Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Sparse estimation”

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 289 records · Page 16

Equation Discovery Using Fast Function Extraction: a Deterministic Symbolic Regression Approach

Advances in machine learning (ML) coupled with increased computational power have enabled identification of patterns in data extracted from complex systems. ML algorithms are actively being sought in recovering physical models or mathematical equations from data. This is a highly valuable technique where models cannot be built using physical reasoning alone. In this paper, we investigate the application of fast function extraction (FFX), a fast, scalable, deterministic symbolic regression algorithm to recover partial differential equations (PDEs). FFX identifies active bases among a huge set of candidate basis functions and their corresponding coefficients from recorded snapshot data. This approach uses a sparsity-promoting technique from compressive sensing and sparse optimization called pathwise regularized learning to perform feature selection and parameter estimation. Furthermore, it recovers several models of varying complexity (number of basis terms). FFX finally filters out many identified models using non-dominated sorting and forms a Pareto front consisting of optimal models with respect to minimizing complexity and test accuracy. Numerical experiments are carried out to recover several ubiquitous PDEs such as wave and heat equations among linear PDEs and Burgers, Korteweg–de Vries (KdV), and Kawahara equations among higher-order nonlinear PDEs. Additional simulations are conducted on the same PDEs under noisy conditions to test the robustness of the proposed approach.

42 ENGINEERING↗

Ocean Warming: From the Surface to the Deep in Observations and Models

The ocean is the primary heat sink of the global climate system. Since 1971, it has been responsible for storing more than 90% of the excess heat added to the Earth system by anthropogenic greenhouse-gas emissions. Adding this heat to the ocean contributes substantially to sea level rise and affects vital marine ecosystems. Considering the global ocean’s large role in ongoing climate variability and change, it is a good place to focus in order to understand what observed changes have occurred to date and, by using models, what future changes might arise under continued anthropogenic forcing of the climate system. While sparse measurement coverage leads to enhanced uncertainties with long-term historical estimates of change, modern measurements are beginning to provide the clearest picture yet of ongoing global ocean change. Observations show that the ocean is warming from the near-surface through to the abyss, a conclusion that is strengthened with each new analysis. In this assessment, we revisit here observation- and model-based estimates of ocean warming from the industrial era to the present and show a consistent, full-depth pattern of change over the observed record that is likely to continue at an ever-increasing pace if effective actions to reduce greenhouse-gas emissions are not taken.

54 ENVIRONMENTAL SCIENCES↗

Technical Report Series on Global Modeling and Data Assimilation: Soil Moisture Active Passive (SMAP) Project Assessment Report for the Beta-Release L4_SM Data Product - Volume 40

During the post-launch SMAP calibration and validation (Cal/Val) phase there are two objectives for each science data product team: 1) calibrate, verify, and improve the performance of the science algorithm, and 2) validate the accuracy of the science data product as specified in the science requirements and according to the Cal/Val schedule. This report provides an assessment of the SMAP Level 4 Surface and Root Zone Soil Moisture Passive (L4_SM) product specifically for the product's public beta release scheduled for 30 October 2015. The primary objective of the beta release is to allow users to familiarize themselves with the data product before the validated product becomes available. The beta release also allows users to conduct their own assessment of the data and to provide feedback to the L4_SM science data product team. The assessment of the L4_SM data product includes comparisons of SMAP L4_SM soil moisture estimates with in situ soil moisture observations from core validation sites and sparse networks. The assessment further includes a global evaluation of the internal diagnostics from the ensemble-based data assimilation system that is used to generate the L4_SM product. This evaluation focuses on the statistics of the observation-minus-forecast (O-F) residuals and the analysis increments. Together, the core validation site comparisons and the statistics of the assimilation diagnostics are considered primary validation methodologies for the L4_SM product. Comparisons against in situ measurements from regional-scale sparse networks are considered a secondary validation methodology because such in situ measurements are subject to upscaling errors from the point-scale to the grid cell scale of the data product. Based on the limited set of core validation sites, the assessment presented here meets the criteria established by the Committee on Earth Observing Satellites for Stage 1 validation and supports the beta release of the data. The validation against sparse network measurements and the evaluation of the assimilation diagnostics address Stage 2 validation criteria by expanding the assessment to regional and global scales.

ubRMSE↗

Developing Deep Learning Models for System Remaining Useful Life Predictions: Application to Aircraft Engines

Prognostics and health management (PHM) is an important part of ensuring reliable operations of complex safety- critical systems. System-level remaining useful life (RUL) estimation is a much more complex problem than making estimations at the component level, and system-level RUL methodologies remain sparse in the literature. Model-based approaches have traditionally worked in the past for components such as capacitors, MOSFETs, batteries, or hard-drives (to name a few examples), but developing high fidelity dynamics models of cyber physical systems that can be used to study the effects of multiple degrading components in the system remains a challenging task. Some initial work on model-based System RUL predictions was demonstrated in Khorasgani, et al [1], but, to generalize the system-level prognostics problem, we have to resort to pure data driven and hybrid approaches. In this work, we propose an end-to-end data- driven framework for developing deep learning models to predict remaining useful life of cyber physical systems operating under unknown faulty conditions. The raw data is organized with a data schema that improves the model development process and down stream data analysis tasks. Due to the unknown faulty conditions, the raw sensor data is transformed into signals that expose the underlying degradation processes, which are then used for model development. Bayesian Optimization is used to tune the model parameters prior to training and validation. We show that this approach results in accurate predictions within 3 cycles to end of life (EOL). We demonstrate the effectiveness of our approach by applying it to the N-CMAPSS turbofan engine dataset recently released by NASA, which includes high fidelity degradation modeling, real world operating conditions, and a large set of fault operating modes.

Prognostics↗

Assessing the Performance of the South American Land Data Assimilation System Version 2 (SALDAS-2) Energy Balance across Diverse Biomes

Understanding the exchange of energy between the surface and the atmosphere is important in view of the climate scenario. However, it becomes a challenging task due to a sparse network of observations. This study aims to improve the energy balance estimates for the Amazon, Cerrado, and Pampa biomes located in South America using the radiation and precipitation forcing obtained from the Clouds and the Earth’s Radiant Energy System (CERES) and the precipitation CPTEC/MERGE datasets. We employed three surface models—Noah-MP, Community Land Model (CLSM), and Integrated Biosphere Simulator (IBIS)—and conducted modeling experiments, termed South America Land Data Assimilation System (SALDAS-2). The results showed that SALDAS-2 radiation estimates had the smallest errors. Moreover, SALDAS-2 precipitation estimates were better than the Global Land Data Assimilation System (GLDAS) in the Cerrado (MBE = −0.16) and Pampa (MBE = −0.19). Noah-MP presented improvements compared with CLSM and IBIS in 100% of towers located in the Amazon. CLSM tends to overestimate the latent heat flux and underestimate the sensible heat flux in the Amazon. Noah-MP and Ensemble outperformed GLDAS in terms latent and sensible heat fluxes. The potential of SALDAS-2 should be emphasized to provide more accurate estimates of surface energy balance.

modeling↗

Evaluation of multiple tracer methods to estimate low groundwater flow velocities

Here, four different tracer methods were used to estimate groundwater flow velocity at a multiple-well site in the saturated alluvium south of Yucca Mountain, Nevada: (1) two single-well tracer tests with different rest or “shut-in” periods, (2) a cross-hole tracer test with an extended flow interruption, (3) a comparison of two tracer decay curves in an injection borehole with and without pumping of a downgradient well, and (4) a natural-gradient tracer test. Such tracer methods are potentially very useful for estimating groundwater velocities when hydraulic gradients are flat (and hence uncertain) and also when water level and hydraulic conductivity data are sparse, both of which were the case at this test location. The purpose of the study was to evaluate the first three methods for their ability to provide reasonable estimates of relatively low groundwater flow velocities in such low-hydraulic-gradient environments. The natural-gradient method is generally considered to be the most robust and direct method, so it was used to provide a “ground truth” velocity estimate. However, this method usually requires several wells, so it is often not practical in systems with large depths to groundwater and correspondingly high well installation costs. The fact that a successful natural gradient test was conducted at the test location offered a unique opportunity to compare the flow velocity estimates obtained by the more easily deployed and lower risk methods with the ground-truth natural-gradient method. The groundwater flow velocity estimates from the four methods agreed very well with each other, suggesting that the first three methods all provided reasonably good estimates of groundwater flow velocity at the site. We discuss the advantages and disadvantages of the different methods, as well as some of the uncertainties associated with them.

58 GEOSCIENCES↗

Climate change-resilient snowpack estimation in the Western United States

Abstract In the 21st century, warmer temperatures and changing atmospheric circulation will likely produce unprecedented changes in Western United States snowfall 1–3 , with impacts on the timing, amount, and spatial patterns of snowpack 4–7 . The ~900 snow pillow stations are indispensable to water resource management by measuring snow-water equivalent (SWE) 8,9 in strategic but fixed locations 10,11 . However, this network may not be impacted by climate change in the same way as the surrounding area 12 and thus fail to accurately represent unmeasured locations; climate change thereby threatens our ability to measure the effects of climate change on snow. In this work, we show that maintaining the current peak SWE estimation skill is nonetheless possible. We find that explicitly including spatial correlations—either from gridded observations or learned by the model—improves skill at predicting distributed snowpack from sparse observations by 184%. Existing artificial intelligence methods can be useful tools to harness the many available sources of snowpack information to estimate snowpack in a nonstationary climate.

54 ENVIRONMENTAL SCIENCES↗

Quantifying uncertainty for deep learning based forecasting and flow-reconstruction using neural architecture search ensembles

Classical problems in computational physics such as data-driven forecasting and signal reconstruction from sparse sensors have recently seen an explosion in deep neural network (DNN) based algorithmic approaches. However, most DNN models do not provide uncertainty estimates, which are crucial for establishing the trustworthiness of these techniques in downstream decision making tasks and scenarios. In recent years, ensemble-based methods have achieved significant success for the uncertainty quantification in DNNs on a number of benchmark problems. However, their performance on real-world applications remains under-explored. In this work, we present an automated approach to DNN discovery and demonstrate how this may also be utilized for ensemble-based uncertainty quantification. Specifically, we propose the use of a scalable neural and hyperparameter architecture search for discovering an ensemble of DNN models for complex dynamical systems. We highlight how the proposed method not only discovers high-performing neural network ensembles for our tasks, but also quantifies uncertainty seamlessly. This is achieved by using genetic algorithms and Bayesian optimization for sampling the search space of neural network architectures and hyperparameters. Subsequently, a model selection approach is used to identify candidate models for an ensemble set construction. Afterwards, a variance decomposition approach is used to estimate the uncertainty of the predictions from the ensemble. We demonstrate the feasibility of this framework for two tasks — forecasting from historical data and flow reconstruction from sparse sensors for the sea-surface temperature. In conclusion, we demonstrate superior performance from the ensemble in contrast with individual high-performing models and other benchmarks.

Deep ensembles↗

Global analysis of night marine air temperature and its uncertainty since 1880

An updated version of the Met Office Hadley Centre's monthly night marine air temperature data set is presented. It is available on a 5° latitude-longitude grid from 1880 as anomalies relative to 1961–1990 calendar-monthly climatological average night marine air temperature (NMAT). Adjustments are made for changes in observation height; these depend on estimates of the stability of the near surface atmospheric boundary layer. In previous versions of the data set, ad hoc adjustments were also made for three periods and regions where poor observational practice was prevalent. These adjustments are re-examined. Estimates of uncertainty are calculated for every grid box and result from measurement errors, uncertainty in adjustments applied to the observations, uncertainty in the measurement height, and under-sampling. The new data set is a clear improvement over previous versions in terms of coverage because of the recent digitization of historical observations from ships' logbooks. However, the periods prior to about 1890 and around World War II remain particularly uncertain, and sampling is still sparse in some regions in other periods. A further improvement is the availability of uncertainty estimates for every grid box and every month. Previous versions required adjustments that were dependent on contemporary measurements of sea surface temperature (SST); to avoid these, the new data set starts in 1880 rather than 1856. Overall agreement with variations of SST is better for the updated data set than for previous versions, supporting existing estimates of global warming and increasing confidence in the global record of temperature variability and change.

54 ENVIRONMENTAL SCIENCES↗

Assessment of the Impact of Discontinuity in Satellite Instruments and Retrievals on Global PM2.5

Exposure to fine particulate matter (PM2.5) is the leading environmental risk factor for mortality globally. Satellite-derived estimates of surface PM2.5 developed from a combination of satellites, simulations, and ground monitor data are relied upon for health impact studies. The ability to develop satellite-derived PM2.5 estimates requires the continued availability of aerosol optical depth (AOD) sources. This work examines the impact of the addition or loss of satellite AOD data sources on global PM2.5 estimation and the impact of continuing the longterm record with AOD from the Visible Infrared Imaging Radiometer Suite (VIIRS) on the Suomi-National Polar orbiting Partnership (S-NPP) satellite after the loss of the MODIS (MODerate resolution Imaging Spectroradiometer) and MISR (Multi-angle Imaging Spectroradiometer) instruments on board the Terra and Aqua satellites. We find that the addition of VIIRS S-NPP AOD products to geophysical PM2.5 estimates from satellites and simulations causes regional differences that correspond to differences in the VIIRS and MODIS Deep Blue AOD algorithms and sampling. Changes in long-term trends and timeseries due to the addition or loss of AOD data sources are generally within their uncertainties. Statistical fusion with ground monitor data partially corrects for changes due to sampling differences when introducing the VIIRS AOD products, but uncertainty remains over desert regions where ground monitor coverage is sparse. This work provides promise for the sustained development of global satellite-derived PM2.5 estimates, despite discontinuities in instruments and retrieval methods.

Remote sensing↗

Towards real-time monitoring: data assimilated time-lapse full waveform inversion for seismic velocity and uncertainty estimation

SUMMARY Rapid development of time-lapse seismic monitoring instrumentations has made it possible to collect dense time-lapse data for tomographically retrieving time-lapse (even continuous) images of subsurface changes. While traditional time-lapse full waveform inversion (TLFWI) algorithms are designed for sparse time-lapse surveys, they lack of effective temporal constraint on time-lapse data, and, more importantly, lack of the uncertainty estimation of the TLFWI results that is critical for further interpretation. Here, we propose a new data assimilation TLFWI method, using hierarchical matrix powered extended Kalman filter (HiEKF) to quantify the image uncertainty. Compared to existing Kalman filter algorithms, HiEKF allows to store and update a data-sparse representation of the cross-covariance matrices and propagate model errors without expensive operations involving covariance matrices. Hence, HiEKF is computationally efficient and applicable to 3-D TLFWI problems. Then, we reformulate TLFWI in the framework of HiEKF (termed hereafter as TLFWI-HiEKF) to predict time-lapse images of subsurface spatiotemporal velocity changes and simultaneously quantify the uncertainty of the inverted velocity changes over time. We demonstrate the validity and applicability of TLFWI–HiEKF with two realistic CO2 monitoring models derived from Frio-II and Cranfield CO2 injection sites, respectively. In both 2-D and 3-D examples, the inverted high-resolution time-lapse velocity results clearly reveal a continuous velocity reduction due to the injection of CO2. Moreover, the accuracy of the model is increasing over time by assimilating more time-lapse data while the standard deviation is decreasing over lapsed time. We expect TLFWI-HiEKF to be equipped with real-time seismic monitoring systems for continuously imaging the distribution of subsurface gas and fluids in the future large-scale CO2 sequestration experiments and reservoir management.

58 GEOSCIENCES↗

Estimating rates and patterns of diversification with incomplete sampling: a case study in the rosids

Premise Recent advances in generating large‐scale phylogenies enable broad‐scale estimation of species diversification. These now common approaches typically are characterized by (1) incomplete species coverage without explicit sampling methodologies and/or (2) sparse backbone representation, and usually rely on presumed phylogenetic placements to account for species without molecular data. We used empirical examples to examine the effects of incomplete sampling on diversification estimation and provide constructive suggestions to ecologists and evolutionary biologists based on those results. Methods We used a supermatrix for rosids and one well‐sampled subclade (Cucurbitaceae) as empirical case studies. We compared results using these large phylogenies with those based on a previously inferred, smaller supermatrix and on a synthetic tree resource with complete taxonomic coverage. Finally, we simulated random and representative taxon sampling and explored the impact of sampling on three commonly used methods, both parametric (RPANDA and BAMM) and semiparametric (DR). Results We found that the impact of sampling on diversification estimates was idiosyncratic and often strong. Compared to full empirical sampling, representative and random sampling schemes either depressed or inflated speciation rates, depending on methods and sampling schemes. No method was entirely robust to poor sampling, but BAMM was least sensitive to moderate levels of missing taxa. Conclusions We suggest caution against uncritical modeling of missing taxa using taxonomic data for poorly sampled trees and in the use of summary backbone trees and other data sets with high representative bias, and we stress the importance of explicit sampling methodologies in macroevolutionary studies.

59 BASIC BIOLOGICAL SCIENCES↗

Sparse Coding for N-Gram Feature Extraction and Training for File Fragment Classification

File fragment classification is an important step in the task of file carving in digital forensics. In file carving, files must be reconstructed based on their content as a result of their fragmented storage on disk or in memory. Existing methods for classification of file fragments typically use hand-engineered features such as byte histograms or entropy measures. In this paper, we propose an approach using sparse coding that enables automated feature extraction. Sparse coding, or sparse dictionary learning, is an unsupervised learning algorithm, and is capable of extracting features based simply on how well those features can be used to reconstruct the original data. With respect to file fragments, we learn sparse dictionaries for n-grams, continuous sequences of bytes, of different sizes. These dictionaries may then be used to estimate n-gram frequencies for a given file fragment, but for significantly larger n-gram sizes than are typically found in existing methods which suffer from combinatorial explosion. To demonstrate the capability of our sparse coding approach, we used the resulting features to train standard classifiers such as support vector machines (SVMs) over multiple file types. Experimentally, we achieved significantly better classification results with respect to existing methods, especially when the features were used in supplement to existing hand-engineered features.

97 MATHEMATICS AND COMPUTING↗

Gaussian Process Regression Method for Costing SmallSat Bus Capabilities

NASA is responding to the growing interest in, andcapabilities of, small satellites for science applications with an increasingnumber and frequency of Announcements of Opportunityfor small satellite space missions. Estimating the probabilitythat these mission concepts will fit within the small cost capsof these opportunities is largely driven by the probability thatone of the burgeoning number of small satellite providers will beable to meet the payload’s accommodation requirements withinthe budget for the spacecraft. JPL has collected a databasecontaining technical specifications and cost of commerciallyavailable Smallsat buses across various vendors. The primarypurpose of the database is for use in JPL’s Team X architecturestudies to inform cost estimates of a spacecraft bus which fitsthe customer’s technical requirements for their payload andmission. Customer needs are often unique and don’t alignperfectly with an off-the-shelf commercial spacecraft bus, whichmotivates the need to develop a cost model across the continuoustechnical parameter space.Al’s Bus Cost Distribution Estimator (ABCDE) uses Gaussianprocess regression (GPR) to predict commercial Smallsat spacecraftbus cost based on a subset of a customer’s technicalrequirements (payload mass, payload power, delta V, pointingcontrol, and downlink rate). GPR is implemented in ABCDE asa Bayesian method which fits an implied multivariate regressionon the technical parameters and uses kriging to intentionally“overfit” the residuals. Overfitting the residuals allows costestimates to collapse in uncertainty closer to the data pointswhile maintaining larger uncertainty intervals in regions of parameterspace with fewer data records. The data used to fit thismodel is sensitive and represents cost estimates for off-the-shelfcommercial buses. GPR simultaneously protects the sensitivityof the database and uses the sparse nature of the database toaccount for uncertainty in cost in a useful way. For a givenset of customer technical requirements, the tool provides a costestimate distribution, the percentiles of which can be interpretedas a confidence level of finding a commercial bus under a specifiedcost cap. ABCDE dramatically pushes the boundaries ofspacecraft cost estimation models due to its Bayesian methodology(accounting for the maximum uncertainty in the underlyingregression), the mathematically advanced kriging methodology,and the novelty of its application in Team X architecture tradestudies.

Austin, Alex↗

A Computational Information Criterion for Particle-Tracking with Sparse or Noisy Data

Traditional probabilistic methods for the simulation of advection-diffusion equations (ADEs) often overlook the entropic contribution of the discretization, e.g., the number of particles, within associated numerical methods. Many times, the gain in accuracy of a highly discretized numerical model is outweighed by its associated computational costs or the noise within the data. Herein, we address the question of how many particles are needed in a simulation to best approximate and estimate parameters in one-dimensional advective-diffusive transport. To do so, we use the well-known Akaike Information Criterion (AIC) and a recently-developed correction called the Computational Information Criterion (COMIC) to guide the model selection process. Random-walk and mass-transfer particle tracking methods are employed to solve the model equations at various levels of discretization. Numerical results demonstrate that the COMIC provides an optimal number of particles that can describe a more efficient model in terms of parameter estimation and model prediction compared to the model selected by the AIC even when the data is sparse or noisy, the sampling volume is not uniform throughout the physical domain, or the error distribution of the data is non-IID Gaussian.

97 MATHEMATICS AND COMPUTING↗

Use of ground-based remotely sensed data for surface energy balance calculations during Monsoon '90

Surface energy balance was evaluated at a semiarid watershed using direct and indirect measurements of the turbulent fluxes, a remote technique based on measurements of surface reflectance and temperature, and conventional meteorological information. Comparison of remote estimates of net radiant flux and soil heat flux densities with measured values showed errors on the order of +/-40 W/sq m. To account for the effects of sparse vegetation, semi-empirical adjustments to aerodynamic resistance were required for evaluation of sensible heat flux density (H). However, a significant scatter in estimated versus measured latent heat flux density (LE) was still observed, +/-75 W/sq m over a range from 100-400 W/sq m. The errors of H and LE estimates were reduced to +/-50 W/sq m when observations were restricted to clear sky conditions.

Moran, M. S.↗

Adaptive sampling for accelerating neutron diffraction-based strain mapping *

Abstract Neutron diffraction is a useful technique for mapping residual strains in dense metal objects. The technique works by placing an object in the path of a neutron beam, measuring the diffracted signals and inferring the local lattice strain values from the measurement. In order to map the strains across the entire object, the object is stepped one position at a time in the path of the neutron beam, typically in raster order, and at each position a strain value is estimated. Typical dwell times at neutron diffraction instruments result in an overall measurement that can take several hours to map an object that is several tens of centimeters in each dimension at a resolution of a few millimeters, during which the end users do not have an estimate of the global strain features and are at risk of incomplete information in case of instruments outages. In this paper, we propose an object adaptive sampling strategy to measure the significant points first. We start with a small initial uniform set of measurement points across the object to be mapped, compute the strain in those positions and use a machine learning technique to predict the next position to measure in the object. Specifically, we use a Bayesian optimization based on a Gaussian process regression method to infer the underlying strain field from a sparse set of measurements and predict the next most informative positions to measure based on estimates of the mean and variance in the strain fields estimated from the previously measured points. We demonstrate our real-time measure-infer-predict workflow on additively manufactured steel parts—demonstrating that we can get an accurate strain estimate even with 30%–40% of the typical number of measurements—leading the path to faster strain mapping with useful real-time feedback. We emphasize that the proposed method is general and can be used for fast mapping of other material properties such as phase fractions from time-consuming point-wise neutron measurements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Field testing and validation of a low-cost MPC for demand flexibility for grid-interactive K-12 schools

K-12 school buildings account for the highest energy consumption within the public sector. Implementing advanced HVAC controls in grid-interactive K-12 schools could bring substantial economic advantages and grid flexibility. Our previous study demonstrated that a low-cost model predictive control (MPC) solution, which coordinates multiple packaged units, can enable demand flexibility without major hardware upgrades. However, a significant gap remains between academic pilots and market-ready scalable solutions. This paper extends the previous single-site pilot to a multi-site demonstration involving three school campuses (95 total units) through a commercial technology transfer process. Addressing the challenge of verifying performance with sparse field data, we present a new statistical approach using Bayesian methods to estimate the MPC’s effect on peak demand. Unlike traditional methods, this approach robustly quantifies uncertainty in non-normal, limited datasets. The results confirm the solution’s replicability, achieving a 21.6–38.9% reduction in HVAC peak demand (10.8–22.1% at the site-level) with > 98% probability across diverse locations. Finally, we document critical barriers to scaling software-as-a-service (SaaS) solutions–such as API instability and diverse legacy systems–and offer practical strategies to accelerate the commercial adoption of grid-interactive efficient buildings.

Ham, Sang Woo↗