Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data quality analysis”

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 631 records · Page 35
All sourcesNASA NTRSDOE OSTI
NASA NTRS2004

Precipitation Analysis at Fine Time Scales using TRMM and Other Satellites: Real-time and Research Products and Applications

Quasi-global precipitation analyses at fine time scales (3-hr) are described. TRMM observations (radar and passive microwave) are used to calibrate polar-orbit microwave observations from SSM/I (and other satellites instruments, including AMSR and AMSU) and geosynchronous IR observations. The individual data sets are then merged using a priority order based on quality to form the TRMM Multi-satellite Precipitation Analysis (MPA). Raingauge information is used to help constrain the satellite-based estimates over land. The TRMM standard research product (Version 6 3B-42 of the TRMM products) will be available for the entire TRMM period (January 1998-present) by the end of 2004. The real-time version of this merged product has been produced over the past two years and is available on the U.S. TRMM web site (trmm.gsfc.nasa.gov) at 0.25" latitude-longitude resolution over the latitude range from 5O0N-50"S. Validation of daily totals indicates good results, with limitations noted in mid-latitude winter over land and regions of shallow, orographic precipitation. Various applications of these estimates are described, includmg: 1) detecting potential floods in near real-time; 2) analyzing Indian Ocean precipitation variations related to the initiation of El Nino; 3) determining characteristics of the African monsoon; and 4) analysis of diurnal variations.

Adler, Robert↗
NASA NTRS2010

The Proposed Mars Astrobiology Explorer - Cacher [MAX-C] Rover: First Step in a Potential Sample Return Campaign

Sample return from Mars has been advocated by numerous scientific advisory panels for over 30 years, most prominently beginning with the National Research Council s [1] strategy for the exploration of the inner solar system, and most recently by the Mars Exploration Program Analysis Group (MEPAG s) Next Decade Science Analysis Group [2]. Analysis of samples here on Earth would have enormous advantages over in situ analyses in producing the data quality needed to address many of the complex scientific questions the community has posed about Mars. Instead of a small, predetermined set of analytical techniques, state of the art preparative and instrumental resources of the entire scientific community could be applied to the samples. The analytical emphasis could shift as the meaning of each result becomes better appreciated. These arguments apply both to igneous rocks and to layered sedimentary materials, either of which could contain water and other volatile constituents. In 2009 MEPAG formed the Mid-Range Rover Science Analysis Group (MRR-SAG) to formulate a mission concept that would address two general objectives: (1) conduct high-priority in situ science and (2) make concrete steps towards the potential return of samples to Earth. This analysis resulted in a mission concept named the Mars Astrobiology Explorer-Cacher (MAX-C), which was envisioned for launch in the 2018 opportunity. After extensive discussion, this group concluded that by far the most definitive contribution to sample return by this mission would be to collect and cache, in an accessible location, a suite of compelling samples that could potentially be recovered and returned by a subsequent mission. This would have the effect of separating two of the essential functions of MSR, the acquisition of the sample collection and its delivery to martian orbit, into two missions.

Allen, Carlton C.↗
NASA NTRS2002

Evaluation of GEOSAT Follow-On Precise Orbit Ephemeris

The GEOSAT Follow-On spacecraft (GFO-l), launched in 1998, began continuous radar altimeter coverage of the oceans in 2000, and after an extensive series of calibration campaigns in 1999 and 2000 was finally accepted by the US Navy on November 29, 2000. By providing high quality altimeter data, GFO can supplement TOPEX/POSEIDON(T/P), ERS-1, and ERS-2 data, providing a different synoptic sampling of the oceans with its 17-day ground track repeat cycle. Altimeter analysis suggests GFO is capable of POSEIDON class altimetry, with orbit error the largest contributor to the GFO altimeter error budget. Satellite laser ranging (SLR), especially in combination with altimeter crossover data, offer the only means of determining high-quality precise orbits. SLR tracking is also augmented by Doppler (Tranet style) beacons. These data were used to tune the gravity field and macromodel (3D representation of the spacecraft geometry and surface properties). Beginning with January 2000 GSFC has produced a 2.5 year span of the Precision Orbit Ephemeris (POE) intended for use on the GFO GDR and for the Pathfinder Project. Using GFO GDR and IGDR altimeter data from NOAA, a two year span of POE orbits are evaluated with respect to the TOPEX orbits by adjusting the GFO data into the TOPEX frame with the help of TOPEX/GFO altimeter crossovers. A 2.5 year span of orbits are also evaluated using tracking data residual analysis, GFO altimeter crossover residual analysis, and direct orbit comparison. In an effort to improve the current POE orbit several ideas are considered, including using recent gravity models which include CHAMP data, the reduced-dynamic approach, and including TOPEX/GFO altimeter crossover data in the orbit solution.

Lemoine, Frank G.↗
NASA NTRS2014

Incorporating Handling Qualities Analysis into Rotorcraft Conceptual Design

This paper describes the initial development of a framework to incorporate handling qualities analyses into a rotorcraft conceptual design process. In particular, the paper describes how rotorcraft conceptual design level data can be used to generate flight dynamics models for handling qualities analyses. Also, methods are described that couple a basic stability augmentation system to the rotorcraft flight dynamics model to extend analysis to beyond that of the bare airframe. A methodology for calculating the handling qualities characteristics of the flight dynamics models and for comparing the results to ADS-33E criteria is described. Preliminary results from the application of the handling qualities analysis for variations in key rotorcraft design parameters of main rotor radius, blade chord, hub stiffness and flap moment of inertia are shown. Varying relationships, with counteracting trends for different handling qualities criteria and different flight speeds are exhibited, with the action of the control system playing a complex part in the outcomes. Overall, the paper demonstrates how a broad array of technical issues across flight dynamics stability and control, simulation and modeling, control law design and handling qualities testing and evaluation had to be confronted to implement even a moderately comprehensive handling qualities analysis of relatively low fidelity models. A key outstanding issue is to how to 'close the loop' with an overall design process, and options for the exploration of how to feedback handling qualities results to a conceptual design process are proposed for future work.

Analysis into Rotorcraft↗
NASA NTRS2005

NASA Technical Interchange Meeting (TIM): Advanced Technology Lifecycle Analysis System (ATLAS) Technology Tool Box

The objective of this Technical Interchange Meeting was to increase the quantity and quality of technical, cost, and programmatic data used to model the impact of investing in different technologies. The focus of this meeting was the Technology Tool Box (TTB), a database of performance, operations, and programmatic parameters provided by technologists and used by systems engineers. The TTB is the data repository used by a system of models known as the Advanced Technology Lifecycle Analysis System (ATLAS). This report describes the result of the November meeting, and also provides background information on ATLAS and the TTB.

ONeil, D. A.↗
NASA NTRS2019

Air Pollution Inputs to the Mojave Desert By Fusing Surface Mobile and Airborne in Situ and Airborne and Satellite Remote Sensing: A Case Study of Interbasin Transport With Numerical Model Validation

Deserts are fragile and highly sensitive ecosystems that increasingly are affected by upwind urban areas and industrial activities. The Los Angeles Basin (LAB) contributes to poor air quality in downwind deserts including the Mojave Desert. Few studies have investigated potential air pollution inputs to the Mojave, whose fragile ecosystem includes endangered plant and animal species. Data were collected on 19 August 2015 by a mobile air quality laboratory, AMOG (AutoMObile trace Gas) Surveyor, that observed inputs can arise from the LAB as well as the San Joaquin Valley (SJV), California. The campaign used a strong methane (CH4) plume as a tracer for the downwind fate of emissions from Bakersfield area petroleum production and also measured ozone (O3). Additional in situ concurrent airborne GHG and O3 data were collected by AJAX - Alpha Jet Atmospheric eXperiment. Both AMOG and AJAX measure winds. Mojave Desert air quality was very poor (visibility ~4 km). Based on the winds, an additional source was inferred beyond the LAB and SJV Basins. Numerical transport modeling and analysis of aerosol lidar data collected the same day by the Cloud Profiling LiDAR onboard the Earth Research-2 stratospheric airplane demonstrated that fires in Northern California were responsible, with prevailing winds transporting air southwards along the eastern Sierra Nevada Range (Bishop Valley) to the Mojave. Whereas the southern and eastern Mojave are impacted by SJV and LAB outflow, the north Mojave generally avoids these inputs. This study shows it can be affected by even distant wildfires, which likely will increase in occurrence and intensity from climate change. Thus, regulatory efforts to reduce air quality impacts on the endangered Mojave ecosystem must include wildfires and also account for the significant differences between different regions of the Mojave. Currently, there is a paucity of studies, highlighting the critical need for field research.

Ira Leifer↗
NASA NTRS2022

Retrieving Particulate Matter Concentrations over the Contiguous United States Using CALIOP Observations

Using twelve years (2007-2018) of NASA Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) near-surface 532 nm aerosol extinction retrievals, multi-year mean and trends of particulate matter (PM) concentrations are derived over the contiguous United States (CONUS). Different from past studies that use column integrated aerosol optical thickness, here only near-surface CALIOP aerosol extinction is used for deriving near-surface PM with aerodynamic diameters less than 2.5 µm (PM2.5) concentrations using an innovative, bulk-mass-modeling-based method. Compared against ground based PM2.5 measurements from the U.S. Environmental Protection Agency (EPA), an encouraging relationship between CALIOP-derived PM2.5 and EPA-observed PM2.5 (Deming slope = 0.89; RMSE = 3.42 µg/m3; mean bias = -1.00 µg/m3) is found using combined daytime/nighttime CALIOP data. Also, comparable trends in PM2.5 concentrations from the EPA and daytime and nighttime CALIOP data are found for most of the eastern CONUS and imply that air quality is generally improving over this region for the study period. Over the western CONUS, a seasonal analysis reveals that PM2.5 trends are positive during the more active wildfire season (June through November) but negative for other months. This study suggests that lidar data show promise in their use for obtaining PM2.5 estimates and provides motivation to further explore aerosol extinction-based PM concentration retrievals in anticipation of future space-based lidar missions.

CALIOP↗
NASA NTRSDate not supplied

The INSTEP Monitoring Network: Merging High-and-Low Cost Measurements to Characterize California Wildfires

Despite challenges with data quality and scope, low-cost sensor networks have skyrocketed in popularity over the last 15 years, making air quality data available on refined spatial scales. More recently, studies have leveraged both high and low-quality instruments to create stronger “hybrid” models, with most studies focusing on particulate matter. Low-cost measurements typically represent ground-level emissions only, providing context for human health issues from climate change-driven events such as wildfires. Since low-cost sensors’ capabilities are localized, daily events and microclimates tend to dominate the data rather than larger regional or atmospheric trends. Likewise, their low cost explains their high uncertainty. In contrast, some regulatory-grade instruments produce column measurements as well, providing reliable information on a broader scope. To bridge this gap while expanding into gas-phase measurements, we deployed 12 air quality sensor packages in California, USA during the 2022 wildfire season. These INSTEP (Inexpensive Network Sensor Technology Exploring Pollution) monitors measure carbon monoxide (CO), carbon dioxide (CO2), ozone (O3), nitrogen dioxide (NO2), and several hydrocarbons including methane (CH4) and formaldehyde (HCHO). Half of the monitors were co-located with remote sensing spectrometers: NASA Pandora and Total Column Carbon Observing Network (TCCON). The overlap in pollutants includes NO2, O3, and HCHO between the INSTEP monitors and the Pandora column measurements. TCCON covers column CO, CO2, and CH4, rounding out our comparison. Most of the monitors were distributed throughout the San Francisco Bay area, and an additional three were located within 100 km of Los Angeles. The sites ranged in geographic and population characteristics, including desert, mountainous, coastal, and urban locations. Since varying environmental conditions such as temperature and pressure are known to challenge sensor performance, we will apply newer sensor “calibration” techniques meant to combat this. We will normalize our sensor signals by z-scoring them prior to applying a single calibration model in the form of multivariate linear regression or an artificial neural network. While this technique has been validated for the hydrocarbon and ozone sensor types (metal oxide), it has not yet been tested on electrochemical and non-dispersive infrared sensors, which are also used in the INSTEP monitors. This will serve as a test to see if this normalization technique – or another – is most effective in accounting for environmental differences among sensors. Related data analysis efforts have found success with a variety of geospatial analysis techniques, including weighted network models in which high-quality instruments are given higher weights than their low-cost counterparts. Our preliminary analysis will focus on kriging, which uses a Gaussian algorithm to assign weights, providing estimated pollution levels at locations between monitors. Smoke trajectory and evolution will also be considered using both measurement types. We also aim to baseline subtract our emission estimates from each region to determine which portion of emissions are regional and local, further characterizing burn differences in northern and southern California fires. Future directions include using INSTEP jointly with TEMPO satellite data, and mobile deployments on aircraft and uncrewed aerial vehicles (UAV).

Low-cost sensors↗
DOE OSTI2025

Multiscale Characterization of Additive Manufacturing Components with Computed Tomography, 3D X-ray Microscopy, and Deep Learning

Additive manufacturing (AM) facilitates the creation of complex-geometry parts, driving advancements in lightweight aerospace components, high-efficiency engine cooling channels, and customized medical implants. However, ensuring the quality and reliability of AM parts remains challenging due to internal defects, surface irregularities, porosity, and residual trapped powder, which are often inaccessible to traditional inspection methods. Recent developments in X-ray computed tomography (XCT) and 3D X-ray microscopy (XRM), particularly systems equipped with resolution-at-a-distance (RaaD™) capabilities, enable high-resolution, non-destructive evaluation of AM components across multiple scales, from sub-micrometer to macroscopic levels. This paper explores modern XCT and XRM techniques for multiscale characterization of AM parts, focusing on their ability to detect and analyze defects such as porosity, cracks, inclusions, and surface roughness, while offering insights into defect formation mechanisms, material properties, and process-induced variations. The integration of deep learning (DL) frameworks, including Simurgh, DeepRecon, and DeepScout, enhances XCT/XRM workflows by reducing scan times, improving resolution recovery, and enabling accurate defect detection even with limited projection data. These DL-based methods overcome limitations of traditional reconstruction techniques, enabling faster, more reliable characterization of dense materials like Inconel 718 and novel alloys such as AlCe. Applications include process parameter optimization, high-throughput quality control, and multistage AM process evaluation, with DL-enhanced workflows accelerating analysis times from weeks to days. Correlative imaging approaches further validate XCT and XRM data against scanning electron microscopy (SEM) images of physically sectioned samples, confirming the accuracy of DL-based reconstructions and enabling comprehensive defect analysis. While challenges remain in generalizing DL models to diverse materials and imaging conditions, improvements in resolution, noise reduction, and defect detection highlight the transformative potential of these methods. This multiscale and correlative approach enables precise identification and correlation of microstructural features with the overall performance of AM components. By integrating advanced XCT, XRM, and DL techniques, this paper demonstrates a significant leap forward in AM characterization, offering valuable insights into the relationships between processing parameters, microstructure, and part performance, and driving innovations that enhance the quality and reliability of AM products for demanding industrial applications.

Additive manufacturing↗
NASA NTRS2012

Analysis of Water and Energy Budgets and Trends Using the NLDAS Monthly Data Sets

The North American Land Data Assimilation System (NLDAS) is a collaborative project between NASA GSFC, NOAA, Princeton University, and the University of Washington. NLDAS has created surface meteorological forcing data sets using the best-available observations and reanalyses. The forcing data sets are used to drive four separate land-surface models (LSMs), Mosaic, Noah, VIC, and SAC, to produce data sets of soil moisture, snow, runoff, and surface fluxes. NLDAS hourly data, accessible from the NASA GES DISC Hydrology Data Holdings Portal, http://disc.sci.gsfc.nasa.gov/hydrology/data-holdings, are widely used by various user communities in modeling, research, and applications, such as drought and flood monitoring, watershed and water quality management, and case studies of extreme events. More information is available at http://ldas.gsfc.nasa.gov/. To further facilitate analysis of water and energy budgets and trends, NLDAS monthly data sets have been recently released by NASA GES DISC.

Vollmer, Bruce E.↗
NASA NTRS2012

A l% and 1cm Perspective Leads to a Novel CDOM Absorption Algorithm

A next-generation in-water profiler designed to measure the apparent optical properties of seawater was developed and validated across a wide dynamic range of water properties. This new Compact-Optical Profiling System (C-OPS) design uses a novel, kite-shaped, free-falling backplane with adjustable buoyancy and is based on 19 state-of-the-art microradiometers, spanning 320-780 nm. Data collected as part of the field commissioning were of a previously unachievable quality and showed that systematic uncertainties in the sampling protocols were discernible at the 1% optical and 1cm depth resolution levels. A sensitivity analysis as a function of three water types, established by the peak in the remote sensing reflectance spectra, revealed which water types and spectral domains were the most indicative of data acquisition uncertainties. The unprecedented vertical resolution of C-OPS measurements provided near-surface data products at the spectral endpoints with a quality level that has not been obtainable. The improved data allowed development of an algorithm for predicting the spectral absorption due to chromophoric dissolved organic matter (CDOM) using ratios of diffuse attenuation coefficients with over 99% of the variance in the data explained.

Morrow, J. H.↗
NASA NTRS1993

Algorithms and sensitivity analyses for Stratospheric Aerosol and Gas Experiment II water vapor retrieval

The operational inversion algorithm used for the retrieval of the water-vapor vertical profiles from the Stratospheric Aerosol and Gas Experiment II (SAGE II) occultation data is presented. Unlike the algorithm used for the retrieval of aerosol, O3, and NO2, the water-vapor retrieval algorithm accounts for the nonlinear relationship between the concentration versus the broad-band absorption characteristics of water vapor. Problems related to the accuracy of the computational scheme, the accuracy of the removal of other interfering species, and the expected uncertainty of the retrieved profile are examined. Results are presented on the error analysis of the SAGE II water vapor retrieval, indicating that the SAGE II instrument produced good quality water vapor data.

Chu, W. P.↗
NASA NTRS1975

Computer analysis of railcar vibrations

Computer models and techniques for calculating railcar vibrations are discussed along with criteria for vehicle ride optimization. The effect on vibration of car body structural dynamics, suspension system parameters, vehicle geometry, and wheel and rail excitation are presented. Ride quality vibration data collected on the state-of-the-art car and standard light rail vehicle is compared to computer predictions. The results show that computer analysis of the vehicle can be performed for relatively low cost in short periods of time. The analysis permits optimization of the design as it progresses and minimizes the possibility of excessive vibration on production vehicles.

Vlaminck, R. R.↗
NASA NTRS1998

In-Flight System Identification

A method is proposed and studied whereby the system identification cycle consisting of experiment design and data analysis can be repeatedly implemented aboard a test aircraft in real time. This adaptive in-flight system identification scheme has many advantages, including increased flight test efficiency, adaptability to dynamic characteristics that are imperfectly known a priori, in-flight improvement of data quality through iterative input design, and immediate feedback of the quality of flight test results. The technique uses equation error in the frequency domain with a recursive Fourier transform for the real time data analysis, and simple design methods employing square wave input forms to design the test inputs in flight. Simulation examples are used to demonstrate that the technique produces increasingly accurate model parameter estimates resulting from sequentially designed and implemented flight test maneuvers. The method has reasonable computational requirements, and could be implemented aboard an aircraft in real time.

Morelli, Eugene A.↗
DOE OSTI2024

A framework to enhance disaster debris estimation with AI and aerial photogrammetry

This study addresses the critical need to enhance disaster preparedness and response, focusing on hurricane impact assessment and debris estimation. Accurate assessments in this context are critical for post-event search-and-rescue (SAR) operations and resource distribution. Recent computing advancements are revealing the potential of unmanned aerial vehicles (UAVs) and artificial intelligence (AI) technologies in collecting data and assisting with post-hurricane reconnaissance. However, the use of AI and UAV photogrammetry for accurate disaster impact analysis remains underexplored. To this end, this study proposes a damage and debris analysis framework harnessing reality capture through aerial imagery and photogrammetry. Within this framework, a region-based neural network is leveraged to detect debris locations in aerial imagery with favorable performance. In a testbed within the Beaumont-Port Arthur region, in Southeast Texas, this study performs 3D reality captures of the built environment. Since the accuracy of the 3D reality capture is of importance in research areas associated with time-sensitive disaster response, we further investigate the optimal 2D aerial imagery overlap ratio required to generate a sufficiently accurate 3D model for disaster impact analysis and debris volume estimation. Results indicate that, in the case of aerial imagery for infrastructure systems, a minimum of 60 % overlap is recommended for damage assessment and debris analysis. In contrast, for flat green areas, a minimum of 50 % overlap is adequate. Overall, for disaster response applications, our study reveals that an overlap ratio between 60 % and 70 % is optimal for achieving a balance between time efficiency and data quality in aerial data collection. Furthermore, these quantitative recommendations are crucial for enabling efficient disaster response efforts. Additionally, our study outcomes will improve disaster impact analysis and facilitating timely and effective response strategies.

Artificial intelligence↗
DOE OSTI2024

Multicriteria Measures to Assess the Sustainability of Diets: A Systematic Review

Abstract Context Assessing the overall sustainability of a diet is a challenging undertaking requiring a holistic approach capable of addressing the multicriteria nature of this concept. Objective The aim was to identify and summarize the multicriteria measures used to assess the sustainability characteristics of diets reported at the individual level by healthy adults. Data Sources Articles were identified via PubMed, Scopus, and Web of Science. The search strategy consisted of key words and MeSH terms, and was concluded in September 2022, covering references in English, Spanish, and Portuguese. Data Extraction This systematic review followed the PRISMA guidelines. The search identified 5663 references, from which 1794 were duplicates. Two reviewers independently screened the titles and abstracts of each of the 3869 records and the full-text of the 144 references selected. Of these, 7 studies met the inclusion criteria. Data Analysis A total of 6 multicriteria measures were identified: 3 different Sustainable Diet Indices, the Quality Environmental Costs of Diet, the Quality Financial Costs of Diet, and the Environmental Impact of Diet. All of these incorporated a health/nutrition dimension, while the environmental and economic dimensions were the second and the third most integrated, respectively. A sociocultural sustainability dimension was included in only 1 of the measures. Conclusion Despite some methodological concerns in the development and validation process of the identified measures, their inclusion is considered indispensable in assessing the transition towards sustainable diets in future studies. Systematic Review Registration PROSPERO registration no. CRD42022358824.

Rei, Mariana (ORCID:0000000189453708)↗
DOE OSTI2022

Making Invisible Visible: Data-Driven Seismic Inversion With Spatio-Temporally Constrained Data Augmentation

Deep learning and data-driven approaches have shown great potential in scientific domains. The promise of data-driven techniques relies on the availability of a large volume of high-quality training datasets. Due to the high cost of obtaining data through expensive physical experiments, instruments, and simulations, data augmentation techniques for scientific applications have emerged as a new direction for obtaining scientific data recently. However, existing data augmentation techniques originating from computer vision yield physically unacceptable data samples that are not helpful for the domain problems that we are interested in. In this article, we develop new data augmentation techniques based on convolutional neural networks. Specifically, our generative models leverage different physics knowledge (such as governing equations, observable perception, and physics phenomena) to improve the quality of the synthetic data. To validate the effectiveness of our data augmentation techniques, we apply them to solve a subsurface seismic full-waveform inversion using simulated CO 2 leakage data. Our interest is to invert for subsurface velocity models associated with very small CO 2 leakage. We validate the performance of our methods using comprehensive numerical tests. Here via comparison and analysis, we show that data-driven seismic imaging can be significantly enhanced by using our data augmentation techniques. Particularly, the imaging quality has been improved by 15% in test scenarios of general-sized leakage and 17% in small-sized leakage when using an augmented training set obtained with our techniques.

58 GEOSCIENCES↗
DOE OSTI2020

A Pattern-Recognition-Based Ensemble Data Imputation Framework for Sensors from Building Energy Systems

Building operation data are important for monitoring, analysis, modeling, and control of building energy systems. However, missing data is one of the major data quality issues, making data imputation techniques become increasingly important. There are two key research gaps for missing sensor data imputation in buildings: the lack of customized and automated imputation methodology, and the difficulty of the validation of data imputation methods. In this paper, a framework is developed to address these two gaps. First, a validation data generation module is developed based on pattern recognition to create a validation dataset to quantify the performance of data imputation methods. Second, a pool of data imputation methods is tested under the validation dataset to find an optimal single imputation method for each sensor, which is termed as an ensemble method. The method can reflect the specific mechanism and randomness of missing data from each sensor. The effectiveness of the framework is demonstrated by 18 sensors from a real campus building. The overall accuracy of data imputation for those sensors improves by 18.2% on average compared with the best single data imputation method.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗