Search NASA⌕ Search

SEARCH · Search NASA

Results for “Linear Regression”

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 847 records · Page 47

Uncertainty Quantification and Sensitivity Analysis for Quantitative Risk Assessments of Hydrogen Infrastructure

Typical QRAs provide deterministic estimates and understanding of risks posed but are constructed using significant assumptions and uncertainties due to limited data availability and historical momentum of using nominal estimates. This report presents a hydrogen QRA analysis using HyRAM+ that incorporates uncertainty with Latin hypercube sampling and sensitivity analysis using linear regression.

08 HYDROGEN↗

Multi-Parameter Optical Fiber for Distributed Sensing of Humidity, CH4, CO2, and Corrosion

This work describes the use of the optical fiber sensor (OFS) for successful monitoring of humidity, CH4, and CO2 based on the strain produced along the single-mode fiber (SMF) sensor. This is enabled by absorption of H2O/gases onto the commercially available polyacrylate coated jacketed portion of the fiber resulting in a change in strain. Under equilibrium, a differential microstrain was observed along the jacketed portion of the SMF with N2, CH4, and CO2 at different relative humidity (RH) conditions. The strain response of the SMF under various mixed gas composition and different RH conditions were also measured and made calibration curves accordingly. Linear regression and principal component analysis of the strain datasets provided deconvolution of the impact of strain from H2O, N2, CH4, and CO2. Additionally, modified OFS comprised of the Fe coated fiber section was employed to monitor corrosion based on the increase in backscattered intensity amplitude of the light being passed once corrosion of Fe occurs. Also, corrosion of Fe was studied under soil by installing the fiber in soil with protective measures which would prevent the sensor being mechanically disturbed or broken during installation. The corrosion rates were studied by monitoring the rate at which the intensity of backscattered light amplitude attains a steady state value when complete corrosion of Fe with a specific coating thickness occurs.

Mainali, Badri↗

Predicting seismic amplitudes with machine learning

The accurate estimation of seismic wave amplitude is vital to precisely determine the yield, magnitude, and event discrimination possible for a given network – a critical element in nuclear explosion monitoring. This task is complicated by several factors, including but not limited to radiation pattern, scattering effects, and crustal variations, which can lead to the attenuation or amplification of amplitude along a given raypath. In this report, we explore the novel application of machine learning to the task of seismic amplitude estimation by training a simple Artificial Neural Network (ANN) on an S-wave amplitude dataset from Lai et al. (2019). Attributes from this dataset used as input to the ANN included event-station distances, station locations (latitude, longitude), event locations (latitude, longitude), event depths, event magnitudes, radiation patterns, signal-to noise ratio (SNR) measurements (average-amplitude, peak-to-trough, maximum peak), and signal periods. We find that the trained ANN predicts S-wave amplitudes with a modest tendency toward underestimating the actual values, as indicated by a linear regression between predicted and actual data (slope: 0.892, intercept: -0.651). These results suggest that an ANN can perform this task, with potential for significant improvements through improved datasets, architectures, and parameter tuning.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Comparative Analysis of HEATNETS for Geothermal Network Performance

Thermal energy networks (TENs), also known as 5th generation district energy systems, or more specifically geothermal networks when exchanging heat with geothermal boreholes, are an important technology for decarbonization. In these networks an ambient loop connects buildings and thermal sources, such as a borehole field, to exchange energy and maintain a desired loop temperature. Water-source heat pumps are used at the buildings to connect to the ambient or thermal loop to meet to the building heating and cooling loads and maintain comfort. A semi-transient, reduced-order technical model and techno-economic model, called HEATNETS, has been developed at NREL that captures the flow of energy around a TEN. In this work, a comparison of the HEATNETS technical model and a well-known coding platform used for modeling geothermal networks, TRNSYS, has been completed for a proposed geothermal network as a verification and validation process. Hourly data provided from the TRNSYS simulation included building loads, pumping power, heat pump power, temperature entering and leaving the borehole field, and mass flow rates. The hourly borehole temperatures were used to create a linear regression model utilized in HEATNETS to estimate the borehole field heat exchange. The building loads and mass flow rates were direct inputs to HEATNETS while the pumping power, heat pump power, borehole temperatures, and coefficients of performance were all simulated and calculated by HEATNETS, allowing for direct comparison of the thermal energy transfer HEATNETS considers the full process from design inputs to economic outputs and can provide modeling options for high-level initial system design and operational optimization. This study focuses on a validation of HEATNETS using results from TRNSYS. HEATNETS is not intended to replace other modeling tools, but this work demonstrates, via a comparison with an industry standard code, that HEATNETS can be a unique, high-level and rapid modeling tool for estimating the performance of a full geothermal network system.

15 GEOTHERMAL ENERGY↗

Machine Learning Analysis of Temperature-Strain Relationships for Structural Health Monitoring of Pipes: Self-powered wireless sensor system for health monitoring of liquid-sodium cooled fast reactors

This report presents machine learning (ML) analysis of temperature-strain relationships for structural health monitoring of nuclear reactor stainless steel (SS) pipes with the strain gauge sensor directly printed on the pipe with a 3D conformal aerosol jet printer. We investigate correlations for two sensor pairs installed on the same SS304 pipe: commercial K-type thermocouple with a printed gold strain gauge (TC3-SG3), and commercial K-type thermocouple with commercial Kyowa strain gauge (TC0-SG0). The temperature ranges for the sensor pairs TC0-SG0 and TC3-SG3 are 20.00°C to 266.37°C and 39.95°C to 219.28°C respectively. ML algorithms in this study include Linear Regression (baseline method), Ridge Regression, Lasso Regression, and Gradient Boosting. Performance evaluation metrics include Root Mean Square Error (RMSE), Mean Square Error (MSE), Mean Absolute Error (MAE), R 2 Score, and Explained Variance. Using advanced feature engineering techniques, we extracted 27 temperature-based features and 30 strategic inclusion features. The best performance was obtained with the Gradient Boosting method, which achieves prediction accuracy of R 2 = 0.9999 and RMSE = 7.69 μStrain for TC0-SG0, and R 2 = 0.9998 and RMSE = 18.03 μStrain for TC3-SG3. While the temperature-strain correlations are weaker for the gauge directly printed on the pipe than for the commercial strain gauge, deployment-ready performance exceeding industry standards is achieved for both sensor pairs.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Use of Dual Frequency Identification Sonar to Determine Adult Chinook Salmon (Oncorhynchus tshawytscha) Escapement in the Secesh River, Idaho ; Annual Report, January 2008 – December 2008.

Chinook salmon in the Snake River basin were listed as threatened under the Endangered Species Act in 1992 (NMFS 1992). The Secesh River represents the only stream in the Snake River basin where natural origin (wild) salmon escapement monitoring occurs at the population level, absent a supplementation program. As such the Secesh River has been identified as a long term salmon escapement and productivity monitoring site by the Nez Perce Tribe Department of Fisheries Resources Management. Salmon managers will use this data for effective population management and evaluation of the effect of conservation actions on a natural origin salmon population. The Secesh River also acts as a reference stream for supplementation program comparison. Dual frequency identification sonar (DIDSON) was used to determine adult spring and summer Chinook salmon escapement in the Secesh River in 2008. DIDSON technology was selected because it provided a non-invasive method for escapement monitoring that avoided listed species trapping and handling incidental mortality, and fish impedance related concerns. The DIDSON monitoring site was operated continuously from June 13 to September 14. The first salmon passage was observed on July 3. DIDSON site total estimated salmon escapement, natural and hatchery fish, was 888 fish {+-} 65 fish (95% confidence interval). Coefficient of variation associated with the escapement estimate was 3.7%. The DIDSON unit was operational 98.1% of the salmon migration period. Adult salmon migration timing in the Secesh River occurred over 74 days from July 3 to September 14, with 5,262 total fish passages observed. The spawning migration had 10%, median, and 90% passage dates of July 8, July 16, and August 12, respectively. The maximum number of net upstream migrating salmon was above the DIDSON monitoring site on August 27. Validation monitoring of DIDSON target counts with underwater optical cameras occurred for species identification. A total of 860 optical camera identified salmon passage observations were identical to DIDSON target counts. However, optical cameras identified eight jack salmon (3 upstream, 5 downstream) less than 55 cm in length that DIDSON did not count as salmon because of the length criteria employed ({ge} 55 cm). Precision of the DIDSON technology was evaluated by comparing estimated net upstream salmon escapement and associated 95% confidence intervals between two DIDSON sonar units operated over a five day period. The DIDSON 1 salmon escapement was 145.7 fish ({+-} 2.3), and the DIDSON 2 escapement estimate was 150.5 fish ({+-} 5). The overlap in the 95% confidence intervals suggested that the two escapement estimates were not significantly different from each other. Known length salmon carcass trials were conducted in 2008 to examine the accuracy of manually measured lengths, obtained using DIDSON software, on high frequency files at a 5 m window length. Linear regression demonstrated a highly significant relationship between known lengths and manually measured salmon carcass lengths (p < 0.0001). A positive bias in manual length measurement of 6.8% to 8% existed among the two observers in the analysis. Total Secesh River salmon escapement (natural origin and hatchery) in 2008 was 912 fish. Natural origin salmon escapement in the entire Secesh River drainage was 847 fish. The estimated natural origin spawner abundance was 836 fish. Salmon spawner abundance in 2008 increased by three fold compared to 2007 abundance levels. The 10 year geometric mean natural origin spawner abundance was 538 salmon and was below the recommended viable population threshold level established by the ICTRT (2007). One additional Snake River basin salmon population was assessed for comparison of natural origin salmon spawner abundance. The Johnson Creek/EFSF Salmon River population had a 10 year geometric mean natural origin spawner abundance of 254 salmon. Salmon spawner abundance levels in both streams were below viable population thresholds. DIDSON technology has been used in the Secesh River to determine salmon escapement over the past five years. The results suggest that DIDSON technology is reliable and can be used to generate accurate and precise estimates of salmon escapement if appropriate methods are used.

13 HYDRO ENERGY↗

Quantification of the association of ventilation rates with sick building syndrome symptoms

Data from published studies were combined and analyzed to develop best-fit equations and curves quantifying the change in sick building syndrome (SBS) symptom prevalence with ventilation rate. For each study, slopes were calculated, representing the fractional change in SBS symptom prevalence per unit change in ventilation rate per person. Values of ventilation rate, associated with each value of slope, were also calculated. Linear regression equations were fit to the resulting data points, after weighting by study size. Integration of the slope-ventilation rate equations yielded curves of relative SBS symptom prevalence versus ventilation rate. Based on these analyses, relative SBS symptom prevalence increases approximately 23percent (12percent to 32percent) as the ventilation rate drops from 10 to 5 L/s-person and relative prevalence decreases approximately 29percent (15percent to 42percent) as ventilation rate increases from 10 to 25 L/s-person.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

TASK 2.5.5 NATURAL EXPOSURE TESTING IN CALIFORNIA

Airborne particulate matter that settles on a roof can either reflect or absorb incoming solar radiation, dependent on the chemical content and size of the particles. These light scattering and absorption processes occur within a few microns of the surface, and can affect the solar reflectance of the roof. Wilkes et al. (2000) tested 24 different roof coatings on a low-slope test stand and observed about a 25% decrease in the solar reflectance of white-coated and aluminum-coated surfaces as the time of exposure increased; however, the decrease leveled off after 2 years. SPRI Inc. and its affiliates studied the effect of climatic exposure on the surface properties of white thermoplastic single-ply membranes and determined that membranes lose from 30 to 50% of their reflectance over 3 years (Miller et al. 2002). The CMRC and its affiliates AISI, NamZAC, MBMA, MCA and NCCA exposed unpainted and painted metal roofing on both steep- and low-slope test roofs and found that after 3 years, the painted polyvinylidene fluoride (PVDF) metal roofs lost less than 5% of their original reflectance (Miller et al. 2004). The results of the three different weathering studies are very interesting in terms of their solar reflectance after 3 years of exposure. The white thermoplastic membrane and white ceramic coating with white topcoat had original reflectance measures that were about 20 percentage points higher than the painted metal; however, after 3-years of field exposure the solar reflectance of the painted metal exceeds that of the thermoplastic membrane and equals that of the coating. The long-term loss of reflectance appears driven by the ability of the particulate matter to cling to the roof and resist being washed off by wind and or rain. Miller et al. (2002) discovered that aerosol deposition introduced biomass of complex microbial consortia onto the test roofs and the combination of contaminants and biomass accelerated the loss of solar reflectance for the thermoplastic membranes and the roof coatings. Airborne contaminants and biomass were also detected on the painted metal roofs; however, the loss of solar reflectance was less than 5% for the painted metal roofs. The chemistry of the PVDF paint resin system uses similar organic film bonding to that responsible for Teflon , making it extremely chemical resistant and dirt shedding. Miller and Rudolph (2003) found the PVDF painted metals maintained solar reflectance even after 30 years of climatic exposure. Therefore the reduction of roof reflectance is closely related to the composition of the roof and to the chemical profile of the contaminants soiling the roof. Contaminants collected from samples of roof products exposed at seven California weathering sites were analyzed for elements and carbons to characterize the chemical profile of the particles soiling each roof sample and to identify those elements that degrade or enhance solar reflectance. The losses in solar reflectance varied from site to site and also varied at a give site based on the color of the coupon. The least drop in reflectance was observed in the alpine climate of McArthur while the largest drop occurred in sites near urban development. Light color samples were soiled after just one year of exposure. The darker color coupons did not show the same seasonal variations in solar reflectance as observed for the lighter colors. However, after an additional year of exposure the samples at all sites regained most of their solar reflectance due to rain and/or wind washing. The loss of reflectance appears cyclical with the onset of seasons having more rainfall. Solar reflectance of the cool pigmented coupons always exceeded that of the conventional pigmented coupons. Climatic soiling did not cause the cool pigmented roof coupons to lose any more solar reflectance than their conventional pigmented counterparts. The effect of roof slope appears to have more of an effect on lighter color roofs whose solar reflectance exceeds at least 0.5 and visually shows the accumulation of airborne contaminants. The thermal emittance remained invariant with time and location and was therefore not affected by climatic soiling. A thin-film deposition model was developed based on first principles, which simulates light interaction with a soiled substrate. This model was used in combination with the measured data to determine the solar absorptance and reflectance of particulate matter at each of the sites calculated using least squares fitting routines. Principal Component Analysis was used to determine the most important combinations of chemicals correlated with changes in solar absorption. Linear regression helped extract an approximate correlation using chromium, iron and elemental carbon concentrations. It appears that chromium ranks first, iron ranks second, and elemental carbon ranks third in importance to soil absorptance in the data

09 BIOMASS FUELS↗

Methods for Analyzing Electric Load Shape and its Variability

Current methods of summarizing and analyzing electric load shape are discussed briefly and compared. Simple rules of thumb for graphical display of load shapes are suggested. We propose a set of parameters that quantitatively describe the load shape in many buildings. Using the example of a linear regression model to predict load shape from time and temperature, we show how quantities such as the load?s sensitivity to outdoor temperature, and the effectiveness of demand response (DR), can be quantified. Examples are presented using real building data.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

YALINA-booster subcritical assembly pulsed-neutron experiments : data processing and spatial corrections.

The YALINA-Booster experiments and analyses are part of the collaboration between Argonne National Laboratory of USA and the Joint Institute for Power & Nuclear Research - SOSNY of Belarus for studying the physics of accelerator driven systems for nuclear energy applications using low enriched uranium. The YALINA-Booster subcritical assembly is utilized for studying the kinetics of accelerator driven systems with its highly intensive D-T or D-D pulsed neutron source. In particular, the pulsed neutron methods are used to determine the reactivity of the subcritical system. This report examines the pulsed-neutron experiments performed in the YALINA-Booster facility with different configurations for the subcritical assembly. The 1141 configuration with 90% U-235 fuel and the 1185 configuration with 36% or 21% U-235 fuel are examined. The Sjoestrand area-ratio method is utilized to determine the reactivities of the different configurations. The linear regression method is applied to obtain the prompt neutron decay constants from the pulsed-neutron experimental data. The reactivity values obtained from the experimental data are shown to be dependent on the detector locations inside the subcritical assembly and the types of detector used for the measurements. In this report, Bell's spatial correction factors are calculated based on a Monte Carlo model to remove the detector dependences. The large differences between the reactivity values given by the detectors in the fast neutron zone of the YALINA-Booster are reduced after applying the spatial corrections. In addition, the estimated reactivity values after the spatial corrections are much less spatially dependent.

43 PARTICLE ACCELERATORS↗

Effect of Off-Target Binding on 18 F-Flortaucipir Variability in Healthy Controls Across the Life Span

Measuring early tau accumulation is important in studying aging and Alzheimer disease and is only as accurate as the signal-to-noise ratio of the tracer. Along with aggregated tau in the form of neurofibrillary tangles, 18 F-flortaucipir has been reported to bind to neuromelanin, monoamine oxidase, calcifications, iron, leptomeningeal melanocytes, and microhemorrages. Although 18 F-flortaucipir successfully differentiates healthy controls (HCs) from subjects with Alzheimer disease, variability exists in the cortical signal in amyloid-negative HCs. We aimed to explore the relationship between off-target binding signal and variability in the cortical signal in HCs. Methods: Subjects ($n$ = 139) received 11 C-Pittsburgh compound B (PIB) and 18 F-flortaucipir PET scans and a magnetization-prepared rapid gradient echo MRI scan. PET frames were realigned and coregistered to the MR images, which were segmented using FreeSurfer. In amyloid-negative HCs ($n$ = 90; age range, 21-94 y), 7 nonspecific or off-target binding regions were considered: caudate, pallidum, putamen, thalamus, cerebellar white matter, hemispheric white matter, and choroid plexus. These regions of interest were assigned to 3 similarly behaving groups using principle components analysis, exploratory factor analysis, and Pearson correlations for caudate, putamen, and pallidum (also correlated with age); thalamus and white matter; and choroid plexus. In amyloid-negative HCs with 11 C-PIB and 18 F-flortaucipir scans, correlations were calculated between white and gray matter before and after partial-volume correction. Results: The correlation between white and gray matter disappeared after partial-volume correction in 11 C-PIB (r 2 = 0) but persisted for 18 F-flortaucipir (r 2 = 0.27), demonstrating that the correlation between white and gray matter signal in 18 F-flortaucipir is not solely due to partial-volume effects. A linear regression showed that off-target signal from putamen and thalamus together explained 64% of the variability in the cortical signal in amyloid-negative HCs (not seen in amyloid-positive HCs). Variability in amyloid-negative HCs but not amyloid-positive HCs correlated with white matter signal (unrelated to partial-volume effects) and age-related off-target signal (possibly related to iron load). Conclusion: The noise in the 18 F-flortaucipir measurement could pose challenges when studying early tau accumulation.

18F-flortaucipir PET↗

Evaluation of Relationships between Growth Rate, Tree Size, Lignocellulose Composition, and Enzymatic Saccharification in Interspecific Corymbia Hybrids and Parental Taxa

In order for a lignocellulosic bioenergy feedstock to be considered sustainable,it must possess a high rate of growth to supply biomass for conversion. Despite the desirability of a fast growth rate for industrial application,it is unclear what effect growth rate has on biomass composition or saccharification. We characterized Klason lignin,glucan,and xylan content with response to growth in Corymbia interspecific F1 hybrid families (HF) and parental species Corymbia torelliana and C. citriodora subspecies variegata and measured the effects on enzymatic hydrolysis from hydrothermally pretreated biomass. Analysis of biomass composition within Corymbia populations found similar amounts of Klason lignin content (19.7–21.3%) among parental and hybrid populations,whereas glucan content was clearly distinguished within C. citriodora subspecies variegata (52%) and HF148 (60%) as compared to other populations (28–38%). Multiple linear regression indicates that biomass composition is significantly impacted by tree size measured at the same age,with Klason lignin content increasing with diameter breast height (DBH) (+0.12% per cm DBH increase),and glucan and xylan typically decreasing per DBH cm increase (-0.7 and -0.3%,respectively). Polysaccharide content within C. citriodora subspecies variegata and HF-148 were not significantly affected by tree size. High-throughput enzymatic saccharification of hydrothermally pretreated biomass found significant differences among Corymbia populations for total glucose production from biomass,with parental Corymbia torelliana and hybrids HF-148 and HF-51 generating the highest amounts of glucose (~180 mg/g biomass,respectively),with HF-51 undergoing the most efficient glucan-to-glucose conversion (74%). Based on growth rate,biomass composition,and further optimization of enzymatic saccharification yield,high production Corymbia hybrid trees are potentially suitable for fast-rotation bioenergy or biomaterial production.

60 APPLIED LIFE SCIENCES↗

Generation of Data-Driven Expected Energy Models for Photovoltaic Systems

Although unique expected energy models can be generated for a given photovoltaic (PV) site, a standardized model is also needed to facilitate performance comparisons across fleets. Current standardized expected energy models for PV work well with sparse data, but they have demonstrated significant over-estimations, which impacts accurate diagnoses of field operations and maintenance issues. This research addresses this issue by using machine learning to develop a data-driven expected energy model that can more accurately generate inferences for energy production of PV systems. Irradiance and system capacity information was used from 172 sites across the United States to train a series of models using Lasso linear regression. The trained models generally perform better than the commonly used expected energy model from international standard (IEC 61724-1), with the two highest performing models ranging in model complexity from a third-order polynomial with 10 parameters (Radj2 = 0.994) to a simpler, second-order polynomial with 4 parameters (Radj2=0.993), the latter of which is subject to further evaluation. Subsequently, the trained models provide a more robust basis for identifying potential energy anomalies for operations and maintenance activities as well as informing planning-related financial assessments. We conclude with directions for future research, such as using splines to improve model continuity and better capture systems with low (≤1000 kW DC) capacity.

14 SOLAR ENERGY↗

Relationships between Vehicle Pricing and Features: Data Driven Analysis of the Chinese Vehicle Market

A full-scale understanding of the dynamics of the Chinese vehicle market can benefit stakeholders with respect to rational decision-making and effective long-term investment. This study attempts to discover the common vehicle pricing patterns in the Chinese market by quantifying statistical correlations among critical vehicle features from intrinsic powertrain systems to extrinsic market positioning. The data samples involve almost all passenger vehicle models sold in 2013 to 2019. After comparing multiple statistical methodologies, a log-transformation variant of the multinomial linear regression model was found to be the best one, and the goodness of fit shows that this model can offer stable estimates, which were validated using 2019 market data. The insights achieved are: (1) The price and major performance features of SUVs/crossovers are similar to those of sedans; (2) If all other explicit features remain the same, the price of a Japanese midsize sedan is 62% higher than that of a Chinese midsize sedan, and European midsize vehicles have the highest prices overall. (3) The incremental price of fuel consumption varies by vehicle class and fuel economy. For example, from 30 to 50 MPG, the vehicle price increases by $119 for a Chinese brand sedan vehicle, by $69 for a Chinese brand SUV.

33 ADVANCED PROPULSION SYSTEMS↗

Development and Application of Schema Based Occupant-Centric Building Performance Metrics

Occupant behavior can significantly influence the operation and performance of buildings. Many occupant-centric key performance indicators (KPIs) rely on having accurate counts of the number of occupants in a building, which is very different to how occupancy information is currently collected in the majority of buildings today. To address this gap, the authors develop a standardized methodology for the calculation of percent space utilization for buildings, which is formulated with respect to two prevalent operational data schemas: the Brick Schema and Project Haystack. The methodology is scalable across different levels of spatial granularity and irrespective of sensor placement. Moreover, the methods are intended to make use of typical occupancy sensors that capture presence level occupancy and not counts of people. Since occupant-hours is a preferable metric to use in KPI calculations, a method to convert between percent space utilization and occupant-hours using the design occupancy for a space is also developed. The methodology is demonstrated on a small commercial office space in Boulder, Colorado using data collected between June 2018 and February 2019. A multiple linear regression is performed that shows strong evidence for a relationship between building energy consumption and percent space utilization.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Initial Findings from US Department of Energy’s Better Plants Virtual in-Plant Training on 50001 Ready

Manufacturing facilities use about 35% of the domestic energy in the United States every year. Implementing an effective energy management system (EnMS) is one of the most important approaches to improve energy efficiency. However, the implementation of EnMS is low for many countries (including the US) and even for energy-intensive sectors. The reasons for the low implementation rate of energy management systems had been investigated by multiple researchers, but very few studies have focused on the barriers and challenges of implementing ISO 50001-based energy management systems. To contribute to this understudied area, this paper discusses the implementation and outcomes of the first Better Plants 50001 Ready Virtual In-plant Training. This paper first provides an overview of 50001 Ready and the 50001 Ready Navigator Tool. Then, it provides details on this training event and its outcomes. Finally, it discusses findings from the responses to 40 live polling questions about the status of the 25 tasks of the 50001 Ready Navigator for participating companies, key components of the participating manufacturing companies’ energy management systems, and challenges and barriers that these companies are facing. The findings suggest that although many companies understood the importance of an effective energy management system, about half of them do not understand the required resources for building energy management systems, and most of them have only just begun establishing these systems and need more assistance and resources in multiple areas. More specifically, more assistance is necessary for the following: (1) improving corporate management’s understanding of the time and resources needed to build an EnMS as well as the benefits; (2) creating linear regression models for more accurate energy performance tracking; (3) understanding energy use, collecting and analyzing energy performance data; (4) optimizing equipment operational controls, and creating action plans.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Urban Green Spaces, Greenness Exposure and Species Richness in Residential Environments and Relations with Physical Activity and BMI in Portuguese Adolescents

Environmental factors play an important role in obesity-related behaviors. Evidence indicates significant associations between weight and urban green spaces in adults, but it is not clear whether this relationship applies to adolescents. Therefore, our aim was to determine the associations between urban green spaces, greenness exposure and species richness in residential environments with physical activity and body mass index. Sixty-two adolescents between 12 and 18 years of age answered a self-administered questionnaire, providing information on height, weight, age, sex and home address. Data on socioeconomic deprivation were obtained from the European Index of Deprivation for Small Portuguese Areas. Physical activity levels were assessed using accelerometers. Urban green space counts and the normalized difference vegetation index values were measured using buffers along the roads with distances of 300, 500, 1000 and 1500 m from each participant’s residence. To quantify the species richness, the species richness index was used. Linear regression models were fitted to analyze whether urban green spaces, exposure to green spaces and species richness counts for each distance were associated with physical activity and self-reported body mass index. We did not find significant associations between the independent variables and the probability of overweight or obesity. The relationship between environmental variables, adolescents’ physical activity and body weight seems to be complex and further studies may contribute to better understanding of the topic.

Melo, Juliana (ORCID:0000000301275340)↗

A Deep Neural Network for Accurate and Robust Prediction of the Glass Transition Temperature of Polyhydroxyalkanoate Homo- and Copolymers

The purpose of this study was to develop a data-driven machine learning model to predict the performance properties of polyhydroxyalkanoates (PHAs), a group of biosourced polyesters featuring excellent performance, to guide future design and synthesis experiments. A deep neural network (DNN) machine learning model was built for predicting the glass transition temperature, Tg, of PHA homo- and copolymers. Molecular fingerprints were used to capture the structural and atomic information of PHA monomers. The other input variables included the molecular weight, the polydispersity index, and the percentage of each monomer in the homo- and copolymers. The results indicate that the DNN model achieves high accuracy in estimation of the glass transition temperature of PHAs. In addition, the symmetry of the DNN model is ensured by incorporating symmetry data in the training process. The DNN model achieved better performance than the support vector machine (SVD), a nonlinear ML model and least absolute shrinkage and selection operator (LASSO), a sparse linear regression model. The relative importance of factors affecting the DNN model prediction were analyzed. Sensitivity of the DNN model, including strategies to deal with missing data, were also investigated. Compared with commonly used machine learning models incorporating quantitative structure–property (QSPR) relationships, it does not require an explicit descriptor selection step but shows a comparable performance. The machine learning model framework can be readily extended to predict other properties.

quantitative structure–property relationship (QSPR↗