Predicting city-scale daily electricity consumption using data-driven models
Not Available
SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Not Available
Anomaly detection (AD) in sensor data is critical to ensure uninterrupted functionality of nuclear power plants (NPPs). Consequently, AD model validation through real-world sensor data is important for applications in nuclear facilities. In this paper, we propose an Autoencoder (AE)—a multi-layered neural network, for AD in sensor data from an operational NPP testbed. Since the dataset lacks labels for irregularities, we introduce random noise and label them to effectively train our model. The proposed AE model assigns a higher reconstruction error to the abnormal samples that deviate from those encountered during the training phase and uses the reconstruction loss to detect anomalies in a representative imbalanced dataset. We also introduce an analytical solution—seasonal trend decomposition (STD)—as another AD scheme for identifying irregularities withinthe same time-series dataset. In contrast to the AE model which relies on reconstruction loss, the STD scheme decomposes the entire dataset into its trend, seasonality, and residual components to pinpoint irregularities. Our findings indicate that the proposed AE and STD models individually achieve recall scores of 97% and 92%, respectively. We validate the performance of the two models on both balanced and imbalanced data. We further solidify the results by picking the combined selected anomalies of the two solutions with an "AND" operator for more reliable predictions.
Anomaly detection (AD) in sensor data is critical to ensure uninterrupted functionality of nuclear power plants (NPPs). Consequently, validation of AD models through real-world sensor data is important for their application in nuclear facilities. In this paper, we propose an Autoencoder (AE)— a multi-layered neural network, for AD in sensor data from an operational NPP testbed. Since the dataset lacks labels for irregularities, we introduce random noise and label them to effectively train our model. The proposed AE model assigns a higher reconstruction error to the abnormal samples that deviate from those encountered during the training phase and uses the reconstruction loss to detect anomalies in a representative imbalanced dataset. We also introduce an analytical solution—seasonal trend decomposition (STD) — as another AD scheme for identifying irregularities within the same time-series dataset. In contrast to the AE model which relies on reconstruction loss, the STD scheme decomposes the entire dataset into its trend, seasonality, and residual components to pinpoint irregularities. Our findings indicate that the proposed AE and STD models individually achieve recall scores of 97% and 92%, respectively. We also validate the performance of the two models on both balanced and imbalanced data. We further solidify the results by picking the combined selected anomalies of the two solutions with an "AND" operator for more reliable predictions.
Analysis of water-level trends in well hydro-graphs can be used to better understand and conceptualize a groundwater-flow system. A water-level trend reflects the summation of all natural and anthropogenic stresses acting on the aquifer at the well location. Common stresses include recharge, evapotranspiration, barometric pressure, earth tides, and pumping. A water-level model (WLM) can be used as a tool to distinguish which stresses are affecting the trend and quantify the impact of each stress on the trend. A WLM is an analytical model that fits measured water levels to a synthetic curve representing the sum of one or more time-series components that explain the water-level trend. Time-series components include recharge pulses simulated with Gamma transforms of precipitation; long-term climatic trends and barometric-pressure responses simulated with moving averages of barometric pressure and water levels in background wells; tidal responses computed from theoretical tidal signals; and pumping drawdown simulated with Theis models. WLM examples are presented to show how to: (1) determine and decompose stresses affecting long-term water levels in wells; and (2) estimate short-term draw-downs in observation wells where drawdown is masked by environmental water-level fluctuations. (authors)
Many-to-one maps are ubiquitous in machine learning, from the image recognition model that assigns a multitude of distinct images to the concept of ``cat'' to the time series forecasting model which assigns a range of distinct time-series to a single scalar regression value. While the primary use of such models is naturally to associate correct output to each input, in many problems it is also useful to be able to explore, understand, and sample from a model's fibers, which are the set of input values x such that f(x) = y, for fixed y in the output space. In this paper we show that popular generative architectures are ill-suited to such tasks. Motivated by this we introduce a novel generative architecture, a Bundle Network, based on the concept of a fiber bundle from (differential) topology. BundleNets exploit the idea of a local trivialization wherein a space can be locally decomposed into a product space that cleanly encodes the many-to-one nature of the map. By enforcing this decomposition in BundleNets and by utilizing state-of-the-art invertible components, investigating a network's fibers becomes natural.
The RAVEN code has been under development at the Idaho National Laboratory since 2012. Its main goal is to create a multi-purpose platform for the deploying of all the capabilities needed for Probabilistic Risk Assessment, uncertainty quantification, data mining analysis and optimization studies. RAVEN has currently reached a good level of maturity in terms of deployed state-of-art and advanced capabilities. The main subject of this report is to show the activities that have been recently accomplished: • Implementation of ensemble modeling for time-series, and • initial implementation of model validation for surrogate models, and • advanced visualization capability for topology based data analysis The development of ensemble modeling for time-series has been performed in order to begin tackling the needs of those RISMC applications that need to communicate 1-Dimensional information (e.g. power histories, etc.) among different models. In this document the implementation details and an application example is reported. The second subject of this report is about the initial development of methods, within the RAVEN framework, to assess the validity of the predictive capabilities of surrogate models. Indeed, after the construction of a surrogate tight to a certain physical model, it is crucial to assess the goodness of its representation, in order to be confident with its prediction. In this initial activity, a cross-validation technique has been employed. This report will highlight the implementation details and proof its correct implementation by an application example. The final subject of this report is about the implementation of advanced visualization capability in RAVEN, for interactive data analysis. Indeed, RAVEN offers several post-processing capabilities that can structurally decompose data extracted from experimental results offering both data clustering/partitioning and dimensionality reduction techniques. A disadvantage of the workflow available in RAVEN is that it treats these as black box operations and the user is expected to know specific information about their data including the number of partitions to expect in some cases or the “correct” parameter settings for a particular algorithm. In order to overcome this limitation, it has been added an interactive user interface that can be run in RAVEN to explore different parameter settings on the fly specifically for the topological post-processor and allows the user to explore the data interactively. The design of this user interface is generalizable to other algorithms and can be used as a model to generate more dynamic and user-friendly visualization capabilities within RAVEN.
Short-term solar forecasting models based solely on global horizontal irradiance (GHI) measurements are often unable to discriminate the forecasting of the factors affecting GHI from those that can be precisely computed by atmospheric models. Our previous study introduced a Physics-based Smart Persistence model for Intra-hour forecasting of solar radiation (PSPI) that decomposed the forecasting of GHI into the computation of extraterrestrial solar radiation and solar zenith angle and the forecasting of cloud albedo and cloud fraction. The extraterrestrial solar radiation and solar zenith angle were accurately computed by the Solar Position Algorithm (SPA) developed at the National Renewable Energy Laboratory (NREL). A cloud retrieval technique was used to estimate cloud albedo and cloud fraction from surface-based observations of GHI. With the assumption of persistent cloud structures, the cloud albedo and cloud fraction were predicted for future time steps using a two-stream approximation and a 5-minute exponential weighted moving average, respectively. The model evaluation indicated the estimation and forecast of cloud fraction mostly contributed to the uncertainty of the PSPI though it overcame the persistence and smart persistence models in all forecast time horizons between 5 and 60 minutes. This study aims to enhance the PSPI by ingesting surface-based observations of cloud fraction from a total sky imager (TSI). The estimation and forecast of cloud albedo is correspondingly improved by utilizing the cloud fraction observations and thus leads to more accurate GHI forecast. Various time-series analysis methods are also investigated on the forecasting of cloud fraction and cloud albedo for further improving the GHI forecast. These improvements are valuable for many applications, such as forecasting energy use for buildings, grid operations, and ultimately bringing down the cost of solar energy.