhalation Dose Coefficients in Consequence Management: Propagated Dose Uncertainty in ICRP 66 Human Respiratory Tract Model

Reference inhalation dose models rely on deterministic biokinetics and reference computational phantoms, limiting their applicability to the variability present in population-specific exposures encountered in emergency response scenarios. Here, this study introduces REDCAL, a Python-based computational framework developed to propagate uncertainty in inhalation dose coefficients using the International Commission on Radiological Protection (ICRP) Publication 66 Human Respiratory Tract Model. REDCAL integrates ICRP deposition and clearance models, systemic biokinetics, and governing physics principles, and leverages Sandia National Laboratories’ Dakota toolkit for uncertainty quantification via Latin Hypercube Sampling. REDCAL was validated against DCAL, with biokinetic retention results differing by less than 1% and effective dose coefficients by less than 2% across all tested radionuclides. Stochastic sampling introduced variability in dose coefficients, with geometric standard deviations (GSD) in committed effective dose coefficients (CEDC) ranging from 1.0 to 1.5, based on lognormal distribution fits. Analysis demonstrated that variations in the activity median aerodynamic diameter (AMAD) notably influenced the computed CEDC values. Smaller particles (<1 µm) increased doses by 20–30% due to deeper lung deposition and prolonged retention for alpha emitting radionuclides, such as 241 Am and 239 Pu. Radionuclides with fast clearance, such as 133 I, demonstrated a dose reduction exceeding 50%, as AMAD increased beyond 5 µm due to upper airway deposition and rapid mucociliary clearance. The greatest GSD among the radionuclides reported in this study was for 241 Am. In most cases, the largest GSDs in the CEDC were associated with larger particle sizes, an expected outcome, as ICRP Publication 66 defines GSD in particle size as a function of AMAD, resulting in an extended tail of the lognormal distribution. The findings support improved inhalation dose assessments and enhance consequence management strategies for the U.S. Federal Radiological Monitoring and Assessment Center by quantifying uncertainty in dose coefficients and strengthening decision-making for emergency response scenarios.

Biokinetic Modeling↗
DOE OSTI2020

Validation of the High-Resolution Salish Sea Tidal Hydrodynamic Model

In this study, a tidal hydrodynamic model was developed and validated to simulate tidal currents in Puget Sound, Washington, to support tidal energy resource characterization using the unstructured-grid, Finite Volume Community Ocean Model (FVCOM). The Salish Sea tidal hydrodynamic model was driven by tides along two open boundaries at the entrance of the Strait of Juan de Fuca and north end of Georgia Strait, and river flows from 19 major rivers in the Salish Sea. To simulate the tidal current in Puget Sound, a high-resolution model grid is required to accurately represent the complex coastlines and bathymetry. The spatial resolution of the model grid varies from ~10 m near river boundaries and ~30 m in small tidal channels and estuaries to near 1000 m inside Georgia Strait and at the open boundaries. Model validation was carried out by comparing simulated and observed water levels at 12 tidal stations and currents at 135 Acoustic Doppler Current Profiler stations in the model domain. A set of model performance metrics, including root mean square error, scatter index, bias, and linear correlation coefficient, were used to quantify the model skills in simulating the tidal hydrodynamics in Puget Sound. Error statistics showed an overall good agreement between simulated and observed tidal elevations and currents, which demonstrated that the Puget Sound tidal model can be used to accurately characterize the tidal stream energy resource in Puget Sound.

16 TIDAL AND WAVE POWER↗