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 37 records · Page 2

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.↗

Several recursive techniques for observer/Kalman filter system identification from data

This paper derives algorithms for identifying autoregressive models, with external input, of multi-input multi-output systems from data using a fast transversal filter or a least-squares lattice filter. The autoregressive models including external inputs are used to identify state-space models and the corresponding observer/Kalman filter gains of the system. The derivation is an extension of scalar autoregressive model approaches, modified to cope with multivariables, external inputs and an extra direct-influence term. Comparisons between the fast transversal filter, the least-squares lattice filter and the classical least-squares method are made in terms of complexity, computational cost and practical applications issues. A numerical example is included to illustrate the approach.

Chen, Chung-Wen↗

Real-time processing of radar return on a parallel computer

NASA is working with the FAA to demonstrate the feasibility of pulse Doppler radar as a candidate airborne sensor to detect low altitude windshears. The need to provide the pilot with timely information about possible hazards has motivated a demand for real-time processing of a radar return. Investigated here is parallel processing as a means of accommodating the high data rates required. A PC based parallel computer, called the transputer, is used to investigate issues in real time concurrent processing of radar signals. A transputer network is made up of an array of single instruction stream processors that can be networked in a variety of ways. They are easily reconfigured and software development is largely independent of the particular network topology. The performance of the transputer is evaluated in light of the computational requirements. A number of algorithms have been implemented on the transputers in OCCAM, a language specially designed for parallel processing. These include signal processing algorithms such as the Fast Fourier Transform (FFT), pulse-pair, and autoregressive modelling, as well as routing software to support concurrency. The most computationally intensive task is estimating the spectrum. Two approaches have been taken on this problem, the first and most conventional of which is to use the FFT. By using table look-ups for the basis function and other optimizing techniques, an algorithm has been developed that is sufficient for real time. The other approach is to model the signal as an autoregressive process and estimate the spectrum based on the model coefficients. This technique is attractive because it does not suffer from the spectral leakage problem inherent in the FFT. Benchmark tests indicate that autoregressive modeling is feasible in real time.

Aalfs, David D.↗

Utilization of Model Predictive Control to Balance Power Absorption Against Load Accumulation

Wave energy converter (WEC) control strategies have been primarily focused on maximizing power absorption. The use of model predictive control strategies allows for a finite-horizon, multiterm objective function to be solved. This work utilizes a multiterm objective function to maximize power absorption while minimizing the structural loads on the WEC system. Furthermore, a Kalman filter and autoregressive model were used to estimate and forecast the wave exciting force and predict the future dynamics of the WEC. The WEC's power-take-off time-averaged power and structural loads under a perfect forecast assumption in irregular waves were compared against results obtained from the Kalman filter and autoregressive model to evaluate model predictive control performance.

16 TIDAL AND WAVE POWER↗

Utilization of Model Predictive Control to Balance Power Absorption Against Load Accumulation: Preprint

Wave energy converter (WEC) control strategies have been primarily focused on maximizing power absorption. The use of model predictive control strategies allows for a finite-horizon, multiterm objective function to be solved. This work utilizes a multiterm objective function to maximize power absorption while minimizing the structural loads on the WEC system. Furthermore, a Kalman filter and autoregressive model were used to estimate and forecast the wave exciting force and predict the future dynamics of the WEC. The WEC's power-take-off time-averaged power and structural loads under a perfect forecast assumption in irregular waves were compared against results obtained from the Kalman filter and autoregressive model to evaluate model predictive control performance.

16 TIDAL AND WAVE POWER↗

Fault detection using a two-model test for changes in the parameters of an autoregressive time series

This article describes an investigation of a statistical hypothesis testing method for detecting changes in the characteristics of an observed time series. The work is motivated by the need for practical automated methods for on-line monitoring of Deep Space Network (DSN) equipment to detect failures and changes in behavior. In particular, on-line monitoring of the motor current in a DSN 34-m beam waveguide (BWG) antenna is used as an example. The algorithm is based on a measure of the information theoretic distance between two autoregressive models: one estimated with data from a dynamic reference window and one estimated with data from a sliding reference window. The Hinkley cumulative sum stopping rule is utilized to detect a change in the mean of this distance measure, corresponding to the detection of a change in the underlying process. The basic theory behind this two-model test is presented, and the problem of practical implementation is addressed, examining windowing methods, model estimation, and detection parameter assignment. Results from the five fault-transition simulations are presented to show the possible limitations of the detection method, and suggestions for future implementation are given.

Scholtz, P.↗

Improvements in the perturbation simulations of the global reference atmospheric model

The Global Reference Atmospheric Model (GRAM) program includes the capability for simulating pseudo-random perturbations in density, temperature, pressure, or wind components along a simulated reentry trajectory or other path through the atmosphere. Some concerns were expressed by GRAM users, however, that the mean-square perturbation gradients may be too large for small values of the vertical separation Delta z. The present GRAM perturbation simulations, based on a one-step autoregressive model, yield a power spectrum versus wavenumber k which is proportional to k sup -2 at high wavenumbers. This feature also produces mean-square perturbation differences which are directly proportional to Delta z, and mean-square perturbation gradients which are inversely proportional to Delta z. Thus, root-mean-square gradients, (Delta f/Delta z) sub rms, increase with decreasing Delta a as Delta z sup -1/2. A simple modification to GRAM is suggested which overcomes this problem, i.e., which produce root-mean-square gradient that remain bound as Delta z approaches zero. Possible applications of more sophisticated simulation approaches, e.g., second order autoregressive models, or fractal model techniques, were also explored briefly but found to yield improvements which appear too small to justify their considerable added complexity for use in the GRAM programs.

Justus, C. G.↗

Sapling growth rates reveal conspecific negative density dependence in a temperate forest

Local tree species diversity is maintained in part by conspecific negative density dependence (CNDD). This pervasive mechanism occurs in a variety of forms and ecosystems, but research to date has been heavily skewed toward tree seedling survival in tropical forests. To evaluate CNDD more broadly, we investigated how sapling growth rates were affected by conspecific adult neighbors in a fully mapped 25.6 ha temperate deciduous forest. We examined growth rates as a function of the local adult tree neighborhood (via spatial autoregressive modeling) and compared the spatial positioning of faster-growing and slower-growing saplings with respect to adult conspecific and heterospecific trees (via bivariate point pattern analysis). In addition, to determine whether CNDD-driven variation in growth rates leaves a corresponding spatial signal, we extended our point pattern analysis to a static, growth-independent comparison of saplings and the next larger size class. We found that negative conspecific effects on sapling growth were most prevalent. Five of the nine species that were sufficiently abundant for analysis exhibited CNDD, while only one species showed evidence of a positive conspecific effect, and one or two species, depending on the analysis, displayed heterospecific effects. There was general agreement between the autoregressive models and the point pattern analyses based on sapling growth rates, but point pattern analyses based on single-point-in-time size classes yielded results that differed markedly from the other two approaches. Our work adds to the growing body of evidence that CNDD is an important force in temperate forests, and demonstrates that this process extends to sapling growth rates. Further, our findings indicate that point pattern analyses based solely on size classes may fail to detect the process of interest (e.g., neighborhood-driven variation in growth rates), in part due to the confounding of tree size and age.

54 ENVIRONMENTAL SCIENCES↗

Quantifying and simulating the weather forecast uncertainty for advanced building control

Weather forecast uncertainty is unavoidable despite technological advancements. Accurately quantifying and modelling this uncertainty is essential for developing and comparing advanced building controllers. In this study, we present a structured approach using a first-order autoregressive model (AR(1)) to model uncertainty in ambient temperature and global solar irradiation (GHI) forecasts. We analyzed weather data from four cities and employed Jensen–Shannon divergence (JSD) to evaluate the similarity between synthetic and actual forecast errors. The average JSD values for temperature are 0.027 (Berkeley), 0.021 (Leuven), 0.018 (Berlin), and 0.008 (Oslo), and for GHI, the average JSD values are 0.016 (Berkeley), 0.058 (Leuven), and 0.013 (Berlin). The low JSD values indicate a high similarity between the synthetic and real forecast error distributions. Further, our approach successfully generates synthetic weather forecasts that mirror the statistical properties of actual forecasts. The implementation of our method for uncertain forecast generation is being added to the BOPTEST framework.

54 ENVIRONMENTAL SCIENCES↗

Failure monitoring in dynamic systems: Model construction without fault training data

Advances in the use of autoregressive models, pattern recognition methods, and hidden Markov models for on-line health monitoring of dynamic systems (such as DSN antennas) have recently been reported. However, the algorithms described in previous work have the significant drawback that data acquired under fault conditions are assumed to be available in order to train the model used for monitoring the system under observation. This article reports that this assumption can be relaxed and that hidden Markov monitoring models can be constructed using only data acquired under normal conditions and prior knowledge of the system characteristics being measured. The method is described and evaluated on data from the DSS 13 34-m beam wave guide antenna. The primary conclusion from the experimental results is that the method is indeed practical and holds considerable promise for application at the 70-m antenna sites where acquisition of fault data under controlled conditions is not realistic.

Smyth, P.↗

Development and assessment of a model predictive controller enabling anticipatory control strategies for a heat-pipe system

To support the reliable and resilient operation of modular reactors and microreactors, anticipatory control strategies have been proposed for achieving faster-than-real-time predictions and decision-making capabilities in anticipation of potential anomalies, including setpoint changes and cyber incidents. Here this work presents how anticipatory control strategies can be implemented via model predictive control (MPC) of a single heat pipe’s temperature. Considering the uncertainty in developing and applying MPC, this work evaluates MPC performance given three different model forms: a linear response surface model, an artificial neural network (ANN), and an autoregressive model with exogenous input (ARX). This work also evaluates the impacts of different input biases and variance on MPC performance in order to account for potential sensor reading variations due to cyber incidents. We observe that nonparametric models such as the ANN and ARX result in more fluctuated control actions compared to the MPC applied to the linear response surface model. However, when the cyber incidents are of a large magnitude, the linear response surface model produces smaller feasible regions than the nonparametric models under identical constraints.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Computing the Critical Temperature of the Affine-Transformed $D=3$ Ising Model Using Masked Autoregressive Flow

The simple Ising model provides a rich environment to build and study lattice field theories. As part of an ongoing project to construct a conformal field theory (CFT) on an arbitrarily curved manifold, in this work we develop methods to measure the critical temperature $β_c$ of the affine-transformed Ising model on the face-centered cubic (FCC) lattice. The main challenge in this endeavor is finding a computationally efficient and accurate method of interpolating and extrapolating Monte Carlo observables with respect to coupling coefficients and temperature. Herein, we compare two such methods. A traditional statistical approach uses the multiple histogram (MH) method, while a newer machine learning approach uses a masked autoregressive flow (MAF) to estimate the underlying probability density function of a set of observables. While the MH method is specifically designed to interpolate and extrapolate Monte Carlo observables, we find that MAF is a viable alternative for measuring $β_c$ with a computational cost that scales more favorably. Furthermore, we comment on additional advantages of MAF relevant to our work, such as extrapolating in system volume.

Svenson, Kai [Texas U.]↗

Linear and nonlinear trending and prediction for AVHRR time series data

The variability of AVHRR calibration coefficient in time was analyzed using algorithms of linear and non-linear time series analysis. Specifically we have used the spline trend modeling, autoregressive process analysis, incremental neural network learning algorithm and redundancy functional testing. The analysis performed on available AVHRR data sets revealed that (1) the calibration data have nonlinear dependencies, (2) the calibration data depend strongly on the target temperature, (3) both calibration coefficients and the temperature time series can be modeled, in the first approximation, as autonomous dynamical systems, (4) the high frequency residuals of the analyzed data sets can be best modeled as an autoregressive process of the 10th degree. We have dealt with a nonlinear identification problem and the problem of noise filtering (data smoothing). The system identification and filtering are significant problems for AVHRR data sets. The algorithms outlined in this study can be used for the future EOS missions. Prediction and smoothing algorithms for time series of calibration data provide a functional characterization of the data. Those algorithms can be particularly useful when calibration data are incomplete or sparse.

Smid, J.↗

Integrated system identification and modal state estimation for control of flexible space structures

A novel approach of integrated system identification and modal state estimation is proposed for control of linear dynamical systems including flexible space structures. There are four steps involved in this approach. First, the relation between a stochastic state space model of a dynamical system and the coefficients of its autoregressive model with exogenous input is derived. Second, an adaptive least-squares transversal predictor is used to estimate the coefficients of the model. Third, a state space model and a steady state Kalman filter gain of the dynamical system are then identified from the coefficients of the model by using the eigensystem realization algorithm. Fourth, a modal state estimator is constructed using the modal parameters of the identified model. On-line implementation of this algorithm can continually improve the modal parameters and the filter gain. It can also gradually update the system model when the system characteristics are slowly changing. A numerical example is used to illustrate the feasibility of the new approach.

Chen, Chung-Wen↗

An Equivalent Time-Variant Storage Model to Harness EV Flexibility: Forecast and Aggregation

The demand for vehicle charging will necessitate large investments in power distribution, transmission, and generation. However, this demand is often also flexible in time, and can be actively managed to reduce the needed investments, and to better integrate renewable electricity. Harnessing this flexibility requires forecasting and controlling electric vehicle (EV) charging at thousands of stations. In this work, we address the problem of forecasting and management of the aggregate flexible demand from tens to thousands of EV supply equipment (EVSEs). First, it presents an equivalent time-variant storage model for flexible demand at an aggregation of EVSEs. The proposed model is generalizable to different markets, and also to different flexible loads. Model parameters representing multiple EVSEs can be easily aggregated by summation, and forecasted using autoregressive models. The forecastability of uncontrolled demand and storage parameters is evaluated using data from 1341 nonresidential EVSEs located in Northern California. The median coefficient of variation is as low as 24% for the forecast of uncontrolled demand at the highest aggregation and 10-15% for the storage parameters. The benefits of aggregation and forecastability are demonstrated using an energy arbitrage scenario. Purchasing energy day ahead is less expensive than in the real-time market, but relies on a uncertain forecast of charging availability. The results show that the forecastability significantly improves for larger aggregations. This helps the aggregator make a better forecast, and decreases the cost of charging in comparison to an uncontrolled case by 60% with respect to an oracle scenario.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Multi-fidelity classification using Gaussian processes: Accelerating the prediction of large-scale computational models

We report that machine learning techniques typically rely on large datasets to create accurate classifiers. However, there are situations when data is scarce and expensive to acquire. This is the case of studies that rely on state-of-the-art computational models which typically take days to run, thus hindering the potential of machine learning tools. In this work, we present a novel classifier that takes advantage of lower fidelity models and inexpensive approximations to predict the binary output of expensive computer simulations. We postulate an autoregressive model between the different levels of fidelity with Gaussian process priors. We adopt a fully Bayesian treatment for the hyper-parameters and use Markov Chain Monte Carlo samplers. We take advantage of the probabilistic nature of the classifier to implement active learning strategies. We also introduce a sparse approximation to enhance the ability of the multi-fidelity classifier to handle a large amount of low fidelity samples. We test these multi-fidelity classifiers against their single-fidelity counterpart with synthetic data, showing a median computational cost reduction of 23% for a target accuracy of 90%. In an application to cardiac electrophysiology, the multi-fidelity classifier achieves an F1 score, the harmonic mean of precision and recall, of 99.6% compared to 74.1% of a single-fidelity classifier when both are trained with 50 samples. In general, our results show that the multi-fidelity classifiers outperform their single-fidelity counterpart in terms of accuracy in all cases. Finally, we envision that this new tool will enable researchers to study classification problems that would otherwise be prohibitively expensive. Source code is available at https://github.com/fsahli/MFclass.

97 MATHEMATICS AND COMPUTING↗