Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “time series encoding”

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 19 records

Power System Event Classification and Localization Using a Convolutional Neural Network

Detection and timely identification of power system disturbances are essential for situation awareness and reliable electricity grid operation. Because records of actual events in the system are limited, ensemble simulation-based events are needed to provide adequate data for building event-detection models through deep learning; e.g., a convolutional neural network (CNN). An ensemble numerical simulation-based training data set have been generated through dynamic simulations performed on the Polish system with various types of faults in different locations. Such data augmentation is proven to be able to provide adequate data for deep learning. The synchronous generators’ frequency signals are used and encoded into images for developing and evaluating CNN models for classification of fault types and locations. With a time-domain stacked image set as the benchmark, two different time-series encoding approaches, i.e., wavelet decomposition-based frequency-domain stacking and polar coordinate system-based Gramian Angular Field (GAF) stacking, are also adopted to evaluate and compare the CNN model performance and applicability. The various encoding approaches are suitable for different fault types and spatial zonation. With optimized settings of the developed CNN models, the classification and localization accuracies can go beyond 84 and 91%, respectively.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Results from field tests of the one-dimensional Time-Encoded Imaging System.

A series of field experiments were undertaken to evaluate the performance of the one dimensional time encoded imaging system. The significant detection of a Cf252 fission radiation source was demonstrated at a stand-off of 100 meters. Extrapolations to different quantities of plutonium equivalent at different distances are made. Hardware modifications to the system for follow on work are suggested.

Marleau, Peter↗

Results from laboratory tests of the two-dimensional Time-Encoded Imaging System.

A series of laboratory experiments were undertaken to demonstrate the feasibility of two dimensional time-encoded imaging. A prototype two-dimensional time encoded imaging system was designed and constructed. Results from imaging measurements of single and multiple point sources as well as extended source distributions are presented. Time encoded imaging has proven to be a simple method for achieving high resolution two-dimensional imaging with potential to be used in future arms control and treaty verification applications.

Marleau, Peter↗

Uncertainty-Guided Prediction Horizon of Phase-Resolved Ocean Wave Forecasting Under Data Sparsity: Experimental and Numerical Evaluation

Accurate short-term wave forecasting is critical for the safe and efficient operation of marine structures that rely on real-time, phase-resolved ocean wave information for control and monitoring purposes (e.g., digital twins). These systems often depend on environmental sensors (e.g., waverider buoys, wave-sensing LIDAR). Challenges arise when upstream sensor data are missing, sparse, or phase-shifted due to drift. This study investigates the performance of two machine learning models, time-series dense encoder (TiDE) and long short-term memory (LSTM), for forecasting phase-resolved ocean surface elevations under varying degrees of data degradation. We introduce the τ-trimming algorithm, which adapts the prediction horizon based on uncertainty thresholds derived from historical forecasts. Numerical wave tank (NWT) and wave basin experiments are used to benchmark model performance under short- and long-term data masking, spatially coarse sensor grids, and upstream phase shifts. Results show under a 50% probability of upstream data loss, the τ-trimmed TiDE model achieves a 46% reduction in error at the most upstream target, compared to 22% for LSTM. Furthermore, phase misalignment in upstream data introduces a near-linear increase in forecast error. Under moderate model settings, a ±3 s misalignment increases the mean absolute error by approximately 0.5 m, while the same error is accumulated at ±4 s using the more conservative approach. These findings inform the design of resilient, uncertainty-aware wave forecasting systems suited for realistic offshore sensing environments.

42 ENGINEERING↗

Characterizing system dynamics with a weighted and directed network constructed from time series data

In this work, we propose a novel method to transform a time series into a weighted and directed network. For a given time series, we first generate a set of segments via a sliding window, and then use a doubly symbolic scheme to characterize every windowed segment by combining absolute amplitude information with an ordinal pattern characterization. Based on this construction, a network can be directly constructed from the given time series: segments corresponding to different symbol-pairs are mapped to network nodes and the temporal succession between nodes is represented by directed links. With this conversion, dynamics underlying the time series has been encoded into the network structure. We illustrate the potential of our networks with a well-studied dynamical model as a benchmark example. Results show that network measures for characterizing global properties can detect the dynamical transitions in the underlying system. Moreover, we employ a random walk algorithm to sample loops in our networks, and find that time series with different dynamics exhibits distinct cycle structure. That is, the relative prevalence of loops with different lengths can be used to identify the underlying dynamics.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Wavelet-based surrogate time series for multiscale simulation of heterogeneous catalysis

We propose a wavelet-based scheme that encodes the essential dynamics of discrete microscale surface reactions in a form that can be coupled with continuum macroscale flow simulations with high computational efficiency. This makes it possible to simulate the dynamic behavior of reactor-scale heterogeneous catalysis without requiring detailed concurrent simulations at both the surface and continuum scales using different models. Our scheme is based on the application of wavelet-based surrogate time series that encodes the essential temporal and/or spatial fine-scale dynamics at the catalyst surface. The encoded dynamics are then used to generate statistically equivalent, randomized surrogate time series, which can be linked to the continuum scale simulation. As a result, we illustrate an application of this approach using two different kinetic Monte Carlo simulations with different characteristic behaviors typical for heterogeneous chemical reactions.

36 MATERIALS SCIENCE↗

Establishing resolution-improved NMR spectroscopy in high magnetic fields with unknown spatiotemporal variations

A half-century quest for higher magnetic fields has been an integral part of the progress undergone in the Nuclear Magnetic Resonance (NMR) study of materials’ structure and dynamics. Because 2D NMR relies on systematic changes in coherences’ phases as a function of an encoding time varied over a series of independent experiments, it generally cannot be applied in temporally unstable fields. This precludes most NMR methods from being used to characterize samples situated in hybrid or resistive magnets that are capable of achieving extremely high magnetic field strength. Recently, “ultrafast” NMR has been developed into an effective and widely applicable methodology enabling the acquisition of a multidimensional NMR spectrum in a single scan; it can therefore be used to partially mitigate the effects of temporally varying magnetic fields. Nevertheless, the strong interference of fluctuating fields with the spatial encoding of ultrafast NMR still severely restricts measurement sensitivity and resolution. Here, we introduce a strategy for obtaining high resolution NMR spectra that exploits the immunity of intermolecular zero-quantum coherences (iZQCs) to field instabilities and inhomogeneities. The spatial encoding of iZQCs is combined with a J-modulated detection scheme that removes the influence of arbitrary field inhomogeneities during acquisition. This new method can acquire high-resolution one-dimensional NMR spectra in large inhomogeneous and fluctuating fields, and it is tested with fields experimentally modeled to mimic those of resistive and resistive-superconducting hybrid magnets.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Autoregressive modeling with state-space embedding vectors for damage detection under operational and environmental variability

A nonlinear time series approach is presented to detect damage in systems by using a state-space reconstruction to infer the geometrical structure of a deterministic dynamical system from observed time series response at multiple locations. The unique contribution of this approach is using a Multivariate Autoregressive (MAR) model of a baseline condition to predict the state space, where the model encodes the embedding vectors rather than scalar time series. A hypothesis test is established that the MAR model will fail to predict future response if damage is present in the test condition, and this test is investigated for robustness in the context of operational and environmental variability. The applicability of this approach is demonstrated using acceleration time series from a base-excited 3-story frame structure.

42 ENGINEERING↗

SigTime: Learning and Visually Explaining Time Series Signatures

Understanding and distinguishing temporal patterns in time series data is essential for scientific discovery and decision-making. For example, in biomedical research, uncovering meaningful patterns in physiological signals can improve diagnosis, risk assessment, and patient outcomes. However, existing methods for time series pattern discovery face major challenges, including high computational complexity, limited interpretability, and difficulty in capturing meaningful temporal structures. Here, to address these gaps, we introduce a novel learning framework that jointly trains two Transformer models using complementary time series representations: shapelet-based representations to capture localized temporal structures and traditional feature engineering to encode statistical properties. The learned shapelets serve as interpretable signatures that differentiate time series across classification labels. Additionally, we develop a visual analytics system—SigTime—with coordinated views to facilitate exploration of time series signatures from multiple perspectives, aiding in useful insights generation. We quantitatively evaluate our learning framework on eight publicly available datasets and one proprietary clinical dataset. Additionally, we demonstrate the effectiveness of our system through two usage scenarios along with the domain experts: one involving public ECG data and the other focused on preterm labor analysis.

97 MATHEMATICS AND COMPUTING↗

Dynamic energy system modeling using hybrid physics-based and machine learning encoder–decoder models

Three model configurations are presented for multi-step time series predictions of the heat absorbed by the water and steam in a thermal power plant. The models predict over horizons of 2, 4, and 6 steps into the future, where each step is a 5-minute increment. The evaluated models are a pure machine learning model, a novel hybrid machine learning and physics-based model, and the hybrid model with an incomplete dataset. The hybrid model deconstructs the machine learning into individual boiler heat absorption units: economizer, water wall, superheater, and reheater. Each configuration uses a gated recurrent unit (GRU) or a GRU-based encoder–decoder as the deep learning architecture. Mean squared error is used to evaluate the models compared to target values. The encoder–decoder architecture is over 11% more accurate than the GRU only models. The hybrid model with the incomplete dataset highlights the importance of the manipulated variables to the system. The hybrid model, compared to the pure machine learning model, is over 10% more accurate on average over 20 iterations of each model. Automatic differentiation is applied to the hybrid model to perform a local sensitivity analysis to identify the most impactful of the 72 manipulated variables on the heat absorbed in the boiler. The models and sensitivity analyses are used in a discussion about optimizing the thermal power plant.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Autonomous extraction of millimeter-scale deformation in InSAR time series using deep learning

Systematically characterizing slip behaviours on active faults is key to unraveling the physics of tectonic faulting and the interplay between slow and fast earthquakes. Interferometric Synthetic Aperture Radar (InSAR), by enabling measurement of ground deformation at a global scale every few days, may hold the key to those interactions. However, atmospheric propagation delays often exceed ground deformation of interest despite state-of-the art processing, and thus InSAR analysis requires expert interpretation and a priori knowledge of fault systems, precluding global investigations of deformation dynamics. Here, we show that a deep auto-encoder architecture tailored to untangle ground deformation from noise in InSAR time series autonomously extracts deformation signals, without prior knowledge of a fault’s location or slip behaviour. Applied to InSAR data over the North Anatolian Fault, our method reaches 2 mm detection, revealing a slow earthquake twice as extensive as previously recognized. We further explore the generalization of our approach to inflation/deflation-induced deformation, applying the same methodology to the geothermal field of Coso, California.

58 GEOSCIENCES↗

On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars

Despite the utility of neural networks (NNs) for astronomical time-series classification, the proliferation of learning architectures applied to diverse data sets has thus far hampered a direct intercomparison of different approaches. Here we perform the first comprehensive study of variants of NN-based learning and inference for astronomical time series, aiming to provide the community with an overview on relative performance and, hopefully, a set of best-in-class choices for practical implementations. In both supervised and self-supervised contexts, we study the effects of different time-series-compatible layer choices, namely the dilated temporal convolutional neural network (dTCNs), long-short term memory NNs, gated recurrent units and temporal convolutional NNs (tCNNs). Additionally, we also study the efficacy and performance of encoder-decoder (i.e., autoencoder) networks compared to direct classification networks, different pathways to include auxiliary (non-time-series) metadata, and different approaches to incorporate multi-passband data (i.e., multiple time series per source). Performance—applied to a sample of 17,604 variable stars (VSs) from the MAssive Compact Halo Objects (MACHO) survey across 10 imbalanced classes—is measured in training convergence time, classification accuracy, reconstruction error, and generated latent variables. We find that networks with recurrent NNs generally outperform dTCNs and, in many scenarios, yield to similar accuracy as tCNNs. In learning time and memory requirements, convolution-based layers perform better. We conclude by discussing the advantages and limitations of deep architectures for VS classification, with a particular eye toward next-generation surveys such as the Legacy Survey of Space and Time, the Roman Space Telescope, and Zwicky Transient Facility.

79 ASTRONOMY AND ASTROPHYSICS↗

Relation Inference among Sensor Time Series in Smart Buildings with Metric Learning

Smart Building Technologies hold promise for better livability for residents and lower energy footprints. Yet, the rollout of these technologies, from demand response controls to fault detection and diagnosis, significantly lags behind and is impeded by the current practice of manual identification of sensing point relationships, e.g., how equipment is connected or which sensors are co-located in the same space. This manual process is still error-prone, albeit costly and laborious.We study relation inference among sensor time series. Our key insight is that, as equipment is connected or sensors co-locate in the same physical environment, they are affected by the same real-world events, e.g., a fan turning on or a person entering the room, thus exhibiting correlated changes in their time series data. To this end, we develop a deep metric learning solution that first converts the primitive sensor time series to the frequency domain, and then optimizes a representation of sensors that encodes their relations. Built upon the learned representation, our solution pinpoints the relationships among sensors via solving a combinatorial optimization problem. Extensive experiments on real-world buildings demonstrate the effectiveness of our solution.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Controller area network decoder (CAN-D)

A system and method for decoding an unknown automotive controller area network (“CAN”) message definitions. CAN data vehicle signal mappings are typically held in secret and varied by automotive model and year. Without knowledge of the mappings, the wealth of real-time vehicle data hidden in the automotive CAN packets is uninterpretable—impeding research, after-market tuning, efficiency and performance monitoring, fault diagnosis, and privacy-related technologies. This technology can ascertain the CAN signals' boundaries (start bit and length), endianness (byte ordering), signedness (binary-to-integer encoding) from raw CAN data. This allows conversion of CAN data to time series. Interpreting the translated CAN data's physical meaning and finding a linear mapping to standard units (e.g., knowing the signal is speed and scaling values to represent units of miles per hour) can be achieved for many signals by leveraging diagnostic standards to obtain real-time measurements of in-vehicle systems. The system and method can be integrated into lightweight hardware enabling an OBD-II plugin for real-time in-vehicle CAN decoding or run on standard computers. The system can output a standard DBC file with the signal definition information.

Verma, Kiren E.↗

System for controller area network payload decoding

A system for decoding an unknown automotive controller area network (“CAN”) message definitions. CAN data vehicle signal mappings are typically held in secret and varied by automotive model and year. Without knowledge of the mappings, the wealth of real-time vehicle data hidden in the automotive CAN packets is uninterpretable—impeding research, after-market tuning, efficiency and performance monitoring, fault diagnosis, and privacy-related technologies. This system can ascertain the CAN signals' boundaries (start bit and length), endianness (byte ordering), signedness (binary-to-integer encoding) from raw CAN data. This allows conversion of CAN data to time series. Interpreting the translated CAN data's physical meaning and finding a linear mapping to standard units (e.g., knowing the signal is speed and scaling values to represent units of miles per hour) can be achieved for many signals by leveraging diagnostic standards to obtain real-time measurements of in-vehicle systems. The system can be integrated into lightweight hardware enabling an OBD-II plugin for real-time in-vehicle CAN decoding or run on standard computers. The system can output a standard DBC file with the signal definition information.

Verma, Kiren E.↗

Optical delay encoding for fast timing and detector signal multiplexing in PET

Purpose: The large number of detector channels in modern positron emission tomography (PET) scanners poses a challenge in terms of readout electronics complexity. Multiplexing schemes are typically implemented to reduce the number of physical readout channels, but often result in performance degradation. Novel methods of multiplexing in PET must be developed to avoid this data degradation. The preservation of fast timing information is especially important for time-of-flight PET. Methods: A new multiplexing scheme based on encoding detector interaction events with a series of extremely fast overlapping optical pulses with precise delays is demonstrated in this work. Encoding events in this way potentially allows many detector channels to be simultaneously encoded onto a single optical fiber that is then read out by a single digitizer. A two channel silicon photomultiplier-based prototype utilizing this optical delay encoding technique along with dual threshold time-over-threshold is demonstrated. Results: The optical encoding and multiplexing prototype achieves a coincidence time resolution of 160 ps full width at half maximum (FWHM) and an energy resolution of 13.1% FWHM at 511 keV with 3 × 3 × 5 mm{sup 3} LYSO crystals. All interaction information for both detectors, including timing, energy, and channel identification, is encoded onto a single optical fiber with little degradation. Conclusions: Optical delay encoding and multiplexing technology could lead to time-of-flight PET scanners with fewer readout channels and simplified data acquisition systems.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Hardware-Based Emulator with Deep Learning Model for Building Energy Control and Prediction Based on Occupancy Sensors’ Data

Heating, ventilation, and air conditioning (HVAC) is the largest source of residential energy consumption. Occupancy sensors’ data can be used for HVAC control since it indicates the number of people in the building. HVAC and sensors form a typical cyber-physical system (CPS). In this paper, we aim to build a hardware-based emulation platform to study the occupancy data’s features, which can be further extracted by using machine learning models. In particular, we propose two hardware-based emulators to investigate the use of wired/wireless communication interfaces for occupancy sensor-based building CPS control, and the use of deep learning to predict the building energy consumption with the sensor data. We hypothesize is that the building energy consumption may be predicted by using the occupancy data collected by the sensors, and question what type of prediction model should be used to accurately predict the energy load. Another hypothesis is that an in-lab hardware/software platform could be built to emulate the occupancy sensing process. The machine learning algorithms can then be used to analyze the energy load based on the sensing data. To test the emulator, the occupancy data from the sensors is used to predict energy consumption. The synchronization scheme between sensors and the HVAC server will be discussed. We have built two hardware/software emulation platforms to investigate the sensor/HVAC integration strategies, and used an enhanced deep learning model—which has sequence-to-sequence long short-term memory (Seq2Seq LSTM)—with an attention model to predict the building energy consumption with the preservation of the intrinsic patterns. Because the long-range temporal dependencies are captured, the Seq2Seq models may provide a higher accuracy by using LSTM architectures with encoder and decoder. Meanwhile, LSTMs can capture the temporal and spatial patterns of time series data. The attention model can highlight the most relevant input information in the energy prediction by allocating the attention weights. The communication overhead between the sensors and the HVAC control server can also be alleviated via the attention mechanism, which can automatically ignore the irrelevant information and amplify the relevant information during CNN training. Our experiments and performance analysis show that, compared with the traditional LSTM neural network, the performance of the proposed method has a 30% higher prediction accuracy.

Ye, Zhijing↗

Improving Quality of Observational Streaming Medical Data by Using Long Short-Term Memory Networks (LSTMs)

We present an exploration of the encoder-decoder structured Long Short-Term Memory Network (LSTM) as a detector of the anomalous missing observations in streaming medical data by using the difference between the LSTM-reconstructed and observed values as the anomaly detector. We experiment with time-series data from bedside monitoring devices from the available Medical Information Mart for Intensive Care Database (MIMIC). Our results show that not only encoder-decoder LSTM approach works well for detecting the difference between anomalous and normal missing observations in streaming medical data, but also has an imputation potential for the missing data.

Bowie, Michael B.↗