Large-scale Building Research Articles

A Diverse Large-Scale Building Dataset and a Novel Plug-and-Play Domain Generalization Method for Building Extraction

In this paper, we introduce a new building dataset and propose a novel domain generalization method to facilitate the development of building extraction from high-resolution remote sensing images. The problem with the current building datasets involves that they lack diversity to train a practical learning model with good generalization ability, and the quality of the labels is unsatisfactory. To address these issues, we built a diverse, large-scale, and high-quality building dataset named the WHU-Mix building dataset, which is more practice-oriented. The WHU-Mix building dataset consists of a training/validation set containing 43,727 diverse images collected from all over the world, and a test set containing 8402 images from five other cities on five continents. In addition, to further improve the generalization ability of a building extraction model, we propose a domain generalization method named batch style mixing (BSM), which can be embedded as an efficient plug-and-play module in the front-end of a building extraction model, providing the model with a progressively larger data distribution to learn data-invariant knowledge. The experiments conducted in this study confirmed the potential of the WHU-Mix building dataset to improve the performance of a building extraction model, resulting in a 6–36% improvement in mIoU, compared to the other existing datasets. The adverse impact of the inaccurate labels in the other datasets can cause about 20% IoU decrease. The experiments also confirmed the high performance of the proposed BSM module in enhancing the generalization ability and robustness of a model, exceeding the baseline model without domain generalization by 13% and the recent domain generalization methods by 4–15% in mIoU.

Decoupling Semantic and Edge Representations for Building Footprint Extraction From Remote Sensing Images

Very high-resolution (VHR) earth observation systems provide an ideal data source for man-made structure detection such as building footprint extraction. Manually delineating building footprints from the remotely sensed VHR images, however, is laborious and time-intensive; thus, automation is needed in the building extraction process to increase productivity. Recently, many researchers have focused on developing building extraction algorithms based on the encoder-decoder architecture of convolutional neural networks. However, we observe that this widely adopted architecture cannot well preserve the precise boundaries and integrity of the extracted buildings. Moreover, features obtained by shallow convolutional layers contain irrelevant background noises that degrade building feature representations. This paper addresses these problems by presenting a feature decoupling network (FD-Net) that exploits two essential building information from the input image, including semantic information that concerns building integrity and edge information that improves building boundaries. The proposed FD-Net improves the existing encoder-decoder framework by decoupling image features into the edge subspace and the semantic subspace; the decoupled features are then integrated by a supervision-guided fusion process considering the heterogeneity between edge and semantic features. Furthermore, a lightweight and effective global context attention module is introduced to capture contextual building information and thus enhance feature representations. Comprehensive experimental results on three real-world datasets confirm the effectiveness of FD-Net in large-scale building mapping. We applied the proposed method to various encoder-decoder variants to verify the generalizability of the proposed framework. Experimental results show remarkable accuracy improvements with less computational cost.

CFNet: An Eigenvalue Preserved Approach to Multiscale Building Segmentation in High-Resolution Remote Sensing Images

In recent years, AI and Deep Learning (DL) methods have been widely used for object classification, recognition, and segmentation of high-resolution multispectral remote sensing images. These DL-based solutions perform better compare to traditional spectral algorithms but still suffer from insufficient optimization of global and local features of object context. In addition, failure of code-data isolation and/or disclosure of detailed eigenvalues causes serious privacy and even secret leakage due to the sensitivity of high-resolution remote sensing data and their processing mechanisms. In this paper, Class Feature (CF) modules have been presented in the decoder part of an attention-based CNN network to distinguish between building and non-building (background) area. In this way, context features of a focused object can be extracted with more details being processed, whilst the resolution of images is maintained. The reconstructed local and global feature values and dependencies in the proposed model are maintained by reconfiguring multiple effective attention modules with contextual dependencies to achieve better results for the eigenvalue. According to quantitative results and their visualization, the proposed model has depicted better performance over others' work using two large-scale building remote sensing datasets. The F1-score of this model reached 87.91 and 89.58 on WHU Buildings Dataset and Massachusetts Buildings Dataset, respectively, which exceeded the other semantic segmentation models.

Improving building inventory with a machine learning approach: application in southern Italy

The compiling of large-scale building inventory is a fundamental step in the evaluation of seismic risk in a region of interest. For classification of buildings in vulnerability classes, different levels of detail could be employed depending on the model adopted for the vulnerability assessment. Generally, the key structural characteristics of exposed building needed for the application of more refined vulnerability models are not easily available. Aimed to improve the information quality relevant to exposure, the interview-based form Cartis was recently implemented by ReLUIS under supervision of Italian Civil Protection Department within “Territorial Themes” DPC-ReLUIS project. Being based on an interview protocol, this form allows to rapidly detect relevant buildings data at urban level. Despite Cartis is an extremely useful tool for the compiling of regional scale inventories, nowadays the database covers only the 5% of Italian territory. This study aims to exploit the actual data available in Cartis database to build an exposure model that could be applied for preliminary inventory evaluation also in towns where the Cartis form is not compiled yet. To this end, a machine learning approach is applied. Cartis data available for several municipalities in Campania region are used together with basic information on buildings and population provided by Italian census data (ISTAT) to train a supervised learning algorithm for identification of homogenous urban sectors in terms of buildings features (i.e., Town Compartments) in municipal area. Once the algorithm is implemented, it may be employed in other municipalities to delimit relative Town Compartments and to identify most common building typologies within them.

Characteristics of Space Planning for Non-housing Remodeling Rental Housing

The purpose of the study was to provide basic data for activating the supply of non-housing remodeling rental housing in the future by identifying the basic status of non-housing remodeling rental housing cases and analyzing spatial characteristics in terms of remodeling. The results of the case analysis are as follows, First, The exterior of the building visually determines the image and plays an important role in providing a pleasant environment, so even if a part of the building is used, the design of the whole building needs to be improved. Second, when remodeling an entire building or commercial space, it has the advantage that it can be freely configured and utilized according to the plan intention of the designer or operator, resulting in more diverse types of personal and community space. Third, Currently, it is being supplied to young single-person households, but it is also necessary to expand the supply target to newlyweds and the elderly, so that various generations can coexist together, such as in Japanese Collective House. In addition, large-scale buildings should be remodeled into local community local community spaces such as small libraries, theaters, etc. that local that local residents can use together to change the perception of rental housing and take advantage of locational advantages.

Characterization of the dynamic mechanical properties of low-iron float glass through Split-Hopkinson-Pressure-Bar tests

Low-iron ultra-clear float glass (LIFG) has been widely used in landmark and large-scale buildings in recent years due to its aesthetic characteristics. A better understanding of the dynamic mechanical properties of LIFG is essential for the blast resistance analysis and design of glass facades. This paper presents a series of quasi-static tests and dynamic tests (using Split-Hopkinson-Pressure-Bar) to study the dynamic compressive and tensile behavior of LIFG. Strain rate effect has been investigated on compressive strength in the range of 10-5 s−1 to 103 s−1 and splitting tensile strength in the range of 10-5 s−1 to 40 s−1. During the tests, an ultra-high-speed camera was employed to capture the crack initiation and propagation. The test results show that both the dynamic compressive and tensile strengths of LIFG are strain-rate dependent, nevertheless the dynamic tensile strength is more sensitive to strain rate than the compressive strength. The strain rate effect is insignificant on the Young’s modulus of LIFG. In addition, the upper limits of strain rate are identified for dynamic compression and splitting tension of glass through SHPB facilities based on a conceptual analysis. For LIFG specimens with the length of 8 mm (for compression) or the diameter of 20 mm (for splitting tension), the upper limit of strain rate is about 2500 s−1 for compression and about 40 s−1 for splitting tension. Increasing or reducing the specimen dimension will correspondingly decrease or increase the upper strain rate limits.

Automated Construction of Structures using 3D Printing: A Review

The recent advances in 3D printing techniques from past one decade in terms of materials development, product evolvement methods, industry 4.0, technological up-gradations, manpower free constructions, customised production, waste minimization, freedom from designs at an affordable cost. 3D printing processes are extensively contributing in the field of structural industry. To meet the fast-growing demand of the humans there is a need to understand the concepts and in-depth knowhow of 3D printing processes. The technological developments of 3DP to infrastructure domain are still untouched zone worldwide. It is because of lesser projects successfully carried out since its inception. The article presents insights of additive manufacturing processes and its applicability to the construction industry. In the construction industry, 3D printing has not yet reached the point of commercial availability at global level. Fewer number of 3DP experiments for large-scale buildings can be discovered in the literature. A critical analysis of past and present literature studies is conducted worldwide and also investigated the potential benefits of additive manufacturing processes to construction industry. This article's unique contribution is its examination of the many uses for 3D printing in the construction industry, as well as the advances in technology that have made this possible.

Review: Air-Conditioning in Modern American Architecture, 1890–1970

Review: <i>Air-Conditioning in Modern American Architecture, 1890–1970</i>

A variational inference based learning approach for decentralized building fire estimation

Real-time estimation of the fire status is critical for reliable decision-making in building fire emergencies. Learning-based approaches for fire estimation have achieved great success, benefited from exploration of sensor data. However, existing works are mostly designed for specific building structures and seldom account for false alarms. Such flaws affect their generalization in broader environments and robustness in practical use. In this paper, we propose a decentralized learning-based approach for building fire estimation. The estimation is conducted in a decentralized scheme, where building zones conduct fire estimation respectively. The approach is built on variational inference technique, and can infer the location and dynamic intensity of the fires concurrently. Specifically, the fire estimation problem is interpreted in a latent variable model (LVM), with latent variables for fire intensity features and location features. The solution is then derived from variational inference (VI), and approximated via data learning on a novel deep network (namely FireEst-VAE). The network can produce accurate estimation of the fire intensity, while performing false alarm identification as a bonus effect. We evaluate the decentralized approach on full-scale experimental and simulation-generated data from a multi-compartment residence. The results illustrate that the proposed approach performs well on a wide range of fire scenarios with one or multiple fire locations, while preventing false alarms from common nuisance sources. The decentralized learning-based approach for fire estimation shows great potential in smart firefighting for large-scale buildings. • A decentralized learning approach based on variational inference for fire estimation. • Integrating interpretability from model methods and efficiency from learning methods. • The decentralized scheme benefits edge computing and generalization in practical use. • Promising performances on fire intensity estimation and false alarm identification.

Large-scale individual building extraction from open-source satellite imagery via super-resolution-based instance segmentation approach

Building footprint is a primary dataset of an urban geographic information system (GIS) database. Therefore, it is essential to establish a robust and automated framework for large-scale building extraction. However, the characteristic of remote sensing images complicates the application of the instance segmentation method based on the Mask R-CNN model, which ought to be improved toward extracting and fusing multi-scale features. Moreover, open-source satellite image datasets with wider spatial coverage and temporal resolution than high-resolution images may exhibit different coloration and resolution. This study proposes a large-scale building extraction framework based on super-resolution (SR) and instance segmentation using a relatively lower-resolution (>0.6 m) open-sourced dataset. The framework comprises four steps: color normalization and image super-resolution, scene classification, building extraction, and scene mosaicking. We took Hyogo Prefecture, Japan (19,187 km2) as a test area and extracted 1,726,006 (29.12 km2) of the 3,301,488 buildings (32.46 km2), where the number of buildings and footprint area increased by 3.0 % and 5.0 % respectively. The result indicated that the color normalization and image super-resolution could improve the visual quality of open-source satellite images and contribute to building extraction accuracy. Moreover, the improved Mask R-CNN based on Multi-Path Vision Transformer (MPViT) backbone achieved F1 scores of 0.71, 0.70, 0.81, and 0.67 for non-built-up, rural, suburban, and urban areas, respectively, which is better than those of the baseline model and other mainstream instance segmentation approaches. This study demonstrates the potential of acquiring acceptable building footprint maps from open-source satellite images, which has significant practical implications.

A Systematic Literature Review of Physics-Based Urban Building Energy Modeling (UBEM) Tools, Data Sources, and Challenges for Energy Conservation

Urban building energy modeling (UBEM) is a practical approach in large-scale building energy modeling for stakeholders in the energy industry to predict energy use in the building sector under different design and retrofit scenarios. UBEM is a relatively new large-scale building energy modeling (BEM) approach which raises different challenges and requires more in-depth study to facilitate its application. This paper performs a systematic literature review on physics-based modeling techniques, focusing on assessing energy conservation measures. Different UBEM case studies are examined based on the number and type of buildings, building systems, occupancy schedule modeling, archetype development, weather data type, and model calibration methods. Outcomes show that the existing tools and techniques can successfully simulate and assess different energy conservation measures for a large number of buildings. It is also concluded that standard UBEM data acquisition and model development, high-resolution energy use data for calibration, and open-access data, especially in heating and cooling systems and occupancy schedules, are among the biggest challenges in UBEM adoption. UBEM research studies focused on developing auto-calibration routines, adding feedback loops for real-time updates, future climate data, and sensitivity analysis on the most impactful modeling inputs should be prioritized for future research.

Secure and Reliable Indoor Localization Based on Multitask Collaborative Learning for Large-Scale Buildings

Accurate and reliable indoor location estimate is crucial for many Internet-of-Things (IoT) applications in the era of smart buildings. However, the positioning accuracy and security of the existing positioning works cannot meet the demands in the large-scale smart buildings scenarios covering multiple multifloor buildings. Therefore, in this article, we focus on the reliable and accurate localization under multibuilding and multifloor environments. We propose two novel designs, including a two-step reliable feature selector and a multitask collaborative positioning model. First, we design a two-step reliable feature selector based on an access point (AP) confidence model and manifold learning, to help select the most representative and reliable fingerprint features. Second, we propose a multitask cooperative positioning model, which consists of a multiscale feature fusion module to adaptively fuse multiscale features and a multitask joint learning module to effectively constrain the cumulative error of multiscale position. Finally, based on the above two, we propose a reliable multibuilding and multifloor localization method (RMBMFL), which can achieve accurate and reliable location estimates with low computational complexity in a smart building complex. We did real-world experiments in a 20 000 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${m^{2}}$ </tex-math></inline-formula> site that covers three multistory buildings to evaluate the performance of the proposed RMBMFL. The experimental results show that RMBMFL achieves a building identification accuracy and a floor identification accuracy of 99%, and a room-level indoor localization with an average positioning error within 2 m, and outperforms state-of-the-art solutions.

Development and economic viability analysis of photovoltaic (PV) energy powered decentralized ultrafiltration of rainwater for potable use

This paper approaches long-term experiments for the treatment of rainwater by ultrafiltration (UF) followed by chlorination to produce potable water The water produced met the standards of drinking water in terms of physical, chemical, and microbiological parameters (coliforms and E. coli). The UF (50 kDa) hollow fiber membrane achieved a high permeate flux (135 L h−1 m−2). The use of photovoltaic (PV) energy was evaluated to energize the proposed system for small, medium, and large-scale building catchment areas for rainwater. The economic assessment shows that a water cost of 0.17 US$ m−3, 0.10 US$ m−3, and 0.05 US$ m−3 for 230 m2, 2300 m2, and 11,500 m2 of rooftop areas, respectively, without a PV power source. With the installation of PV panels as the source of power, similar water costs were found, 0.13 US$ m−3, 0.10 US$ m−3, and 0.04 US$ m−3 for 230 m2, 2300 m2, and 11,500 m2, respectively. The results demonstrate that treatment of rainwater using UF followed by chlorination powered by renewable energy (PV) is a technical and economical alternative to supply drinking water through decentralized systems.

Experimental Investigation of Two-Phase Flow Distribution with Different Vertical Header Configurations

In this study, we investigated the behavior of two-phase flow distribution inside a vertical header of a microchannel heat exchanger (MCHX) that functions as an evaporator of a heat pump system. In general, the two-phase flow distribution behavior of the refrigerant differs depending on the target application, which ranges from small-scale automobile air-conditioners to large-scale building heat pump systems. Particularly, it is reported that the distribution characteristics in the vertical header of the MCHX vary extensively according to the inlet flow conditions of the refrigerant and the physical profile of the header. In this study, the physical configurations (header height, branch tube diameter) of four types of vertical headers were considered. Thereafter, the operating conditions in an experimental device that simulates an MCHX with a vertical header were selected. The experiment was performed under R410A as the working fluid, with a saturation temperature of 15 °C, inlet mass flow rate of 50–150 kg h−1 (mass flux of 908–2723 kg m−2 s−1), and an inlet vapor quality of 0.1–0.2. The liquid and vapor flow ratios and the relative standard deviation were adopted as metrics to characterize the uniformity of flow distribution. The distribution characteristics were subsequently described according to Reynolds and Froude numbers. The larger the Reynolds number and the smaller the Froude number, the more uniform the two-phase flow distribution becomes. A correlation was proposed as a function of the Reynolds and Froude numbers to predict the flow distribution characteristics for the considered vertical headers.

Revealing the Impact of Urban Form on COVID-19 Based on Machine Learning: Taking Macau as an Example

The COVID-19 pandemic has led to a re-examination of the urban space, and the field of planning and architecture is no exception. In this study, a conditional generative adversarial network (CGAN) is used to construct a method for deriving the distribution of urban texture through the distribution hotspots of the COVID-19 epidemic. At the same time, the relationship between urban form and the COVID-19 epidemic is established, so that the machine can automatically deduce and calculate the appearance of urban forms that are prone to epidemics and may have high risks, which has application value and potential in the field of planning and design. In this study, taking Macau as an example, this method was used to conduct model training, image generation, and comparison of the derivation results of different assumed epidemic distribution degrees. The implications of this study for urban planning are as follows: (1) there is a correlation between different urban forms and the distribution of epidemics, and CGAN can be used to predict urban forms with high epidemic risk; (2) large-scale buildings and high-density buildings can promote the distribution of the COVID-19 epidemic; (3) green public open spaces and squares have an inhibitory effect on the distribution of the COVID-19 epidemic; and (4) reducing the volume and density of buildings and increasing the area of green public open spaces and squares can help reduce the distribution of the COVID-19 epidemic.

Real-time data-based performance analysis of a large-scale building applied PV system

Solar photovoltaic (PV) is one of the fastest-growing renewable technologies. Building applied photovoltaic systems can significantly help reduce the energy and environmental loads of buildings. This study analyzes the performance of a 425 kW building applied photovoltaic system by investigating important technical, economic, and environmental parameters. These analyses have been carried out with the help of year-long real-time performance data. The technical performance indicators examined in this study include direct current and alternating current power output, array yield, specific (final) yield, references yield, array losses, module and overall PV conversion efficiency, capacity factor, and performance ratio. The main strengths of the study include the large size of the BAPV system, and a comprehensive techno-economic and environmental analysis based upon real-time monitored data. The PV system has been found to annually produce 835.60 MWh and 814.88 MWh of DC and AC power output, respectively. The performance ratio varies from 74.03% in August to 84.62% in January with an annual figure of 78.09%. The annual capacity factor is determined to be 21.85%. For a project life of 25 years, the simple payback period and Net Present Value have been calculated to be 6.92 years and USD 866,565.24, respectively. Over the project lifetime, the savings of CO2, CH4 and N2O are calculated to be 15,666 tons, 483.75 kg, and 83.25 kg respectively.

Short-term load forecasting for multiple buildings: A length sensitivity-based approach

With the rapid development of large-scale building energy monitoring platforms, it is of great significance to develop precise forecasting methods for buildings on a large scale to achieve better energy system design, system operation, energy management, and renewable energy integration in the grid. Traditionally, using all available historical data to train a data-driven model has been widely employed to ensure prediction performance because more historical information can be learned. However, this strategy may introduce more noise, especially for short-term load forecasting. Thus, this study proposes a novel approach for selectively utilizing building historical data to determine the amount of data that should be used to train the data-driven model. First, the CV(RMSE) curve of each building reflecting the relationship between training data length and forecasting accuracy is obtained using LightGBM. Second, clustering techniques such as k-means are used to identify buildings that are sensitive to the training data length based on CV(RMSE) curves. Finally, the optimal training data length for day-ahead forecasting is estimated for each building. The case study shows that approximately 20% of buildings in the Building Data Genome are labeled as length-sensitive buildings, and adopting appropriate training data lengths can reduce the prediction error of these buildings by up to 15%.

Rapid generation of editable engineering drawings from 3D point cloud reconstruction for large-scale buildings

Floor plans are essential for maintaining and renovating existing structures. As a result, there is an increasing requirement for the quick and automated development of up-to-date editable engineering drawings of aged buildings. This research creates a fast and totally automated framework for producing editable engineering drawings from 3D point cloud data obtained by a Lidar laser scanner. The suggested framework achieves reasonably high accuracy in drawing wall lines and marking doors on the floor layout. In this article, experiments were carried out to validate the correctness of the automatically generated engineering drawing. The intersection over union (IoU) metric of the wall lines is up to 98%, and the errors for wall-dimension and door detection F1-measure are approximately 8% and 83%, respectively. The total framework-running time is less than 1 min to generate the final editable engineering drawing of the floor plan in dxf file format.

Technical Feasibility Assessment for a Novel Fifth-Generation District Heating Model of Interconnected Operation with a Large-Scale Building

In this study, a novel fifth-generation district heating (DH) model was proposed that implements the energy-prosumer concept of bilateral heat trading (BHT) process between the DH network and the building. The newly proposed BHT model can be characterized by the feature of using the low temperature of DH return pipe’s water. The technical feasibility of the proposed BHT model was evaluated through operation simulation analysis based on the actual operation data of the hybrid pilot system combined with the fuel cell and heat pump and the annual hourly temperature profile of the existing DH return pipe. The main objective of this study is to examine the technical feasibility of the interconnection operation model with the existing DHN as an alternative to overcome the limitations of the current fuel cell cogeneration model, which suffers from the low production volume caused by the high initial investment cost. From the simulation results, it was confirmed that considerable operational benefit, more than 30% in terms of primary energy savings, can be achieved with the proposed model, and compared to the stand-alone model of the fuel cell cogeneration system for the building, it can provide a more flexible technical environment to improve the system utilization rate by about 40%.

타워크레인 전문신호수의 효율적 운영체계에 대한 연구

Recently, in the case of domestic construction sites, the ratio of construction equipment to construction manpower is increasing due to high-rise and large-scale buildings. As a result, the death toll from construction machinery and equipment accidents accounts for more than 20% of all deaths in the construction industry. Accordingly, from 2017, as a measure to strengthen safety management for tower cranes, the installation and dismantling team registration system was introduced and the right to stop work was strengthened, the professional qualification system was introduced (tower crane installation and dismantling workers), the collision prevention system became mandatory, and the installation, dismantling, and climbing work were recorded. the age limit of old cranes (restriction of use in principle for more than 20 years), and mandatory non-destructive inspection are added and managed, but most of them are limited to the management and technical fields of equipment aspects of tower cranes and have been strengthened. It seems that there are very few things. As with the results of the analysis of disaster cases of tower cranes in the past, the role of a dedicated signal operator is important when lifting weights, but anyone can play a role as a dedicated signal operator signal operator if they only receive 8 hours of special safety training, so it is urgent to supplement this. In this study, based on the review of previous research results and the analysis of disaster cases that occurred in the past 10 years (2010~2019) of tower cranes by the Korea Occupational Safety and Health Agency, we reviewed the signal management system in advanced countries such as the United States and the United Kingdom. An interview was conducted with a tower crane official (manager of a professional construction company, tower crane operator) and a signal officer, and direct and indirect problems were identified. Based on this, an improvement plan was derived by conducting a survey, and the AHP (Analytic Hierarchy Process) analysis method was used to derive the priority of the improvement plans. As a result of the study, the management system for dedicated signalman should be nationally reorganized and built and managed by improving the dedicated signalman system laws and systems. The result was that As in the results of this study, it is expected that tower crane accidents will not occur any more by implementing the improvement of laws and systems for tower crane dedicated signalmen.