Power Capacity Specification for Energy Storage in Wind Application Using Probability-Based Method
Explore the source record for details and available documents.
17 WIND ENERGY↗
SEARCH · Search NASA
Results for “non-parametric probability density estimation”
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.
Probability-Based Method for Power Capacity Specification of Wind Energy Storage Systems
Explore the source record for details and available documents.
17 WIND ENERGY↗
Non-Parametric Statistical Analysis of Current Waveforms through Power System Sensors
The protection, control, and monitoring of the power grid is not possible without accurate measurement devices. As the percentage of renewable energy sources penetrating the existing grid infrastructure increases, so do uncertainties surrounding their effects on the everyday operation of the power system. Many of these devices are sources of high-frequency transients. These transients may be useful for identifying certain events or behaviors otherwise not seen in traditional analysis techniques. Therefore, the ability of sensors to accurately capture these phenomena is paramount. In this work, two commercial-grade power system distribution sensors are investigated in terms of their ability to replicate high-frequency phenomena by studying their responses to three events: a current inrush, a microgrid “close-in”, and a fault on the terminals of a wind turbine. Kernel density estimation is used to derive the non-parametric probability density functions of these error distributions and their adequateness is quantified utilizing the commonly used root mean square error (RMSE) metric. It is demonstrated that both sensors exhibit characteristics in the high harmonic range that go against the assumption that measurement error is normally distributed.
47 OTHER INSTRUMENTATION↗
Detecting Anomalies in Time Series Using Kernel Density Approaches
This paper introduces a novel anomaly detection approach tailored for time series data with exclusive reliance on normal events during training. Our key innovation lies in the application of kernel-density estimation (KDE) to scrutinize reconstruction errors, providing an empirically derived probability distribution for normal events post-reconstruction. This non-parametric density estimation technique offers a nuanced understanding of anomaly detection, differentiating it from prevalent threshold-based mechanisms in existing methodologies. In post-training, events are encoded, decoded, and evaluated against the estimated density, providing a comprehensive notion of normality. In addition, we propose a data augmentation strategy involving variational autoencoder-generated events and a smoothing step for enhanced model robustness. The significance of our autoencoder-based approach is evident in its capacity to learn normal representation without prior anomaly knowledge. Through the KDE step on reconstruction errors, our method addresses the versatility of anomalies, departing from assumptions tied to larger reconstruction errors for anomalous events. Our proposed likelihood measure then distinguishes normal from anomalous events, providing a concise yet comprehensive anomaly detection solution. The extensive experimental results support the feasibility of our proposed method, yielding significantly improved classification performance by nearly 10% on the UCR benchmark data.
Frehner, Robin↗
Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators: Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators
Forecasts of available wind power are critical in key electric power systems operations planning problems, including economic dispatch and unit commitment. Such forecasts are necessarily uncertain, limiting the reliability and cost effectiveness of operations planning models based on a single deterministic or “point” forecast. A common approach to address this limitation involves the use of a number of probabilistic scenarios, each specifying a possible trajectory of wind power production, with associated probability. We present and analyze a novel method for generating probabilistic wind power scenarios, leveraging available historical information in the form of forecasted and corresponding observed wind power time series. We estimate non-parametric forecast error densities, specifically using epi-spline basis functions, allowing us to capture the skewed and non-parametric nature of error densities observed in real-world data. We then describe a method to generate probabilistic scenarios from these basis functions that allows users to control for the degree to which extreme errors are captured.We compare the performance of our approach to the current state-of-the-art considering publicly available data associated with the Bonneville Power Administration, analyzing aggregate production of a number of wind farms over a large geographic region. Finally, we discuss the advantages of our approach in the context of specific power systems operations planning problems: stochastic unit commitment and economic dispatch. Here, our methodology is embodied in the joint Sandia – University of California Davis Prescient software package for assessing and analyzing stochastic operations strategies.
17 WIND ENERGY↗
Quantiles, parametric-select density estimation, and bi-information parameter estimators
A quantile-based approach to statistical analysis and probability modeling of data is presented which formulates statistical inference problems as functional inference problems in which the parameters to be estimated are density functions. Density estimators can be non-parametric (computed independently of model identified) or parametric-select (approximated by finite parametric models that can provide standard models whose fit can be tested). Exponential models and autoregressive models are approximating densities which can be justified as maximum entropy for respectively the entropy of a probability density and the entropy of a quantile density. Applications of these ideas are outlined to the problems of modeling: (1) univariate data; (2) bivariate data and tests for independence; and (3) two samples and likelihood ratios. It is proposed that bi-information estimation of a density function can be developed by analogy to the problem of identification of regression models.
Parzen, E.↗
PHOTOMETRIC REDSHIFTS AND QUASAR PROBABILITIES FROM A SINGLE, DATA-DRIVEN GENERATIVE MODEL
We describe a technique for simultaneously classifying and estimating the redshift of quasars. It can separate quasars from stars in arbitrary redshift ranges, estimate full posterior distribution functions for the redshift, and naturally incorporate flux uncertainties, missing data, and multi-wavelength photometry. We build models of quasars in flux-redshift space by applying the extreme deconvolution technique to estimate the underlying density. By integrating this density over redshift, one can obtain quasar flux densities in different redshift ranges. This approach allows for efficient, consistent, and fast classification and photometric redshift estimation. This is achieved by combining the speed obtained by choosing simple analytical forms as the basis of our density model with the flexibility of non-parametric models through the use of many simple components with many parameters. We show that this technique is competitive with the best photometric quasar classification techniques-which are limited to fixed, broad redshift ranges and high signal-to-noise ratio data-and with the best photometric redshift techniques when applied to broadband optical data. We demonstrate that the inclusion of UV and NIR data significantly improves photometric quasar-star separation and essentially resolves all of the redshift degeneracies for quasars inherent to the ugriz filter system, even when included data have a low signal-to-noise ratio. For quasars spectroscopically confirmed by the SDSS 84% and 97% of the objects with Galaxy Evolution Explorer UV and UKIDSS NIR data have photometric redshifts within 0.1 and 0.3, respectively, of the spectroscopic redshift; this amounts to about a factor of three improvement over ugriz-only photometric redshifts. Our code to calculate quasar probabilities and redshift probability distributions is publicly available.
79 ASTRONOMY AND ASTROPHYSICS↗
A Deterministic Annealing Approach to Clustering AIRS Data
We will examine the validity of means and standard deviations as a basis for climate data products. We will explore the conditions under which these two simple statistics are inadequate summaries of the underlying empirical probability distributions by contrasting them with a nonparametric, method called Deterministic Annealing technique
clustering↗
Likelihood non-Gaussianity in large-scale structure analyses
The standard present-day large-scale structure (LSS) analyses make a major assumption in their Bayesian parameter inference – that the likelihood has a Gaussian form. For summary statistics currently used in LSS, this assumption, even if the underlying density field is Gaussian, cannot be correct in detail. We investigate the impact of this assumption on two recent LSS analyses: the Beutler et al. power spectrum multipole ($P_ℓ$) analysis and the Sinha et al. group multiplicity function ($ζ$) analysis. Using non-parametric divergence estimators on mock catalogues originally constructed for covariance matrix estimation, we identify significant non-Gaussianity in both the $P_ℓ$ and $ζ$ likelihoods. We then use Gaussian mixture density estimation and independent component analysis on the same mocks to construct likelihood estimates that approximate the true likelihood better than the Gaussian pseudo-likelihood. Using these likelihood estimates, we accurately estimate the true posterior probability distribution of the Beutler et al. and Sinha et al. parameters. Likelihood non-Gaussianity shifts the fσ8 constraint by -0.44σ, but otherwise does not significantly impact the overall parameter constraints of Beutler et al. For the $ζ$ analysis, using the pseudo-likelihood significantly underestimates the uncertainties and biases the constraints of the Sinha et al. halo occupation parameters. For log $M_1$ and α, the posteriors are shifted by +0.43σ and -0.51σ and broadened by 42 per cent and 66 per cent, respectively. The divergence and likelihood estimation methods we present provide a straightforward framework for quantifying the impact of likelihood non-Gaussianity and deriving more accurate parameter constraints.
79 ASTRONOMY AND ASTROPHYSICS↗
The Poisson tensor completion non-parametric differential entropy estimator
We introduce the Poisson tensor completion (PTC) estimator, a non-parametric differential entropy estimator. The PTC estimator leverages inter-sample relationships to compute a low-rank Poisson tensor decomposition of the frequency histogram. Our crucial observation is that the histogram bins are an instance of a space partitioning of counts and thus can be identified with a spatial Poisson process. The Poisson tensor decomposition leads to a completion of the intensity measure over all bins—including those containing few to no samples—and leads to our proposed PTC differential entropy estimator. A Poisson tensor decomposition models the underlying distribution of the count data and guarantees non-negative estimated values and so can be safely used directly in entropy estimation. Our estimator is the first tensor-based estimator that exploits the underlying spatial Poisson process related to the histogram explicitly when estimating the probability density with low-rank tensor decompositions for the purpose of tensor completion. Furthermore, we demonstrate that our PTC estimator is a substantial improvement over standard histogram-based estimators for sub-Gaussian probability distributions because of the concentration of norm phenomenon.
42 ENGINEERING↗
Modeling and Visualizing Uncertainty in Continuous Variables Predicted using Remotely Sensed Data
The use of remotely sensed images to map continuous biophysical variables, such as those related to terrestrial vegetation amount, sea surface temperature, and many other targets of NASA s Earth Observing System (EOS), includes variable, parametric, positional, spatial support and structural sources of uncertainty. A complete description of uncertainty will lead to a probability distribution at each location, allowing the exploration of the spatial dimension of uncertainty, that is, where the field is not well quantified. To achieve this purpose, convenient visualization tools are required. We have produced such a tool, called PDFVis, that facilitates the display of probability density functions (pdfs) on a per-grid-cell basis. The density estimate from Monte-Carlo generated realizations is interactively displayed as well as parametric and non-parametric summaries of the pdf field (such as mean, median, quartiles, standard deviation, number of modes, and locations of modes). Shaded surface renderings of pdfs along a transect can also be projected onto a plane. This tool will become more useful as richer descriptions of spatial uncertainty become available.
Dungan, Jennifer L.↗