Search NASA⌕ Search

SEARCH · Search NASA

Results for “Weatherization training”

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 397 records · Page 22

xesn: Echo state networks powered by Xarray and Dask

Xesn is a Python package that allows scientists to easily design Echo State Networks (ESNs) for forecasting problems. ESNs are a Recurrent Neural Network architecture introduced by Jaeger (2001) that are part of a class of techniques termed Reservoir Computing. One defining characteristic of these techniques is that all internal weights are determined by a handful of global, scalar parameters, thereby avoiding problems during backpropagation and reducing training time significantly. Because this architecture is conceptually simple, many scientists implement ESNs from scratch, leading to questions about computational performance. Xesn offers a straightforward, standard implementation of ESNs that operates efficiently on CPU and GPU hardware. The package leverages optimization tools to automate the parameter selection process, so that scientists can reduce the time finding a good architecture and focus on using ESNs for their domain application. Importantly, the package flexibly handles forecasting tasks for out-of-core, multi-dimensional datasets, eliminating the need to write parallel programming code. Xesn was initially developed to handle the problem of forecasting weather dynamics, and so it integrates naturally with Python packages that have become familiar to weather and climate scientists such as Xarray (Hoyer & Hamman, 2017). However, the software is ultimately general enough to be utilized in other domains where ESNs have been useful, such as in signal processing (Jaeger & Haas, 2004).

97 MATHEMATICS AND COMPUTING↗

Real-Time Teleguidance of a Non-Surgeon Crew Medical Officer Performing Orthopedic Surgery at the Amundsen-Scott South Pole Station During Winter-Over

The Amundsen-Scott South Pole Research station located at the geographic South Pole, is the most isolated, permanently inhabited human outpost on Earth. Medical care is provided to station personnel by a non-surgeon crew medical officer (CMO). During the winter-over period from February to October, the station is isolated, with no incoming or outgoing flights due to severe weather conditions. In late June, four months after the station had closed for the austral winter, a 31 year old meteorologist suffered a complete rupture of his patellar tendon while sliding done an embankment. An evacuation was deemed to be too risky to aircrews due to the extreme cold and darkness. A panel of physicians from Massachusetts General Hospital, Johns Hopkins University and the University of Texas Medical Branch were able to assess the patient remotely via telemedicine and agreed that surgery was the only means to restore mobility and prevent long term disability. The lack of a surgical facility and a trained surgical team were overcome by conversion of the clinic treatment area, and intensive preparation of medical laypersons as surgical assistants. The non-surgeon CMO and CMO assistant at South Pole, were guided through the administration of spinal anesthetic, and the two-hour operative repair by medical consultants at Massachusetts General Hospital. Real-time video of the operative field, directions from the remote consultants and audio communication were provided by videoconferencing equipment, operative cameras, and high bandwidth satellite communications. In real-time, opening incision/exposure, tendon relocation, hemostatsis, and operative closure by the CMO was closely monitored and guided and by the remote consultants. The patient s subsequent physical rehabilitation over the ensuing months of isolation was also monitored remotely via telemedicine. This was the first time in South Pole s history that remote teleguidance had been used for surgery and represents a model for real-time guidance of CMO s working at remote duty stations.

Otto, Christian↗

Assessment of Quantum ML Applicability for Climate Actions: Comparison of the Variational Quantum Classifier and the Quantum Support Vector Classifier with Classical ML Models

Climate change refers to significant and long-term alterations in the Earth’s climate patterns, typically resulting from human activities that increase greenhouse gas emissions. Addressing climate change is not merely an option but a necessity, demanding creative solutions and efforts from individuals, researchers, communities, and governments. Despite the capabilities of machine learning (ML) with data-driven solutions promising to combat climate change-related problems, they face challenges stemming from traditional computational methods and prolonged training times, impeding their practical utility. Recent strides in quantum computing have permeated diverse domains, spanning from manufacturing engineering and pharmaceutical discovery to the latest frontier of detecting climate anomalies. With the potential to substantially reduce time and computational complexity, quantum computing shows promise in addressing climate change impacts. Its distinctive features will enable the concurrent exploration of expansive solution spaces, making it well-suited for analyzing extensive climate datasets, simulating intricate climate models, optimizing resource allocation, and discerning patterns in climate data for mitigation and adaptation endeavors. This study explores the potential of using Quantum machine learning (QML) techniques on climate and weather data obtained from NASA Giovannis. We used two QML algorithms, the Quantum Support Vector Classifier (QSVC) and the Variational Quantum Classifier (VQC) models, using the IBM Qiskit ML 0.7.2 ecosystem. We used an actual 127-Qubit IBM Quantum Computer (IBM 127-qubit Eagle) in this study. The methodology and results sections describe the experiences gained from applying and evaluating quantum ML results on climate and weather data obtained from NASA satellites as a novel practical application of quantum computing.

Earth Observational Data↗

Altimetry for the Future: Building on 25 Years of Progress

In 2018 we celebrated 25 years of development of radar altimetry, and the progress achieved by this methodology in the fields of global and coastal oceanography, hydrology, geodesy and cryospheric sciences. Many symbolic major events have celebrated these developments, e.g., in Venice, Italy, the 15th (2006) and 20th (2012) years of progress and more recently, in 2018, in Ponta Delgada, Portugal, 25 Years of Progress in Radar Altimetry. On this latter occasion it was decided to collect contributions of scientists, engineers and managers involved in the worldwide altimetry community to depict the state of altimetry and propose recommendations for the altimetry of the future. This paper summarizes contributions and recommendations that were collected and provides guidance for future mission design, research activities, and sustainable operational radar altimetry data exploitation. Recommendations provided are fundamental for optimizing further scientific and operational advances of oceanographic observations by altimetry, including requirements for spatial and temporal resolution of altimetric measurements, their accuracy and continuity. There are also new challenges and new openings mentioned in the paper that are particularly crucial for observations at higher latitudes, for coastal oceanography, for cryospheric studies and for hydrology. The paper starts with a general introduction followed by a section on Earth System Science including Ocean Dynamics, Sea Level, the Coastal Ocean, Hydrology, the Cryosphere and Polar Oceans and the ‘‘Green” Ocean, extending the frontier from biogeochemistry to marine ecology. Applications are described in a subsequent section, which covers Operational Oceanography, Weather, Hurricane Wave and Wind Forecasting, Climate projection. Instruments’ development and satellite missions’ evolutions are described in a fourth section. A fifth section covers the key observations that altimeters provide and their potential complements, from other Earth observation measurements to in situ data. Section 6 identifies the data and methods and provides some accuracy and resolution requirements for the wet tropospheric correction, the orbit and other geodetic requirements, the Mean Sea Surface, Geoid and Mean Dynamic Topography, Calibration and Validation, data accuracy, data access and handling (including the DUACS system). Section 7 brings a transversal view on scales, integration, artificial intelligence, and capacity building (education and training). Section 8 reviews the programmatic issues followed by a conclusion.

Saleh Abdalla↗

Assessment Overview: Regional Impacts of Using the Day Cloud Phase Distinction RGB for Lightning IDSS

The new generation of GOES satellites has presented a wealth of novel observations and data products to the National Weather Service forecasters throughout the country. While high quality, low latency data is always welcome in operational environments, it is valuable to understand how new information is interpreted, and how it complements and supplements existing datasets. To that end, NASA SPoRT, in partnership with the NWS, hosted a formal assessment of the use of the Day Cloud Phase Distinction RGB to anticipate lightning initiation as indicated by either GLM, NLDN, or both, using an approach outlined by Elsenheimer and Gravelle (2019). As is typical for SPoRT assessments, we focused on a targeted time period (about 6 weeks in June and July 2021) and provided a brief training module to help forecasters become familiar with the application we were targeting and the objectives of our assessment. Unlike typical SPoRT assessments, we focused on a broad geographic region, coordinating with NWS Regional offices to garner participation from several different WFOs throughout the country. Given the field of view and observation angles of GOES-East and GOES-West, which anecdotally impacts GLM lightning observations in particular, having a geographically diverse participant list helps us interpret how these viewing differences may impact interpretation of the RGB and its value in anticipating lightning. Not unexpectedly, we found that the participating forecasters expressed that they could readily interpret the RGB to recognize glaciation and, in the absence of lightning data, use the RGB to anticipate that lightning was possible, for the purpose of impact-based decision support services (IDSS). Lead times from observations of glaciation to first flash were 19 minutes on average, and -10C isothermal reflectivities ranged from 34.5 to 52.5 dBZ. Herein, we present the detailed results of the formal assessment period and provide next steps and recommendations based on forecaster feedback for using the DCPD RGB in the context of anticipating lightning initiation and for IDSS, including regional differences in observations.

satellite remote sensing↗

Estimating Subhourly Inverter Clipping Loss From Satellite-Derived Irradiance Data: Preprint

Photovoltaic system production simulations are conventionally run using hourly weather datasets. Hourly simulations are sufficiently accurate to predict the majority of long-term system behavior but cannot resolve high-frequency effects like inverter clipping caused by short-duration irradiance variability. Direct modeling of this subhourly clipping error is only possible for the few locations with high-resolution irradiance datasets. This paper describes a method of predicting the magnitude of this error using a machine learning model and 30-minute satellite irradiance data. The method predicts a correction for each 30-minute interval with the potential to roll up into 60-minute corrections to match an hourly energy model. The model is trained and validated at locations where the error can be directly simulated from 1-minute ground data. The validation shows low bias at most ground station locations. The model is also applied to gridded satellite irradiance to produce a heatmap of the estimated clipping error across the United States. Finally, the relative importance of each predictor satellite variable is retrieved from the model and discussed.

41 EE - Solar Energy Technologies Office (EE-4S)↗

Enhanced Soundings for Local Coupling Studies: 2015 ARM Climate Research Facility Field Campaign

Matching observed diurnal cycles is a fundamental yet extremely complex test for models. High temporal resolution measurements of surface turbulent heat fluxes and boundary layer properties are required to evaluate the daytime evolution of the boundary layer and its sensitivity to land-atmosphere coupling. To address this need, (12) one-day intensive observing periods (IOP) with enhanced radiosonding will be carried out at the ARM Southern Great Plains (SGP) Central Facility (CF) during summer 2015. Each IOP will comprise a single launch to correspond with the nighttime overpass of the A-Train of satellites (~0830 UTC) and hourly launches during daytime beginning from 1130 UTC and ending at 2130 UTC. At 3-hourly intervals (i.e., 1140 UTC, 1440 UTC, 1740 UTC, and 2040 UTC) a duplicate second radiosonde will be launched 10 minutes subsequent to launch of the on-hour radiosonde for the purpose of assessing horizontal atmospheric variability. In summary, each IOP will have a 14-sounding supplement to the 6-hourly operational sounding schedule at the ARM-SGP CF. The IOP days will be decided before sunset on the preceding day, according to the judgment of the PI’s and taking into consideration daily weather forecasts and the operability of complimentary ARM-SGP CF instrumentation. An overarching goal of the project is to address how ARM could better observe land-atmosphere coupling to support the evaluation and refinement of coupled weather and climate models.

Ferguson, CR↗

Communication Demand in the National Airspace - A Federated Learning Approach

Within the national airspace system (NAS), efficient use of spectrum remains a challenge; as UAS and UAM missions evolve, the amount of mission-critical aircraft communications are expected to significantly grow. To accommodate the increased demand, NASA Glenn Research Center is investigating artificial intelligence approaches that could dynamically allocate spectrum; however, these solutions are driven by communication and aviation data items, many of which are not directly available. One such cornerstone data item is communication demand, parameterizing the needs within a sector in terms that may directly inform spectrum allocation, such as channel access duration, bandwidth, and modulation type. This paper considers the complexity of predicting communication demand as a function of NAS behaviors. Unlike prior prediction work in communications, this research must consider how the NAS may be impacted by external factors - such as convective weather and closures - rather than estimating demand from time-series forecasting alone. Much of this research considers a federated learning design to predict communication events in terms of the type of event occurring (sector coordination, conflict resolution, etc.). To do so, an investigation of products from Sherlock Data Warehouse is conducted, identifying the trends, sufficiency, and correlations of each product to identified events. Additionally, a preliminary discussion for inferring associations between these event types and their communication parameters (duration, bandwidth, modulation) is presented. By utilizing federated learning, imbalances in the types of events and data present throughout the NAS can inform local models without impairing global training. Furthermore, the two-stage approach proposed allows for robust and speculative communication modelling, as communication techniques continue to evolve. As a result, this model enables a generalized approach to understanding NAS communications which is able to inform long-term changes to aviation spectrum management.

Nathan Schimpf↗

A Federated Learning Approach to Predicting Communication Demand in the National Airspace

Within the national airspace system (NAS), efficient use of spectrum remains a challenge; as UAS and UAM missions evolve, the amount of mission-critical aircraft communications are expected to significantly grow. To accommodate the increased demand, NASA Glenn Research Center is investigating artificial intelligence approaches that could dynamically allocate spectrum; however, these solutions are driven by communication and aviation data items, many of which are not directly available. One such cornerstone data item is communication demand, parameterizing the needs within a sector in terms that may directly inform spectrum allocation, such as channel access duration, bandwidth, and modulation type. This paper considers the complexity of predicting communication demand as a function of NAS behaviors. Unlike prior prediction work in communications, this research must consider how the NAS may be impacted by external factors - such as convective weather and closures - rather than estimating demand from time-series forecasting alone. Much of this research considers a federated learning design to predict communication events in terms of the type of event occurring (sector coordination, conflict resolution, etc). To do so, an investigation of products from Sherlock Data Warehouse is conducted, identifying the trends, sufficiency, and correlations of each product to identified events. Additionally, a preliminary discussion for inferring associations between these event types and their communication parameters (duration, bandwidth, modulation) is presented. By utilizing federated learning, imbalances in the types of events and data present throughout the NAS can inform local models without impairing global training. Furthermore, the two-stage approach proposed allows for robust and speculative communication modelling, as communication techniques continue to evolve. As a result, this model enables a generalized approach to understanding NAS communications which is able to inform long-term changes to aviation spectrum management.

Nathan Schimpf↗

Online transfer learning strategy for enhancing the scalability and deployment of deep reinforcement learning control in smart buildings

In recent years, advanced control strategies based on Deep Reinforcement Learning (DRL) proved to be effective in optimizing the management of integrated energy systems in buildings, reducing energy costs and improving indoor comfort conditions when compared to traditional reactive controllers. However, the scalability and implementation of DRL controllers are still limited since they require a considerable amount of time before converging to a near-optimal solution. This issue is currently addressed in literature through the offline pre-training of the DRL agent. However this solution results in two main critical issues: (1) the need to develop a building surrogate model to perform the training task, and (2) the need to perform a fine-tuning process over several training episodes to obtain a near-optimal control policy. In this context, this paper introduces an Online Transfer Learning (OTL) strategy that exploits two knowledge-sharing techniques, weight-initialization and imitation learning, to transfer a DRL control policy from a source office building to various target buildings in a simulation environment coupling EnergyPlus and Python. A DRL controller based on discrete Soft Actor–Critic (SAC) is trained on the source building to manage the operation of a cooling system consisting of a chiller and a thermal storage. Several target buildings are defined to benchmark the performance of the OTL strategy with that of a Rule-Based Controller (RBC) and two DRL-based control strategies, deployed in offline and online fashion. The strategy adopted for OTL emulates the real world implementation with a simulation process by implementing the transferred DRL agent for a single episode in the target buildings. Target buildings have the same geometrical features and are served by the same energy system as the source building, but differ in terms of weather conditions, electricity price schedules, occupancy patterns, and building envelope efficiency levels. The results show that the OTL strategy can reduce the cumulated sum of temperature violations on average by 50% and 80% respectively when compared to RBC and online DRL while enhancing the energy system operation with electricity cost savings ranging between 20% and 40%. Furthermore, the OTL agent performs slightly worse than the offline DRL controller but it does not require any modeling effort and can be implemented directly on target buildings emulating a real-world implementation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

A Privacy-Aware Federated Learning Framework for Distributed Energy Resource Analytics in Constrained Environments

To be resilient against extreme weather events, the rural communities in Puerto Rico are leveraging distributed energy resources (DER). However, computing frameworks sup-porting the grid in critical decision-making are still largely centralized. Sensitive consumer data are transmitted over the Internet or cellular networks to a secondary or tertiary node. It guarantees better situational awareness at the cost of a wider attack surface, jeopardizing user privacy, as more DER come online. Cloud, Edge, and Fog computing all require data aggregation at some level. This paper introduces a privacy-aware federated learning framework that leverages the Fog model by pushing analytics all the way to the DER and load assets. These local models train on individual asset data and transmit only learned parameters (such as weights) over secure communications to a global decision-maker. By abstracting personally identifiable consumer data without impacting decision optimality, this framework better aligns with distributed power generation paradigm.

Sundararajan, Aditya↗

Atmospheric teleconnection processes linking winter air stagnation and haze extremes in China with regional Arctic sea ice decline

Abstract. Recent studies suggested significant impacts of boreal cryosphere changes on wintertime air stagnation and haze pollution extremes in China. However, the underlying mechanisms of such a teleconnection relationship remains unclear. Here we use the Whole Atmosphere Community Climate Model (WACCM) to investigate dynamic processes leading to atmospheric circulation and air stagnation responses to Arctic sea ice changes. We conduct four climate sensitivity experiments by perturbing sea ice concentrations (SIC) and corresponding sea surface temperature (SST) in autumn and early winter over the whole Arctic and three subregions in the climate model. The results indicate distinct responses in circulation patterns and regional ventilation to the region-specific Arctic changes, with the largest increase of both the probability (by 132 %) and the intensity (by 30 %) of monthly air stagnation extremes being found in the experiment driven by SIC and SST changes over the Pacific sector of the Arctic (the East Siberian and Chukchi seas). The increased air stagnation extremes are mainly driven by an amplified planetary-scale atmospheric teleconnection pattern that resembles the negative phase of the Eurasian (EU) pattern. Dynamical diagnostics suggest that convergence of transient eddy forcing in the vicinity of Scandinavia in winter is largely responsible for the amplification of the teleconnection pattern. Transient eddy vorticity fluxes dominate the transient eddy forcing and produce a barotropic anticyclonic anomaly near Scandinavia and wave train propagation across Eurasia to the downstream regions in East Asia. The piecewise potential vorticity inversion analysis reveals that this long-range atmospheric teleconnection of Arctic origin takes place primarily via the middle and upper troposphere. The anomalous ridge over East Asia in the middle and upper troposphere worsens regional ventilation conditions by weakening monsoon northwesterlies and enhancing temperature inversions near the surface, leading to more and stronger air stagnation and pollution extremes over eastern China in winter. Ensemble projections based on state-of-the-art climate models in the Coupled Model Intercomparison Project Phase 6 (CMIP6) corroborate this teleconnection relationship between high-latitude environmental changes and midlatitude weather extremes, though the tendency and magnitude vary considerably among each participating model.

54 ENVIRONMENTAL SCIENCES↗

NeuralMie (v1.0): an aerosol optics emulator

The direct interactions of atmospheric aerosols with radiation significantly impact the Earth's climate and weather and are important to represent accurately in simulations of the atmosphere. This work introduces two contributions to enable a more accurate representation of aerosol optics in atmosphere models: (1) NeuralMie, a neural network Mie scattering emulator that can directly compute the bulk optical properties of a diverse range of aerosol populations and is appropriate for use in atmosphere simulations where aerosol optical properties are parameterized, and (2) TAMie, a fast Python-based Mie scattering code based on the Toon and Ackerman (1981) Mie scattering algorithm that can represent both homogeneous and coated particles. TAMie achieves speed and accuracy comparable to established Fortran Mie codes and is used to produce training data for NeuralMie. NeuralMie is highly flexible and can be used for a wide range of particle types, wavelengths, and mixing assumptions. It can represent core-shell scattering and, by directly estimating bulk optical properties, is more efficient than existing Mie code and Mie code emulators while incurring negligible error compared to existing aerosol optics parameterization schemes (0.08 % mean absolute percentage error).

54 ENVIRONMENTAL SCIENCES↗

Advanced Intelligent System Application to Load Forecasting and Control for Hybrid Electric Bus

The primary motivation for this research emanates from providing a decision support system to the electric bus operators in the municipal and urban localities which will guide the operators to maintain an optimal compromise among the noise level, pollution level, fuel usage etc. This study is backed up by our previous studies on study of battery characteristics, permanent magnet DC motor studies and electric traction motor size studies completed in the first year. The operator of the Hybrid Electric Car must determine optimal power management schedule to meet a given load demand for different weather and road conditions. The decision support system for the bus operator comprises three sub-tasks viz. forecast of the electrical load for the route to be traversed divided into specified time periods (few minutes); deriving an optimal 'plan' or 'preschedule' based on the load forecast for the entire time-horizon (i.e., for all time periods) ahead of time; and finally employing corrective control action to monitor and modify the optimal plan in real-time. A fully connected artificial neural network (ANN) model is developed for forecasting the kW requirement for hybrid electric bus based on inputs like climatic conditions, passenger load, road inclination, etc. The ANN model is trained using back-propagation algorithm employing improved optimization techniques like projected Lagrangian technique. The pre-scheduler is based on a Goal-Programming (GP) optimization model with noise, pollution and fuel usage as the three objectives. GP has the capability of analyzing the trade-off among the conflicting objectives and arriving at the optimal activity levels, e.g., throttle settings. The corrective control action or the third sub-task is formulated as an optimal control model with inputs from the real-time data base as well as the GP model to minimize the error (or deviation) from the optimal plan. These three activities linked with the ANN forecaster proving the output to the GP model which in turn produces the pre-schedule of the optimal control model. Some preliminary results based on a hypothetical test case will be presented for the load forecasting module. The computer codes for the three modules will be made available fe adoption by bus operating agencies. Sample results will be provided using these models. The software will be a useful tool for supporting the control systems for the Electric Bus project of NASA.

Momoh, James↗

SatNet: A Benchmark for Satellite Scheduling Optimization

Satellites provide essential services such as networking and weather tracking, and the number of near-earth and deep space satellites are expected to grow rapidly in the coming years. Communications with terrestrial ground stations is one of the critical functionalities of any space mission. Satellite scheduling is a problem that has been scientifically investigated since the 1970s. A central aspect of this problem is the need to consider resource contention and satellite visibility constraints as they require line of sight. Due to the combinatorial nature of the problem, prior solutions such as linear programs and evolutionary algorithms require extensive compute capabilities to output a feasible schedule for each scenario. Machine learning based scheduling can provide an alternative solution by training a model with historical data and generating a schedule quickly with model inference. We present SatNet, a benchmark for satellite scheduling optimization based on historical data from the NASA Deep Space Network. We propose formulation of the satellite scheduling problem as a Markov Decision Process and use reinforcement learning (RL) policies to generate schedules. The nature of constraints imposed by SatNet differ from other combinatorial optimization problems such as vehicle routing studied in prior literature. Our initial results indicate that RL is an alternative optimization approach that can generate candidate solutions of comparable quality to existing state-of-the-practice results. However, we also find that RL policies overfit to the training dataset and do not generalize well to new data, thereby necessitating continued research on reusable and generalizable agents.

Wilson, Brian↗

SpaceNet 8 - The Detection of Flooded Roads and Buildings

The frequency and intensity of natural disasters (i.e. wildfires, storms, floods) has increased over recent decades. Extreme weather can often be linked to climate change, and human population expansion and urbanization have led to a growing risk. In particular floods due to large amounts of rainfall are of rising severity and are causing loss of life, destruction of buildings and infrastructure, erosion of arable land, and environmental hazards around the world. Expanding urbanization along rivers and creeks often includes opening flood plains for building construction and river straightening and dredging speeding up the flow of water. In a flood event, rapid response is essential which requires knowledge which buildings are susceptible to flooding and which roads are still accessible. To this aim, SpaceNet 8 is the first remote sensing machine learning training dataset combining building footprint detection, road network extraction, and flood detection covering 850km 2, including 32k buildings and 1,300km roads of which 13% and 15% are flooded, respectively.

Arndt, Jacob↗

Reduced-Gravity Simulator for Field Environments - Drone Augmented System

Astronauts need to go through extensive training on Earth before heading into space, and the more accurate the analogue, the better prepared they will be. NASA currently trains astronauts in simulated lower gravity environments in either the Neutral Buoyancy Laboratory or the Active Response Gravity Offload System (ARGOS), but neither of these systems allow for testing outdoors in field environments. NASA, in partnership with Aquiline Drones, mechanical engineering, biomedical engineering, and the Krenicki Arts and Engineering Institute, are working to develop a gravity offload device compatible with testing in field environments such as Desert Research and Technology Studies (Desert RATS). This report covers the technical requirements of the gravity offload device, the design choices and justifications, options for scalability, current progress, and future goals. Two separate teams were formed to respond to NASA’s solicitation, and this report covers the work of the weather balloon and drone side of the project. Using a weather balloon to passively offload the user’s weight and a drone to actively respond to changes using a series of sensors, this system should create a consistent and well balanced offloading force. The weather balloon and drone system is connected to a harness system to comfortably lift the user, offload at their center of gravity, and provide attachment points for some of the necessary sensors. Final testing proved the comfort of the harness, mobility of the user, and passive stability of the balloon offload to exceed all expectations of the team. The current system is a scaled model of what NASA would be able to implement with a maximum offload of 25 pounds. The ability to scale up to NASA’s desired value of 100 pounds is easily achievable with linear scaling of the current model. All systems should work similarly on scaled versions, but minor tweaking and enhanced controls would be needed for optimal performance.

Jason Lee↗

Estimating Subhourly Inverter Clipping Loss From Satellite-Derived Irradiance Data

Photovoltaic system production simulations are conventionally run using hourly weather datasets. Hourly simulations are sufficiently accurate to predict the majority of long-term system behavior but cannot resolve high-frequency effects like inverter clipping caused by short-duration irradiance variability. Direct modeling of this subhourly clipping error is only possible for the few locations with high-resolution irradiance datasets. This paper describes a method of predicting the magnitude of this error using a machine learning regressor ensemble model, comprised of a random forest and an XGBoost model, and 30-minute satellite irradiance data. The method predicts a correction for each 30-minute interval with the potential to roll up into 60-minute corrections to match an hourly energy model. The model is trained and validated at locations where the error can be directly simulated from 1-minute ground data. The validation shows low bias at most ground station locations. The model is also applied to gridded satellite irradiance to produce a heatmap of the estimated clipping error across the United States. Finally, the relative importance of each predictor satellite variable is retrieved from the model and discussed.

41 EE - Solar Energy Technologies Office (EE-4S)↗