Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Time Series Data”

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 523 records · Page 29

Data-driven Resilience Characterization of Control Dynamical Systems

In this paper, we define and quantify resiliency of a power network and propose data-driven algorithms for computing the same for the power grid. To do this, we use the Koopman operator framework to lift the controlled dynamical system to an abstract (possibly higher) dimensional space, where the evolution is linear. The linear system representation allows us to relate small time local controllability and observability of a general nonlinear control system to the controllability and observability of the lifted linear system. Finally, we define the resiliency of the underlying power grid in terms of the controllability and observability gramians of the lifted linear system. We illustrate the proposed approach to compute the resiliency metrics on time-series data obtained from a microgrid.

koopman operator, resilience, control↗

Understanding Building Infrastructure and Building Operation through DOE Asset Score Model: Lessons Learned from a Pilot Project

The U.S. Department of Energy (DOE) is developing a national voluntary energy asset score system to help building owners to evaluate the as-built physical characteristics (including building envelope, the mechanical and electrical systems) and overall building energy efficiency, independent of occupancy and operational choices. The energy asset score breaks down building energy use information by simulating building performance under typical operating and occupancy conditions for a given use type. A web-based modeling tool, the energy asset score tool facilitates the implementation of the asset score system. The tool consists of a simplified user interface built on a centralized simulation engine (EnergyPlus). It is intended to reduce both the implementation cost for the users and increase modeling standardization compared with an approach that requires users to build their own energy models. A pilot project with forty-two buildings (consisting mostly offices and schools) was conducted in 2012. This paper reports the findings. Participants were asked to collect a minimum set of building data and enter it into the asset score tool. Participants also provided their utility bills, existing ENERGY STAR scores, and previous energy audit/modeling results if available. The results from the asset score tool were compared with the building energy use data provided by the pilot participants. Three comparisons were performed. First, the actual building energy use, either from the utility bills or via ENERGY STAR Portfolio Manager, was compared with the modeled energy use. It was intended to examine how well the energy asset score represents a building’s system efficiencies, and how well it is correlated to a building’s actual energy consumption. Second, calibrated building energy models (where they exist) were used to examine any discrepancies between the asset score model and the pilot participant buildings’ [known] energy use pattern. This comparison examined the end use breakdowns and more detailed time series data. Third, ASHRAE 90.1 prototype buildings were also used as an industry standard modeling approach to test the accuracy level of the asset score tool. Our analysis showed that the asset score tool, which uses simplified building simulation, could provide results comparable to a more detailed energy model. The buildings’ as-built efficiency can be reflected in the energy asset score. An analysis between the modeled energy use through the asset score tool and the actual energy use from the utility bills can further inform building owners about the effectiveness of their building’s operation and maintenance.

Wang, Na↗

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↗

Comparison of CEAS and Williams-type models for spring wheat yields in North Dakota and Minnesota

The CEAS and Williams-type yield models are both based on multiple regression analysis of historical time series data at CRD level. The CEAS model develops a separate relation for each CRD; the Williams-type model pools CRD data to regional level (groups of similar CRDs). Basic variables considered in the analyses are USDA yield, monthly mean temperature, monthly precipitation, and variables derived from these. The Williams-type model also used soil texture and topographic information. Technological trend is represented in both by piecewise linear functions of year. Indicators of yield reliability obtained from a ten-year bootstrap test of each model (1970-1979) demonstrate that the models are very similar in performance in all respects. Both models are about equally objective, adequate, timely, simple, and inexpensive. Both consider scientific knowledge on a broad scale but not in detail. Neither provides a good current measure of modeled yield reliability. The CEAS model is considered very slightly preferable for AgRISTARS applications.

Barnett, T. L.↗

Data-Driven Operator Theoretic Methods for Phase Space Learning and Analysis

This paper uses data-driven operator theoretic approaches to explore the global phase space of a dynamical system. In this work, we defined conditions for discovering new invariant subspaces in the state space of a dynamical system starting from an invariant subspace based on the spectral properties of the Koopman operator. When the system evolution is known locally in several invariant subspaces in the state space of a dynamical system, a phase space stitching result is derived that yields the global Koopman operator. Additionally, in the case of equivariant systems, a phase space stitching result is developed to identify the global Koopman operator using the symmetry properties between the invariant subspaces of the dynamical system and time-series data from any one of the invariant subspaces. Finally, these results are extended to topologically conjugate dynamical systems; in particular, the relation between the Koopman tuple of topologically conjugate systems is established. The proposed results are demonstrated on several second-order nonlinear dynamical systems including a bistable toggle switch. Our method elucidates a strategy for designing discovery experiments: experiment execution can be done in many steps, and models from different invariant subspaces can be combined to approximate the global Koopman operator.

42 ENGINEERING↗

AI-Based Analytics and Energy Modeling Framework for Characterizing Urban Energy Systems

Developing location-specific district energy models is essential for understanding energy patterns and supporting efficient management and planning decisions. However, accurately characterizing these models remains challenging due to gaps in building characteristics and labor-intensive traditional modeling workflows. To address these challenges, we develop an AI-based framework that integrates top-down and bottom-up building energy data to automate urban energy model characterization. The framework trains multimodal deep learning models using heterogeneous ResStockTM datasets to infer missing building characteristics from varying levels of known information and generate simulation-ready inputs for district-scale energy modeling. It also employs a conditioning-based injection approach to generate ”what-if” scenarios, enabling users to explore retrofit, efficiency, and technology-upgrade pathways. Integrated within URBANoptTM, a bottom-up district energy modeling platform for simulating co-located buildings, the framework infers detailed building-level inputs required for bottom-up simulations. Both localized and generalized AI models are developed to learn relationships across categorical, numerical, and time-series data, enabling reconstruction of missing attributes and generation of targeted upgrade scenarios. We demonstrate this methodology on a residential neighborhood in Baltimore, MD, assessing internal consistency against ResStock reference data and URBANopt simulation, and comparing selected attributes against real-world building characteristics. Results show strong overall predictive accuracy in data completion and scenario generation, with localized and generalized models offering complementary trade-offs between precision and scalability. Overall, our automated framework streamlines energy modeling and provides a reliable framework for urban building energy characterization.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Unsupervised Deep Learning Can Identify Physical Parameters

Parametrizing a problem is one of the biggest obstacles toward the application of physical methods to novel systems. In this research, we show that deep learning methods can be used to accelerate this parametrization process. We use an unsupervised learning algorithm, a variational autoencoder, to extract meaningful physical parameters from unlabeled time-series generated from classical physical systems. For example, given time series data of projectile motion and the instruction to represent the data in a small number of unspecified parameters, the model can discover, without further direction, the concepts of velocity and gravity. Using this method, one can uncover information about the physics of a system, even when the relevant physical parameters are unknown to the experimenter. This has obvious applications to novel systems in which parameters are not known a priori. Moreover, errors in the reconstruction of the parameters can be used to reveal meaningful info rmation about state transitions, which can be conceptualized as a generalization of similar unsupervised learning methods used to detect phase transitions in the thermodynamics of particle systems to systems that evolve in the time domain. While our proof-of-concept applications are modest, this method shows potential for making progress in a wide array of problems both in physics and outside of the field.

Veillette, John↗

A review of privacy in energy applications

As the distribution system continues to experience an increase in distributed energy resource (DER) and electric vehicle (EV) penetration, so does the need for new solutions that can help grid operators manage and leverage their capabilities. This will undoubtedly lead to new operational schemes and business opportunities that will transform the traditional consumer into a prosumer who will be more actively engaged in grid operations. Although the field is still under active development, many of the potential use cases presented in literature or industry are built upon edge computing, two-way communications, and other innovative computational constructs to attain their goals. However, at their core, many use cases assume a great level of data access to aid with the decision-making process, an assumption that may need to be revised to ensure fair and equitable operational processes are maintained. This may be particularly true as edge resources are predicted to participate in retail-side, many-to-many, or peer-to-peer markets and thus may lead to financial impacts if data access considerations are ignored. The need to revise data access mechanisms can be further justified by the introduction of new participants into the operational process, who do not have the same level of trust, nor the incentives to focus on energy delivery as their primary objective. At the same time, more consumers are becoming aware of their own data, and the potential impacts of its abuse. To help solution developers better understand these risks, this report has been developed to offer an initial introduction to the topic of privacy. This is achieved by 1) Highlighting the need for privacy-aware solutions; 2) Encouraging system designers to be inquisitive about the status quo; 3) Documenting the existing threat space; 4) Presenting and evaluating tools that may be helpful towards enabling better privacy postures; and 5) Making recommendations to encourage the adoption of better practices. From a technical perspective, the report focuses on evaluating two potential techniques by applying them to the Transactive Energy Space. Based on the obtained results, it can be established that differential privacy (DP) methods may have limited applicability when highly correlated, time-series data records need to be protected. However, DP may be a powerful tool when it is used to aggregate and analyze mid-size and large-size data sets in a more traditional statistical environment. The second tool under evaluation is threshold cryptography, which can guarantee complete secrecy (and thus privacy) but requires the establishment of key management procedures and dedicated communication channels for key coordination. Therefore, due to its increased computational overhead, the use of threshold cryptography must be weighted using a cost/benefit analysis on a per-application basis.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Modeling Air Handling Units to Create a Diverse Fault Dataset for FDD Innovation: Lessons Learned and Recommendations

As energy management and information systems (e.g., automated fault detection and diagnostics [AFDD] tools) become more prevalent in the commercial building stock, it is important to determine the effectiveness of these technologies by benchmarking their performance. The authors have been working to develop the largest publicly available dataset of HVAC fault datasets for performance benchmarking applications, covering the most common HVAC systems and designs including chiller plants, rooftop packaged units, dual duct air handling unit and single duct air handling units. This study covers the development, modeling, and validation of a synthetic fault dataset for the air handling unit (AHU), one of the most common HVAC configurations found in the commercial building stock. Despite this being a common system, real-world time series data are scarce and usually do not span a wide range of weather conditions. Due to this limitation, two detailed AHU models, which included the single duct AHU and dual duct AHU developed in the Modelica language and HVACSIM+ were employed to carry out annual simulations of numerous common sensor faults, mechanical faults, and control sequence faults. The fault inclusive data were then validated by comparing fault effects on system performance to expected symptoms. We summarize the nature of each fault and their impacts under different weather and operation conditions. We report some lessons learnt during the efforts of validating the high volumes of the FDD data sets. Finally, we highlight considerations for FDD developers that may want to use this dataset to assess their algorithms’ performance and their improvement over time.

Casillas, Armando↗

Stiff neural ordinary differential equations

Neural Ordinary Differential Equations (ODEs) are a promising approach to learn dynamical models from time-series data in science and engineering applications. This work aims at learning neural ODEs for stiff systems, which are usually raised from chemical kinetic modeling in chemical and biological systems. We first show the challenges of learning neural ODEs in the classical stiff ODE systems of Robertson’s problem and propose techniques to mitigate the challenges associated with scale separations in stiff systems. We then present successful demonstrations in stiff systems of Robertson’s problem and an air pollution problem. The demonstrations show that the usage of deep networks with rectified activations, proper scaling of the network outputs as well as loss functions, and stabilized gradient calculations are the key techniques enabling the learning of stiff neural ODEs. The success of learning stiff neural ODEs opens up possibilities of using neural ODEs in applications with widely varying time-scales, such as chemical dynamics in energy conversion, environmental engineering, and life sciences.

97 MATHEMATICS AND COMPUTING↗

Robust Power System Stability Assessment Against Adversarial Machine Learning-Based Cyberattacks via Online Purification

The increasing complexity associated with renewable generation brings more challenges to power system stability assessment (SA). Data-driven approaches based on machine learning (ML) techniques for stability assessment have received significant research interest and shown their promising performance. However, ML-based models are recognized to be vulnerable to adversarial disturbances, where a slight perturbation to power system measurements could lead to unacceptable errors. To address this issue, this paper develops a novel lightweight mitigation strategy, i.e., robust online stability assessment (ROSA), to enhance the ML-based assessment model against both white-box and the black-box adversarial disturbances (i.e., purification) in the online implementation. The ROSA involves a supervised learning-based module for the primary stability assessment and a self-supervised learning-based module. Further, the two modules are trained jointly with different objective (loss) functions and implemented in sequence. A suitable purification objective and various time-series data augmentation methods are designed for SA applications to tackle adversarial disturbances adaptively. Case studies are performed, and the comparative results have clearly illustrated the competitive, robust accuracy against various adversarial scenarios and verified the effectiveness of the proposed online purification strategy.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Deep learning-enabled probing of irradiation-induced defects in time-series micrographs

Modeling time-series data with convolutional neural networks (CNNs) requires building a model to learn in batches as opposed to training sequentially. Coupling CNNs with in situ or operando techniques opens the possibility of accurately segmenting dynamic reactions and mass transport phenomena to understand how materials behave under the conditions in which they are used. In this article, in situ ion irradiation transmission electron microscopy (TEM) images are used as inputs into the CNN to assess the defect generation rate, defect cluster density, and saturation of defects. We then use the output segmentation maps to correlate with conventional TEM micrographs to assess the model’s ability to detail nanoscale interactions. Next, we discuss the implications of preprocessing and hyperparameters on model variability, accuracy when expanded to other datasets, and the role of regularization when controlling model variance. Ultimately, we eliminate human bias when extrapolating physical metrics, speed up analysis time, decouple reactions that happen at 100 ms intervals, and deploy models that are both accurate and transferable to similar experiments.

Burns, Kory (ORCID:0000000158019909)↗

Progress Report on the Airborne Composition Standard Variable Name and Time Series Working Groups of the 2017 ESDSWG

The role of NASA's Earth Science Data Systems Working Groups (ESDSWG) is to make recommendations relevant to NASA's Earth science data systems from users' experiences and community insight. Each group works independently, focusing on a unique topic. Progress of two of the 2017 Working Groups will be presented. In a single airborne field campaign, there can be several different instruments and techniques that measure the same parameter on one or more aircraft platforms. Many of these same parameters are measured during different airborne campaigns using similar or different instruments and techniques. The Airborne Composition Standard Variable Name Working Group is working to create a list of variable standard names that can be used across all airborne field campaigns in order to assist in the transition to the ICARTT Version 2.0 file format. The overall goal is to enhance the usability of ICARTT files and the search ability of airborne field campaign data. The Time Series Working Group (TSWG) is a continuation of the 2015 and 2016 Time Series Working Groups. In 2015, we started TSWG with the intention of exploring the new OGC (Open Geospatial Consortium) WaterML 2 standards as a means for encoding point-based time series data from NASA satellites. In this working group, we realized that WaterML 2 might not be the best solution for this type of data, for a number of reasons. Our discussion with experts from other agencies, who have worked on similar issues, identified several challenges that we would need to address. As a result, we made the recommendation to study the new TimeseriesML 1.0 standard of OGC as a potential NASA time series standard. The 2016 TSWG examined closely the TimeseriesML 1.0 and, in coordination with the OGC TimeseriesML Standards Working Group, identified certain gaps in TimeseriesML 1.0 that would need to be addressed for the standard to be applicable to NASA time series data. An engineering report was drafted based on the OGC Engineering Report template, describing recommended changes to TimeseriesML 1.0, in the form of use cases. In 2017, we are conducting interoperability experiments to implement the use cases and demonstrate the feasibility and suitability of these modifications for NASA and related user communities. The results will be incorporated into the existing draft engineering report.

OGC (Open Geospatial Consortium) WaterML 2 standar↗

Risk-Aware Framework Development for Disruption Prediction: Alcator C-Mod and DIII-D Survival Analysis

Abstract Survival regression models can achieve longer warning times at similar receiver operating characteristic performance than previously investigated models. Survival regression models are also shown to predict the time until a disruption will occur with lower error than other predictors. Time-to-event predictions from time-series data can be obtained with a survival analysis statistical framework, and there have been many tools developed for this task which we aim to apply to disruption prediction. Using the open-source Auton-Survival package we have implemented disruption predictors with the survival regression models Cox Proportional Hazards, Deep Cox Proportional Hazards, and Deep Survival Machines. To compare with previous work, we also include predictors using a Random Forest binary classifier, and a conditional Kaplan-Meier formalism. We benchmarked the performance of these five predictors using experimental data from the Alcator C-Mod and DIII-D tokamaks by simulating alarms on each individual shot. We find that developing machine-relevant metrics to evaluate models is an important area for future work. While this study finds cases where disruptive conditions are not predicted, there are instances where the desired outcome is produced. Giving the plasma control system the expected time-to-disruption will allow it to determine the optimal actuator response in real time to minimize risk of damage to the device.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Phytoplankton Assemblage Patterns in the Southern Mid-Atlantic Bight

As part of the Wallops Coastal Oceans Observing Laboratory (Wa-COOL) Project, we sampled a time-series transect in the southern Mid-Atlantic Bight (MAB) biweekly. Our 2-year time-series data included physical parameters, nutrient concentrations, and chlorophyll a concentrations. A detailed phytoplankton assemblage structure was examined in the second year. During the 2-year study, chlorophyll a concentration (and ocean color satellite imagery) indicated that phytoplankton blooms occurred in January/February during mixing conditions and in early autumn under stratified conditions. The chlorophyll a concentrations ranged from 0.25 microgram 1(exp -1) to 15.49 microgram 1(exp -1) during the 2-year period. We were able to discriminate approximately 116 different species under phase contrast microscopy. Dominant phytoplankton included Skeletonema costatum, Rhizosolenia spp., and Pseudo-nitzschia pungens. In an attempt to determine phytoplankton species competition/succession within the assemblage, we calculated a Shannon Weaver diversity index for our diatom microscopy data. Diatom diversity was greatest during the winter and minimal during the spring. Diatom diversity was also greater at nearshore stations than at offshore stations. Individual genera appeared patchy, with surface and subsurface patches appearing abruptly and persisting for only 1-2 months at a time. The distribution of individual species differed significantly from bulk variables of the assemblage (chlorophyll a ) and total phytoplankton assemblage (cells), which indicates that phytoplankton species may be limited in growth in ways that differ from those of the total assemblage. Our study demonstrated a highly diverse phytoplankton assemblage throughout the year, with opportunistic species dominating during spring and fall in response to seasonal changes in temperature and nutrients in the southern MAB.

Mid-Atlantic↗

Variational data augmentation for a learning-based granular predictive model of power outages

As the trend in climate change continues, extreme weather events are expected to occur with increasing frequency and severity and pose a significant threat to the electric power infrastructure. Regardless of the efforts a utility puts towards hardening the grid, storm-induced damage to the utility assets such as cables and distributed energy resources (DERs) that are particularly vulnerable to such events is unavoidable. Access to a highly granular, in space and time, outage forecasting tool with long lead times (i.e., days ahead) will enhance the efficiency of service restoration efforts. Here, in this study, we propose to develop and implement a multi-model framework as an operational tool based on a granular and multi-day outage forecasting model using operational numerical weather prediction model forecasts and detailed component outage information. An innovative two-layered recurrent neural network, i.e., a long-short-term-memory (LSTM)-based variational autoencoder (VAE) framework and a sliding window are used to address the uneven distribution of different types of weather events and make better use of the time-series data. Case studies are performed to demonstrate the performance of the new framework.

54 ENVIRONMENTAL SCIENCES↗