Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “autoregressive models”

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 109 records · Page 6

Data-Driven Robust M-LS-SVR-Based NARX Modeling for Estimation and Control of Molten Iron Quality Indices in Blast Furnace Ironmaking

Optimal operation of an industrial blast furnace (BF) ironmaking process largely depends on a reliable measurement of molten iron quality (MIQ), which is not feasible using the conventional sensors. This paper proposes a novel data-driven robust modeling method for online estimation and control of MIQ indices. First, to completely capture the nonlinear dynamics of the BF process, the nonlinear autoregressive exogenous (NARX) model is constructed for the MIQ indices. Then, considering that the standard least-squares support vector regression (LS-SVR) cannot directly cope with the multi-output problem, the multi- task transfer learning is proposed to design a novel multi-output LS-SVR (M-LS-SVR) for learning of the NARX model. Furthermore, a novel M-estimator is proposed to reduce the interference of outliers and improve the robustness of the M-LS- SVR model. Since the weights of different outlier data are properly given by the weight function, their corresponding contributions on modeling can be properly distinguished, thus a robust modeling result can be achieved. Finally, a novel multi- objective evaluation index on modeling performance is developed by comprehensively considering the root mean square error of modeling and the correlation coefficient on trend fitting, based on which the NSGA-II algorithm is used to globally optimize the model parameters. Both experiments using industrial data and industrial applications illustrate that the proposed method can eliminate the adverse effect caused by the fluctuation of data in BF process efficiently, which indicates its stronger robustness and higher accuracy. Moreover, control testing shows that the developed model can be well applied to realize data-driven control of the BF process.

blast cavities, Control↗

Watch and learn—a generalized approach for transferrable learning in deep neural networks via physical principles

Transfer learning refers to the use of knowledge gained while solving a machine learning task and applying it to the solution of a closely related problem. Such an approach has enabled scientific breakthroughs in computer vision and natural language processing where the weights learned in state-of-the-art models can be used to initialize models for other tasks which dramatically improve their performance and save computational time. Here we demonstrate an unsupervised learning approach augmented with basic physical principles that achieves fully transferrable learning for problems in statistical physics across different physical regimes. By coupling a sequence model based on a recurrent neural network to an extensive deep neural network, we are able to learn the equilibrium probability distributions and inter-particle interaction models of classical statistical mechanical systems. Our approach, distribution-consistent learning, DCL, is a general strategy that works for a variety of canonical statistical mechanical models (Ising and Potts) as well as disordered interaction potentials. Using data collected from a single set of observation conditions, DCL successfully extrapolates across all temperatures, thermodynamic phases, and can be applied to different length-scales. This constitutes a fully transferrable physics-based learning in a generalizable approach.

97 MATHEMATICS AND COMPUTING↗

Multivariate Time Series Anomaly Detection with Few Positive Samples

Given the scarcity of anomalies in real-world applications, the majority of literature has been focusing on modeling normality. The learned representations enable anomaly detection as the normality model is trained to capture certain key underlying data regularities under normal circumstances. In practical settings, particularly industrial time series anomaly detection, we often encounter situations where a large amount of normal operation data is available along with a small number of anomaly events collected over time. This practical situation calls for methodologies to leverage these small number of anomaly events to create a better anomaly detector. In this paper, we introduce two methodologies to address the needs of this practical situation and compared them with recently developed state of the art techniques. Our proposed methods anchor on representative learning of normal operation with autoregressive (AR) model along with loss components to encourage representations that separate normal versus few positive examples. We applied the proposed methods to two industrial anomaly detection datasets and demonstrated effective performance in comparison with approaches from literature. Our study also points out additional challenges with adopting such methods in practical applications.

Xue, Feng↗

A Robust State Estimation Framework Considering Measurement Correlations and Imperfect Synchronization

This paper develops a robust power system state estimation framework with the consideration of measurement correlations and imperfect synchronization. In the framework, correlations of SCADA and PMU measurements are calculated separately through unscented transformation and a vector autoregression (VAR) model. In particular, PMU measurements during the waiting period of two SCADA measurement scans are buffered to develop the VAR model. The latter takes into account the temporal and spatial correlations of PMU measurements and provides redundant measurements to suppress bad data and mitigate imperfect synchronization. In case of synchronization issues between SCADA and PMU measurements, either forecasted PMU measurements or prior SCADA measurements from the last estimation run are leveraged to restore system observability. After that, robust generalized maximum-likelihood (GM)-estimator is developed to integrate full measurement correlations and handle bad data in SCADA and PMU measurements. Thanks to the robustness and flexibility of the proposed robust estimator, bad data as well as imperfect measurement synchronization are mitigated, yielding high statistical efficiency. Comprehensive comparison results with other alternatives under various conditions demonstrate the benefits of the proposed framework.

Zhao, Junbo (ORCID:0000000284989666)↗

A staged deep learning approach to spatial refinement in 3D temporal atmospheric transport

High-resolution spatiotemporal simulations effectively capture the complexities of atmospheric plume dispersion in complex terrain. However, their high computational cost makes them impractical for applications requiring rapid responses or iterative processes, such as optimization, uncertainty quantification, or inverse modeling. To address this challenge, this work introduces the Dual-Stage Temporal Three-dimensional UNet Super-resolution (DST3D-UNet-SR) model, a highly efficient deep learning model for plume dispersion predictions. DST3D-UNet-SR is composed of two sequential modules: the temporal module (TM), which predicts the transient evolution of a plume in complex terrain from low-resolution temporal data, and the spatial refinement module (SRM), which subsequently enhances the spatial resolution of the TM predictions. We train DST3D-UNet-SR using a comprehensive dataset derived from high-resolution large eddy simulations (LES) of plume transport. We propose the DST3D-UNet-SR model to significantly accelerate LES of three-dimensional (3D) plume dispersion by three orders of magnitude. Additionally, the model demonstrates the ability to dynamically adapt to evolving conditions through the incorporation of new observational data, substantially improving prediction accuracy in high-concentration regions near the source.

3D temporal sequences↗

Optimal carbon storage reservoir management through deep reinforcement learning

Model-based optimization plays a central role in energy system design and management. The complexity and high-dimensionality of many process-level models, especially those used for geosystem energy exploration and utilization, often lead to formidable computational costs when the dimension of decision space is also large. This work adopts elements of recently advanced deep learning techniques to solve a sequential decision-making problem in applied geosystem management. Specifically, a deep reinforcement learning framework was formed for optimal multiperiod planning, in which a deep Q-learning network (DQN) agent was trained to maximize rewards by learning from high-dimensional inputs and from exploitation of its past experiences. To expedite computation, deep multitask learning was used to approximate high-dimensional, multistate transition functions. Both DQN and deep multitask learning are pattern based. As a demonstration, the framework was applied to optimal carbon sequestration reservoir planning using two different types of management strategies: monitoring only and brine extraction. Both strategies are designed to mitigate potential risks due to pressure buildup. Results show that the DQN agent can identify the optimal policies to maximize the reward for given risk and cost constraints. Finally, experiments also show that knowledge the agent gained from interacting with one environment is largely preserved when deploying the same agent in other similar environments.

15 GEOTHERMAL ENERGY↗

An algebraic method for constructing stable and consistent autoregressive filters

In this paper, we introduce an algebraic method to construct stable and consistent univariate autoregressive (AR) models of low order for filtering and predicting nonlinear turbulent signals with memory depth. By stable, we refer to the classical stability condition for the AR model. By consistent, we refer to the classical consistency constraints of Adams–Bashforth methods of order-two. One attractive feature of this algebraic method is that the model parameters can be obtained without directly knowing any training data set as opposed to many standard, regression-based parameterization methods. It takes only long-time average statistics as inputs. The proposed method provides a discretization time step interval which guarantees the existence of stable and consistent AR model and simultaneously produces the parameters for the AR models. In our numerical examples with two chaotic time series with different characteristics of decaying time scales, we find that the proposed AR models produce significantly more accurate short-term predictive skill and comparable filtering skill relative to the linear regression-based AR models. These encouraging results are robust across wide ranges of discretization times, observation times, and observation noise variances. Finally, we also find that the proposed model produces an improved short-time prediction relative to the linear regression-based AR-models in forecasting a data set that characterizes the variability of the Madden–Julian Oscillation, a dominant tropical atmospheric wave pattern.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Prediction of global solar irradiance based on time series analysis: Application to solar thermal power plants energy production planning

Due to strong increase of solar power generation, the predictions of incoming solar energy are acquiring more importance. Photovoltaic and solar thermal are the main sources of electricity generation from solar energy. In the case of solar thermal energy plants with storage energy system, its management and operation need reliable predictions of solar irradiance with the same temporal resolution as the temporal capacity of the back-up system. These plants can work like a conventional power plant and compete in the energy stock market avoiding intermittence in electricity production. This work presents a comparisons of statistical models based on time series applied to predict half daily values of global solar irradiance with a temporal horizon of 3 days. Half daily values consist of accumulated hourly global solar irradiance from solar raise to solar noon and from noon until dawn for each day. The dataset of ground solar radiation used belongs to stations of Spanish National Weather Service (AEMet). The models tested are autoregressive, neural networks and fuzzy logic models. Due to the fact that half daily solar irradiance time series is non-stationary, it has been necessary to transform it to two new stationary variables (clearness index and lost component) which are used as input of the predictive models. Improvement in terms of RMSD of the models essayed is compared against the model based on persistence. The validation process shows that all models essayed improve persistence. The best approach to forecast half daily values of solar irradiance is neural network models with lost component as input, except Lerida station where models based on clearness index have less uncertainty because this magnitude has a linear behaviour and it is easier to simulate by models. (author)

14 SOLAR ENERGY↗

Adaptive Reweighted Variance Estimation for Monte Carlo Eigenvalue Simulations

Monte Carlo (MC) simulation is used to solve the eigenvalue form of the Boltzmann transport equation to estimate various parameters such as fuel pin flux distributions that are crucial for the safe and efficient operation of nuclear systems (e.g., a nuclear reactor). Monte Carlo eigenvalue simulation uses a sample mean over many stationary cycles (iterations) to estimate various parameters important to nuclear systems. A variance estimate of the sample mean is often used for calculating the confidence intervals. However, MC eigenvalue simulation variance estimators that ignore the intercycle correlation underestimate the true variance of the estimated quantity. This paper presents novel data-adaptive approaches based on a simple autoregressive (AR) model and sigmoid functions to improve MC variance estimation. The standard MC sample-based variance estimator (or naïve estimator) and the spectral density–based MC variance estimator are enhanced by adding data-adaptive components that reduce their bias and improve performance. By investigating the frequency pattern of the AR(1) (order 1) model, two adaptive spectral estimators and one adaptive naïve estimator are proposed. The proposed estimators manifest superior performance when applied to three test problems compared to the standard spectral density–based estimator previously introduced by the authors. In conclusion, these new estimators are straightforward, as they use online algorithms and do not require storage of tallies from all active cycles.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

A Comparison of Infectious Disease Forecasting Methods across Locations, Diseases, and Time

Accurate infectious disease forecasting can inform efforts to prevent outbreaks and mitigate adverse impacts. This study compares the performance of statistical, machine learning (ML), and deep learning (DL) approaches in forecasting infectious disease incidences across different countries and time intervals. We forecasted three diverse diseases: campylobacteriosis, typhoid, and Q-fever, using a wide variety of features (n = 46) from public datasets, e.g., landscape, climate, and socioeconomic factors. We compared autoregressive statistical models to two tree-based ML models (extreme gradient boosted trees [XGB] and random forest [RF]) and two DL models (multi-layer perceptron and encoder–decoder model). The disease models were trained on data from seven different countries at the region-level between 2009–2017. Forecasting performance of all models was assessed using mean absolute error, root mean square error, and Poisson deviance across Australia, Israel, and the United States for the months of January through August of 2018. The overall model results were compared across diseases as well as various data splits, including country, regions with highest and lowest cases, and the forecasted months out (i.e., nowcasting, short-term, and long-term forecasting). Overall, the XGB models performed the best for all diseases and, in general, tree-based ML models performed the best when looking at data splits. There were a few instances where the statistical or DL models had minutely smaller error metrics for specific subsets of typhoid, which is a disease with very low case counts. Feature importance per disease was measured by using four tree-based ML models (i.e., XGB and RF with and without region name as a feature). The most important feature groups included previous case counts, region name, population counts and density, mortality causes of neonatal to under 5 years of age, sanitation factors, and elevation. This study demonstrates the power of ML approaches to incorporate a wide range of factors to forecast various diseases, regardless of location, more accurately than traditional statistical approaches.

59 BASIC BIOLOGICAL SCIENCES↗

On-line identification and attitude control for SCOLE

This paper documents on-line linear least-square identification and attitude control of SCOLE, a laboratory apparatus representing an offset-feed antenna attached to the Space Shuttle. Identification is done autonomously by starting a slew maneuver in pitch or roll with reaction jets and observing the time history data of associated Euler angles when the jets are quiescent. Linear least-square analysis is used to select the parameters that best fit the output of an Autoregressive (AR) model to the data. The control effectiveness of the jets is determined in a subsequent test, again using linear least squares. The parameters so derived are used to design switching lines for time-optimal attitude control. This report describes the identification and control algorithms and the experimental apparatus and procedures used. Also, experimental data are presented that reflect the performance of the identification algorithms and the attitude control system.

Montgomery, R. C.↗

Least squares linear lags and limited memory filters

Pure autoregressive (AR) models which are linear in the short term, that is when a variable can be predicted by linear regression on a limited number of past observations, are discussed. When evenly spaced observations are available, a fixed set of AR coefficients can be calculated independent of the data. For filtering purposes, such a lag structure can be implemented recursively with an efficient algorithm. The method of computing variance recursively is also derived. A complete algorithm is presented in the appendix.

Discenza, Joseph H.↗

Prediction of muscle performance during dynamic repetitive movement

BACKGROUND: During long-duration spaceflight, astronauts experience progressive muscle atrophy and often perform strenuous extravehicular activities. Post-flight, there is a lengthy recovery period with an increased risk for injury. Currently, there is a critical need for an enabling tool to optimize muscle performance and to minimize the risk of injury to astronauts while on-orbit and during post-flight recovery. Consequently, these studies were performed to develop a method to address this need. METHODS: Eight test subjects performed a repetitive dynamic exercise to failure at 65% of their upper torso weight using a Lordex spinal machine. Surface electromyography (SEMG) data was collected from the erector spinae back muscle. The SEMG data was evaluated using a 5th order autoregressive (AR) model and linear regression analysis. RESULTS: The best predictor found was an AR parameter, the mean average magnitude of AR poles, with r = 0.75 and p = 0.03. This parameter can predict performance to failure as early as the second repetition of the exercise. CONCLUSION: A method for predicting human muscle performance early during dynamic repetitive exercise was developed. The capability to predict performance to failure has many potential applications to the space program including evaluating countermeasure effectiveness on-orbit, optimizing post-flight recovery, and potential future real-time monitoring capability during extravehicular activity.

Physical Endurance/physiology↗

Dynamic regulation of heart rate during acute hypotension: new insight into baroreflex function

To examine the dynamic properties of baroreflex function, we measured beat-to-beat changes in arterial blood pressure (ABP) and heart rate (HR) during acute hypotension induced by thigh cuff deflation in 10 healthy subjects under supine resting conditions and during progressive lower body negative pressure (LBNP). The quantitative, temporal relationship between ABP and HR was fitted by a second-order autoregressive (AR) model. The frequency response was evaluated by transfer function analysis. Results: HR changes during acute hypotension appear to be controlled by an ABP error signal between baseline and induced hypotension. The quantitative relationship between changes in ABP and HR is characterized by a second-order AR model with a pure time delay of 0.75 s containing low-pass filter properties. During LBNP, the change in HR/change in ABP during induced hypotension significantly decreased, as did the numerator coefficients of the AR model and transfer function gain. Conclusions: 1) Beat-to-beat HR responses to dynamic changes in ABP may be controlled by an error signal rather than directional changes in pressure, suggesting a "set point" mechanism in short-term ABP control. 2) The quantitative relationship between dynamic changes in ABP and HR can be described by a second-order AR model with a pure time delay. 3) The ability of the baroreflex to evoke a HR response to transient changes in pressure was reduced during LBNP, which was due primarily to a reduction of the static gain of the baroreflex.

Clinical Trial↗

Stochastic Modeling of Length of Day

We propose a general stochastic model for the UT1-LOD system and derive the corresponding Kalman filter model.

stochastic model autoregressive moving average tra↗

Prediction of Muscle Performance During Dynamic Repetitive Exercise

A method for predicting human muscle performance was developed. Eight test subjects performed a repetitive dynamic exercise to failure using a Lordex spinal machine. Electromyography (EMG) data was collected from the erector spinae. Evaluation of the EMG data using a 5th order Autoregressive (AR) model and statistical regression analysis revealed that an AR parameter, the mean average magnitude of AR poles, can predict performance to failure as early as the second repetition of the exercise. Potential applications to the space program include evaluating on-orbit countermeasure effectiveness, maximizing post-flight recovery, and future real-time monitoring capability during Extravehicular Activity.

Byerly, D. L.↗

JMP Applications in Photovoltaic Reliability (Presentation)

The ability to accurately predict power delivery over the course of time is of vital importance to the growth of the photovoltaic (PV) industry. Two key cost drivers are the efficiency with which sunlight is converted into power and secondly how this relationship develops over time. The accurate knowledge of power decline over time, also known as degradation rates, is essential and important to all stakeholders--utility companies, integrators, investors, and scientist alike. Outdoor testing plays a vital part in quantifying degradation rates of different technologies in various climates. Due to seasonal changes, however, several complete cycles (typically 3-5 years) need to be completed traditionally to obtain reasonably accurate degradation rates. In a rapidly evolving industry such atime span is often unacceptable and the need exists to determine degradation rates more accurately in a shorter period of time. Advanced time series modeling such as ARIMA (Autoregressive Integrated Moving Average) modeling can be utilized to decrease the required time span and is compared with some non-linear modeling. In addition, it will be demonstrated how the JMP 9 map feature was used toreveal important technological trends by climate.

14 SOLAR ENERGY↗

STPMiner: A Highperformance Spatiotemporal Pattern Mining Toolbox

The volume of spatiotemporal data being generated from scientific simulations and observations from sensors is growing at an astronomical rate. This data explosion is going to pose three challenges to the existing data mining infrastructure: algorithmic, computational, and I/O. Over the years we have implemented several spatiotemporal data mining algorithms including: outliers/anomalies, colocation patterns, change patterns, clustering, classification, and prediction algorithms. In this paper we briefly discuss the core spatiotemporal pattern mining algorithms along with some of the computational and I/O challenges associated with the big data.

99 GENERAL AND MISCELLANEOUS↗