Search NASA⌕ Search

SEARCH · Search NASA

Results for “mean squared error”

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 145 records · Page 8

Validation and Spatiotemporal Analysis of CERES Surface Net Radiation Product

The Clouds and the Earth’s Radiant Energy System (CERES) generates one of the few global satellite radiation products. The CERES ARM Validation Experiment (CAVE) has been providing long-term in situ observations for the validation of the CERES products. However, the number of these sites is low and their distribution is globally sparse, and particularly the surface net radiation product has not been rigorously validated yet. Therefore, additional validation efforts are highly required to determine the accuracy of the CERES radiation products. In this study, global land surface measurements were comprehensively collected for use in the validation of the CERES net radiation (R n ) product on a daily (340 sites) and a monthly (260 sites) basis, respectively. The validation results demonstrated that the CERES R n product was, overall, highly accurate. The daily validations had a Mean Bias Error (MBE) of 3.43 W·m -2 , Root Mean Square Error (RMSE) of 33.56 W·m -2 , and R 2 of 0.79, and the monthly validations had an MBE of 3.40 W·m -2 , RMSE of 25.57 W·m -2 , and R 2 of 0.84. The accuracy was slightly lower for the high latitudes. Following the validation, the monthly CERES R n product, from March 2000 to July 2014, was used for a further analysis. The global spatiotemporal variation of the R n , which occurred during the measurement period, was analyzed. In addition, two hot spot regions, the southern Great Plains and south-central Africa, were then selected for use in determining the driving factors or attribution of the R n variation. We determined that R n over the southern Great Plains decreased by -0.33 W·m -2 per year, which was mainly driven by changes in surface green vegetation and precipitation. In south-central Africa, R n decreased at a rate of -0.63 W·m -2 per year, the major driving factor of which was surface green vegetation

54 ENVIRONMENTAL SCIENCES↗

Improving seasonal precipitation forecasts in the Western United States through statistical downscaling

Abstract Seasonal precipitation forecasts in the western United States are critical resources for water resource management, especially during winter. While current seasonal forecasting systems provide monthly precipitation forecasts operationally, their coarse resolution limits their effectiveness in capturing the localized precipitation patterns and snowpack conditions essential for water resource managers in the mountainous regions. Here, analog statistical downscaling is demonstrated as an effective approach to enhance the spatial resolution of operational seasonal forecasts provided by the North American Multi-Model Ensemble. Downscaling was performed by building an analog ‘library’, in which corresponding model forecasts and observed values during the training period were stored. In the testing period, unseen model forecasts referenced the closest historical forecast from the analog library and applied the corresponding observational value for each point. This analysis indicates that downscaled products can capture localized features more accurately than the original coarse resolution forecasts, reducing forecast error across the western United States. Moreover, downscaling individual ensemble members—rather than downscaling the ensemble mean—further reduces forecasting error for their multi-model ensemble mean products. The greatest error reductions in the downscaled product, measured by root mean squared error (RMSE), were observed at low to mid-elevations (500–2000 meters), with 50%–70% improvement relative to the original forecasts. In the higher elevations (2000 meters and above), changes in RMSE relative to the original forecast were limited to 10%–30% improvements. The improvement is more substantial for forecast systems with 10 ensemble members compared to that with 4 members, but this relationship does not hold for the system with 24 ensemble members. These findings show that analog statistical downscaling can effectively address the spatial limitations of seasonal precipitation forecasts with minimal computational cost, providing a valuable framework for enhancing coarse resolution forecasting products while providing insights into the timing of ensemble mean calculations during the downscaling process.

Vernon, B. (ORCID:0009000891670689)↗

MABAL: a Novel Deep-Learning Architecture for Machine-Assisted Bone Age Labeling

Bone age assessment (BAA) is a commonly performed diagnostic study in pediatric radiology to assess skeletal maturity. The most commonly utilized method for assessment of BAA is the Greulich and Pyle method (Pediatr Radiol 46.9:1269–1274, 2016; Arch Dis Child 81.2:172–173, 1999) atlas. The evaluation of BAA can be a tedious and time-consuming process for the radiologist. As such, several computer-assisted detection/diagnosis (CAD) methods have been proposed for automation of BAA. Classical CAD tools have traditionally relied on hard-coded algorithmic features for BAA which suffer from a variety of drawbacks. Recently, the advent and proliferation of convolutional neural networks (CNNs) has shown promise in a variety of medical imaging applications. There have been at least two published applications of using deep learning for evaluation of bone age (Med Image Anal 36:41–51, 2017; JDI 1–5, 2017). However, current implementations are limited by a combination of both architecture design and relatively small datasets. The purpose of this study is to demonstrate the benefits of a customized neural network algorithm carefully calibrated to the evaluation of bone age utilizing a relatively large institutional dataset. In doing so, this study will aim to show that advanced architectures can be successfully trained from scratch in the medical imaging domain and can generate results that outperform any existing proposed algorithm. The training data consisted of 10,289 images of different skeletal age examinations, 8909 from the hospital Picture Archiving and Communication System at our institution and 1383 from the public Digital Hand Atlas Database. The data was separated into four cohorts, one each for male and female children above the age of 8, and one each for male and female children below the age of 10. The testing set consisted of 20 radiographs of each 1-year-age cohort from 0 to 1 years to 14–15+ years, half male and half female. The testing set included left-hand radiographs done for bone age assessment, trauma evaluation without significant findings, and skeletal surveys. A 14 hidden layer-customized neural network was designed for this study. The network included several state of the art techniques including residual-style connections, inception layers, and spatial transformer layers. Data augmentation was applied to the network inputs to prevent overfitting. A linear regression output was utilized. Mean square error was used as the network loss function and mean absolute error (MAE) was utilized as the primary performance metric. MAE accuracies on the validation and test sets for young females were 0.654 and 0.561 respectively. For older females, validation and test accuracies were 0.662 and 0.497 respectively. For young males, validation and test accuracies were 0.649 and 0.585 respectively. Finally, for older males, validation and test set accuracies were 0.581 and 0.501 respectively. The female cohorts were trained for 900 epochs each and the male cohorts were trained for 600 epochs. An eightfold cross-validation set was employed for hyperparameter tuning. Test error was obtained after training on a full data set with the selected hyperparameters. Using our proposed customized neural network architecture on our large available data, we achieved an aggregate validation and test set mean absolute errors of 0.637 and 0.536 respectively. To date, this is the best published performance on utilizing deep learning for bone age assessment. Our results support our initial hypothesis that customized, purpose-built neural networks provide improved performance over networks derived from pre-trained imaging data sets. We build on that initial work by showing that the addition of state-of-the-art techniques such as residual connections and inception architecture further improves prediction accuracy. This is important because the current assumption for use of residual and/or inception architectures is that a large pre-trained network is required for successful implementation given the relatively small datasets in medical imaging. Instead we show that a small, customized architecture incorporating advanced CNN strategies can indeed be trained from scratch, yielding significant improvements in algorithm accuracy. It should be noted that for all four cohorts, testing error outperformed validation error. One reason for this is that our ground truth for our test set was obtained by averaging two pediatric radiologist reads compared to our training data for which only a single read was used. This suggests that despite relatively noisy training data, the algorithm could successfully model the variation between observers and generate estimates that are close to the expected ground truth.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Techno-economic implications and cost of forecasting errors in solar PV power production using optimized deep learning models

Accurate solar Photovoltaic (PV) power forecasting is important for enhancing both the performance and economic feasibility of PV systems. This study evaluates several deep learning models, including Dense Neural Networks (DNN), Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN), and a hybrid LSTMCNN model, for predicting PV power production one day in advance. Prior to optimization, the models exhibited relatively high errors, with the best model (DNN) achieving a Root Mean Square Error (RMSE) of 31.13 kW and a coefficient of determination (R 2 ) of 62.15 %. After employing Bayesian optimization, the LSTM-CNN model demonstrated the best performance, with the RMSE reduced to 9.79 kW and R 2 improved to 97.62 %, showcasing significant enhancement in predictive accuracy. Here, the economic evaluation considered three cases: rewards for underestimation (0.08 USD/kWh), no rewards, and penalties for both over-and underestimation (120 % of the utility tariff). In the rewards scenario, the LSTM-CNN model reduced the Levelized Cost of Electricity (LCOE) by 4 %, while in the penalty scenario, a backup diesel generator would have increased the LCOE by 49 %. Additionally, the LSTM-CNN model minimized financial losses, achieving the lowest penalties and maximizing net cash flow compared to other models, demonstrating its overall technical and economic superiority.

Deep learning↗

An Error-Entropy Minimization Algorithm for Tracking Control of Nonlinear Stochastic Systems with Non-Gaussian Variables

This paper presents an error-entropy minimization tracking control algorithm for a class of dynamic stochastic system. The system is represented by a set of time-varying discrete nonlinear equations with non-Gaussian stochastic input, where the statistical properties of stochastic input are unknown. By using Parzen windowing with Gaussian kernel to estimate the probability densities of errors, recursive algorithms are then proposed to design the controller such that the tracking error can be minimized. The performance of the error-entropy minimization criterion is compared with the mean-square-error minimization in the simulation results.

Entropy↗

Demonstration of an AI-driven workflow for dynamic x-ray spectroscopy

X-ray absorption near edge structure (XANES) spectroscopy is a powerful technique for characterizing the chemical state and symmetry of individual elements within materials, but requires collecting data at many energy points which can be time-consuming. While adaptive sampling methods exist for efficiently collecting spectroscopic data, they often lack domain-specific knowledge about the structure of XANES spectra. Here we demonstrate a knowledge-injected Bayesian optimization approach for adaptive XANES data collection that incorporates understanding of spectral features like absorption edges and pre-edge peaks. We show this method accurately reconstructs the absorption edge of XANES spectra using only 15–20% of the measurement points typically needed for conventional sampling, while maintaining the ability to determine the x-ray energy of the sharp peak after the absorption edge with errors less than 0.03 eV, the absorption edge with errors less than 0.1 eV; and overall root-mean-square errors less than 0.005 compared to traditionally sampled spectra. Our experiments on battery materials and catalysts demonstrate the method’s effectiveness for both static and dynamic XANES measurements, improving data collection efficiency and enabling better time resolution for tracking chemical changes. This approach advances the degree of automation in XANES experiments, reducing the common errors of under- or over-sampling points near the absorption edge and enabling dynamic experiments that require high temporal resolution or limited measurement time.

Bayesian optimization↗

Short-Term Global Horizontal Irradiance Forecasting Based on Sky Imaging and Pattern Recognition

Accurate short-term forecasting is crucial for solar integration in the power grid. In this paper, a classification forecasting framework based on pattern recognition is developed for 1-hour-ahead global horizontal irradiance (GHI) forecasting. Three sets of models in the forecasting framework are trained by the data partitioned from the preprocessing analysis. The first two sets of models forecast GHI for the first four daylight hours of each day. Then the GHI values in the remaining hours are forecasted by an optimal machine learning model determined based on a weather pattern classification model in the third model set. The weather pattern is determined by a support vector machine (SVM) classifier. The developed framework is validated by the GHI and sky imaging data from the National Renewable Energy Laboratory (NREL). Results show that the developed short-term forecasting framework outperforms the persistence benchmark by 16% in terms of the normalized mean absolute error and 25% in terms of the normalized root mean square error.

14 SOLAR ENERGY↗

SU-C-204-06: Monte Carlo Dose Calculation for Kilovoltage X-Ray-Psoralen Activated Cancer Therapy (X-PACT): Preliminary Results

Purpose: X-PACT is an experimental cancer therapy where kV x-rays are used to photo-activate anti-cancer therapeutics through phosphor intermediaries (phosphors that absorb x-rays and re-radiate as UV light). Clinical trials in pet dogs are currently underway (NC State College of Veterinary Medicine) and an essential component is the ability to model the kV dose in these dogs. Here we report the commissioning and characterization of a Monte Carlo (MC) treatment planning simulation tool to calculate X-PACT radiation doses in canine trials. Methods: FLUKA multi-particle MC simulation package was used to simulate a standard X-PACT radiation treatment beam of 80kVp with the Varian OBI x-ray source geometry. The beam quality was verified by comparing measured and simulated attenuation of the beam by various thicknesses of aluminum (2–4.6 mm) under narrow beam conditions (HVL). The beam parameters at commissioning were then corroborated using MC, characterized and verified with empirically collected commissioning data, including: percent depth dose curves (PDD), back-scatter factors (BSF), collimator scatter factor(s), and heel effect, etc. All simulations were conducted for N=30M histories at M=100 iterations. Results: HVL and PDD simulation data agreed with an average percent error of 2.42%±0.33 and 6.03%±1.58, respectively. The mean square error (MSE) values for HVL and PDD (0.07% and 0.50%) were low, as expected; however, longer simulations are required to validate convergence to the expected values. Qualitatively, pre- and post-filtration source spectra matched well with 80kVp references generated via SPEKTR software. Further validation of commissioning data simulation is underway in preparation for first-time 3D dose calculations with canine CBCT data. Conclusion: We have prepared a Monte Carlo simulation capable of accurate dose calculation for use with ongoing X-PACT canine clinical trials. Preliminary results show good agreement with measured data and hold promise for accurate quantification of dose for this novel psoralen X-ray therapy. Funding Support, Disclosures, & Conflict of Interest: The Monte Carlo simulation work was not funded; Drs. Adamson & Oldham have received funding from Immunolight LLC for X-PACT research.

60 APPLIED LIFE SCIENCES↗

Improvement and Verification of Online Cross Section Generation Capability of Griffin for TRISO-fueled Reactors

Griffin, a MOOSE-based reactor multiphysics code jointly developed by Idaho National Laboratory and Argonne National Laboratory under the DOE Office of Nuclear Energy’s NEAMS program, has pursued the development of an online multigroup cross section generation capability for a few years to enable high-fidelity, problem-dependent neutronics analyses of advanced thermal reactors. Recent advancements in Griffin’s online multigroup cross section generation capability have significantly improved the accuracy, robustness, and efficiency of self-shielding calculations for both prismatic and pebble-bed TRISO-fueled reactor applications. Key developments include a unified fuel self-shielding method applicable to both TRISO and annular compact/spherical shell fuel zone geometries; an advanced Dancoff Category-based Equivalence Theory using a bell function for non-fuel resonance treatment, achieving more than an order-of-magnitude speedup compared to the Tone method; an on-the-fly multigroup equivalence approach to mitigate group condensation errors; and a streaming correction method for pebble-bed homogenization. A proof-of-concept demonstration of on-the-fly group condensation with consistent P0 transport correction was also achieved. The method reproduced direct fine-group solutions with excellent accuracy (eigenvalue errors within 10 pcm and pin-power differences within 0.5%), but due to performance limitations of the current fixed-source solver, improvements to solver efficiency will be addressed in future work. Verification tests were performed on graphite-moderated TRISO-fueled two-dimensional core benchmark problems representing gas-cooled microreactors, heat pipe-cooled microreactors, gas-cooled pebble-bed reactors, and fluoride salt-cooled high-temperature reactors. Across all cases, Griffin showed excellent agreement with Serpent2 continuous energy Monte Carlo solutions: eigenvalue errors within 200 pcm, pin-power root-mean-square errors within 2%, and control rod and drum worth errors less than 2%. It should be noted that, for the benchmark problem, cross section generation contributed less than 3% of the total simulation times. These results demonstrate that Griffin’s online cross section generation capability delivers accurate and efficient reactor physics solutions across a wide spectrum of TRISO-fueled advanced reactor designs. With further improvements to the fine-group fixed-source solver and planned extensions to depletion, transients, and coupled neutron–gamma transport, Griffin will be well-positioned to become a powerful and comprehensive tool for advanced reactor analysis.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

A Cabled Acoustic Telemetry System for Detecting and Tracking Juvenile Salmon: Part 2. Three-Dimensional Tracking and Passage Outcomes

In Part 1 of this paper [1], we presented the engineering design and instrumentation of the Juvenile Salmon Acoustic Telemetry System (JSATS) cabled system, a nonproprietary technology developed by the U.S. Army Corps of Engineers, Portland District, to meet the needs for monitoring the survival of juvenile salmonids through the 31 dams in the Federal Columbia River Power System. Here in Part 2, we describe how the JSATS cabled system was employed as a reference sensor network for detecting and tracking juvenile salmon. Time-of-arrival data for valid detections on four hydrophones were used to solve for the three-dimensional (3D) position of fish surgically implanted with JSATS acoustic transmitters. Validation tests demonstrated high accuracy of 3D tracking up to 100 m from the John Day Dam spillway. The along-dam component, used for assigning the route of fish passage, had the highest accuracy; the median errors ranged from 0.06 to 0.22 m, and root mean square errors ranged from 0.05 to 0.56 m at distances up to 100 m. For the case study at John Day Dam during 2008, the range for 3D tracking was more than 100 m upstream of the dam face where hydrophones were deployed, and detection and tracking probabilities of fish tagged with JSATS acoustic transmitters were higher than 98%. JSATS cabled systems have been successfully deployed on several major dams to acquire information for salmon protection and for development of more “fish-friendly” hydroelectric facilities.

13 HYDRO ENERGY↗

DEM Modeling and Validation of Pebble Bed Packing Using Chrono::GPU

Accurate prediction of pebble packing structure is important for pebble bed reactors because the spatial distribution of void fraction directly affects coolant flow, pressure drop, heat transfer, and neutronic behavior. However, experimentally validated DEM studies that directly evaluate local void-fraction structure in reactor-relevant pebble beds remain limited. In this work, the pebble bed experiment conducted at Missouri University of Science and Technology is simulated using the graphics processing unit (GPU)-based discrete element method (DEM) code Chrono::GPU. The study focuses on evaluating the ability of Chrono::GPU to reproduce the packing arrangement and void-fraction distribution of a randomly packed spherical pebble bed. The DEM results are first verified against established radial void-fraction correlations, including the Mueller and Vortmeyer-Schuster models, to assess the predicted bulk porosity, near-wall behavior, and oscillatory packing structure. The simulation is then verified against reference DEM data and validated against gamma-ray computed tomography (CT) experimental data at three axial locations. The Chrono::GPU results reproduce the main features of the experimental packing, including the high void fraction near the wall, the first near-wall trough, and the damped oscillatory radial profile caused by wall-induced ordering. Quantitative comparison with DEM data and the CT-based radial profiles shows good agreement, with mean absolute errors on the order of 0.07 and root-mean-square errors below 0.09 for the averaged profiles. These results demonstrate that Chrono::GPU can accurately capture the void-fraction structure of spherical pebble beds and provides a reliable DEM framework for future pebble bed reactor packing, recycling, and thermal-hydraulic studies.

97 - MATHEMATICS AND COMPUTING↗

Intraoperative Localization of Tantalum Markers for Proton Beam Radiation of Choroidal Melanoma by an Opto-Electronic Navigation System: A Novel Technique

Purpose: External beam proton radiation therapy has been used since 1975 to treat choroidal melanoma. For tumor location determination during proton radiation treatment, surgical tantalum clips are registered with image data. This report introduces the intraoperative application of an opto-electronic navigation system to determine with high precision the position of the tantalum markers and their spatial relationship to the tumor and anatomical landmarks. The application of the technique in the first 4 patients is described. Methods and Materials: A navigated reference base was attached noninvasively to the eye, and a navigated pointer device was used to record the spatial position of the tantalum markers, the tumor, and anatomical landmarks. Measurement accuracy was assessed on ex vivo porcine eye specimen by repetitive recording of the tantalum marker positions. The method was applied intraoperatively on 4 patients undergoing routine tantalum clip surgery. The spatial position information delivered by the navigation system was compared to the geometric data generated by the EYEPLAN software. Results: In the ex vivo experiments, the maximum repetition error was 0.34 mm. For the intraoperative application, the root mean square error of paired-points matching of the marker positions from the navigation system and from the EYEPLAN software was 0.701-1.25 mm. Conclusions: Navigation systems are a feasible tool for accurate localization of tantalum markers and anatomic landmarks. They can provide additional geometric information, and therefore have the potential to increase the reliability and accuracy of external beam proton radiation therapy for choroidal melanoma.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Experiments on a vapor compression air conditioner with liquid desiccants for efficient dehumidification

Buildings require air conditioning systems that not only cool and dehumidify supply air but also provide sufficient ventilation to ensure indoor air quality and occupant comfort. However, standard recirculation systems-which introduce about a 10 % to 20 % fraction of outdoor air-often fail to deliver air that is precisely cooled and dry, particularly because 80-90 % of the ventilation cooling load is latent. Mixing humid ventilation air with recirculated indoor air increases the energy and costs required to condition the air to comfortable levels. Dedicated outdoor air systems (DOASs) are designed to handle this latent dominated ventilation load and thus need to have efficient humidity removal. Many cooling cycles can perform this task. Here we describe a liquid desiccant DOAS, which combines a vapor compression cycle and a liquid desiccant absorber and desorber pair. We present its performance at 26 operating conditions and a thermodynamic model which can accurately predict the moisture removal efficiency. The model's performance predictions have a mean percentage error of 2.5 % and a coefficient of variation of the root mean square error of 7.5 %. We also compare the performance of this vapor-compression-coupled liquid desiccant system with a standard vapor compression system with the same components but no liquid desiccant. For the 26 conditions tested in this study, this comparison shows that adding liquid desiccants lowers the required evaporator cooling load by 21 %, allows for 25 % lower compressor volumetric capacity, and 25 % lower electricity use. Future work will leverage this model to quantify the reduction in annual electricity use across different climates, including the need for a standard vapor compression system to reheat the air during some of the year.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Training the Recurrent neural network by the Fuzzy Min-Max algorithm for fault prediction

In this paper, we present a training technique of a Recurrent Radial Basis Function neural network for fault prediction. We use the Fuzzy Min-Max technique to initialize the k-center of the RRBF neural network. The k-means algorithm is then applied to calculate the centers that minimize the mean square error of the prediction task. The performances of the k-means algorithm are then boosted by the Fuzzy Min-Max technique.

97 MATHEMATICS AND COMPUTING↗

Di-CNN: Domain-Knowledge-Informed Convolutional Neural Network for Manufacturing Quality Prediction

In manufacturing, convolutional neural networks (CNNs) are widely used on image sensor data for data-driven process monitoring and quality prediction. However, as purely data-driven models, CNNs do not integrate physical measures or practical considerations into the model structure or training procedure. Consequently, CNNs’ prediction accuracy can be limited, and model outputs may be hard to interpret practically. This study aims to leverage manufacturing domain knowledge to improve the accuracy and interpretability of CNNs in quality prediction. A novel CNN model, named Di-CNN, was developed that learns from both design-stage information (such as working condition and operational mode) and real-time sensor data, and adaptively weighs these data sources during model training. It exploits domain knowledge to guide model training, thus improving prediction accuracy and model interpretability. A case study on resistance spot welding, a popular lightweight metal-joining process for automotive manufacturing, compared the performance of (1) a Di-CNN with adaptive weights (the proposed model), (2) a Di-CNN without adaptive weights, and (3) a conventional CNN. The quality prediction results were measured with the mean squared error (MSE) over sixfold cross-validation. Model (1) achieved a mean MSE of 6.8866 and a median MSE of 6.1916, Model (2) achieved 13.6171 and 13.1343, and Model (3) achieved 27.2935 and 25.6117, demonstrating the superior performance of the proposed model.

47 OTHER INSTRUMENTATION↗

Machine Learning Techniques Applied to Sensor Data Correction in Building Technologies

Since commercial and residential buildings account for nearly half of the United States' energy consumption, making them more energy-efficient is a vital part of the nation's overall energy strategy. Sensors play an important role in this research by collecting data needed to analyze performance of components, systems, and whole-buildings. Given this reliance on sensors, ensuring that sensor data are valid is a crucial problem. Solutions being researched are machine learning techniques, namely: artificial neural networks and Bayesian Networks. Types of data investigated in this study are: (1) temperature; (2) humidity; (3) refrigerator energy consumption; (4) heat pump liquid pressure; and (5) water flow. These data are taken from Oak Ridge National Laboratory's (ORNL) ZEBRAlliance research project which is composed of four single-family homes in Oak Ridge, TN. Results show that for the temperature, humidity, pressure, and flow sensors, data can mostly be predicted with root-mean-square error (RMSE) of less than 10% of the respective sensor's mean value. Results for the energy sensor are not as good; RMSE are centered about 100% of the mean value and are often well above 200%. Bayesian networks have RSME of less than 5% of the respective sensor's mean value, but took substantially longer to train.

building technologies↗

Energy Performance Assessment of Radiant Cooling System through Modeling and Calibration at Component Level

The paper describes a case study of an information technology office building with a radiant cooling system and a conventional variable air volume (VAV) system installed side by side so that performancecan be compared. First, a 3D model of the building involving architecture, occupancy, and HVAC operation was developed in EnergyPlus, a simulation tool. Second, a different calibration methodology was applied to develop the base case for assessing the energy saving potential. This paper details the calibration of the whole building energy model to the component level, including lighting, equipment, and HVAC components such as chillers, pumps, cooling towers, fans, etc. Also a new methodology for the systematic selection of influence parameter has been developed for the calibration of a simulated model which requires large time for the execution. The error at the whole building level [measured in mean bias error (MBE)] is 0.2%, and the coefficient of variation of root mean square error (CvRMSE) is 3.2%. The total errors in HVAC at the hourly are MBE = 8.7% and CvRMSE = 23.9%, which meet the criteria of ASHRAE 14 (2002) for hourly calibration. Different suggestions have been pointed out to generalize the energy saving of radiant cooling system through the existing building system. So a base case model was developed by using the calibrated model for quantifying the energy saving potential of the radiant cooling system. It was found that a base case radiant cooling system integrated with DOAS can save 28% energy compared with the conventional VAV system.

Building Energy Simulation↗

National soil organic carbon estimates can improve global estimates

Global inventories of spatial and vertical distribution of soil organic carbon (SOC) stocks are being used in national and global initiatives targeted to mitigate climate change and land degradation impacts. Yet, national level high-resolution estimates of SOC stocks can be useful for improving the accuracy of global SOC inventories. We estimated spatially resolved SOC stocks of surface 0–30 cm and subsurface 30–100 cm layers at a spatial resolution of 30m in tropical Island, Sri Lanka using a legacy harmonized soil database of 122 soil profiles. The national estimates were compared with two global estimates derived from WISE30sec and SoilGrids250m. The tropical Island (land area=64,610 km2) occupying 0.03% of global land area showed a considerable heterogeneity in SOC stocks ranging from 2.0-342.5 Mg ha -1 and 2.7-391.7 Mg ha -1 in the surface and subsurface soil layers, respectively. We found, elevation, precipitation and slope angle as main environmental controllers of the spatial distribution of SOC stocks under tropical climate. Incorporating the pedogenic information (derived from soil series level legacy map, soil orders and suborders) with environmental controllers resulted in better regression models of predicting surface (R 2 =0.61) and subsurface (R 2 =0.81) SOC stocks. Geographically weighted regression kriging derived maps of SOC stocks revealed that 0–100 cm soil layer of the tropical Island stored 500 Tg C contributing for 0.04% of the global SOC stocks. The validation results of our estimates showed low Mean Estimation Error (MEE: surface -1.6 and subsurface -1.6 Mg ha -1 ) and Root Mean Square Error (RMSE: surface 29.5 and subsurface 24.9 Mg ha -1 ) indicating a low bias and satisfactory predictions. The relative improvement of the prediction accuracy of the SOC stocks of our geospatial estimates in the 0–30 cm layer in comparison to SoilGrids250m and WISE30sec data derived SOC stocks were 51.7% and 35.2%, respectively. The SOC stocks predictions of the 30-100 cm soil layer showed even better relative improvement compared to SoilGrids250m (78.4%) and WISE30sec (57.4%) SOC estimates. Compared to estimates of total SOC stocks resulted in this study, WISE30sec data derived SOC stock maps showed 30% over estimation of the C stock in surface 0-30 cm (332 Tg C) and 41% overestimation in 30-100 cm layer (343 Tg C). The over estimation of total SOC stocks by the SoilGrids250 SOC stocks map for the surface 0-30 cm layer was 122% (567 Tg C) and for the 30-100 cm layer it was 209% (750 Tg C). We conclude that the fusion of legacy soil information of SOC stocks with appropriate environmental covariates and pedogenic information derived from legacy area-class soil maps at national level can produce more accurate inventories of spatial and vertical distribution of SOC stocks. As a result, these national inventories have a great potential of upgrading global inventories of SOC stocks.

54 ENVIRONMENTAL SCIENCES↗