Search

This study explores the diurnal variations and influencing factors of PM 2.5 , NO 2 , and ozone concentrations in educational buildings. Utilizing an integrated system of indoor and outdoor sensors, building automation control networks, and walk-through inspections, air quality data along with relevant characteristics were collected from ten educational buildings in Central Florida. Advanced Neural Network models (RNNs and CNNs), including the Long Short-Term Memory (LSTM) and the Attention Temporal Convolutional Network (ATCN) algorithms based on the Least Absolute Shrinkage and Selection Operator (LASSO), were developed to accurately identify diurnal patterns in indoor air quality (IAQ) and the differences in influencing factors. The findings indicate greater variability in diurnal differences and factors influencing indoor NO 2 and ozone concentrations compared to PM 2.5. Although the factors influencing day and night PM 2.5 levels were similar, there were significant differences in the contribution weights of these factors. Optimized RNNs and CNNs significantly outperformed standard Artificial Neural Network (ANN) models in dynamically simulating and predicting target pollutants. Comparative analysis of the root-mean-square error (RMSE) demonstrated that LASSO-LSTM models comprehensively outperformed LASSO-ATCN models by averaging 13.4% (p < 0.05). These results can be referenced in studies concerning Indoor Air Quality (IAQ) control conducted in similar environmental settings. • Indoor and outdoor air pollution measurements were conducted in ten BACnet-equipped buildings. • Indoor PM 2.5 , NO 2 , and O 3 exhibited diurnal fluctuations in their influencing factors. • The D/N ratio of indoor NO 2 and its contributions from influencing factors are unstable. • LASSO-based ANNs significantly outperformed standard ANNs in simulating targeted pollutants. • LASSO-LSTM outperformed LASSO-ATCN by averaging 13.4% (p < 0.05). [ABSTRACT FROM AUTHOR]

Speech intelligibility is a critical aspect of building science, particularly in educational buildings where poor sound quality may have a detrimental impact on students' learning and teachers' health. However, considering the numerous building regulations proposing varying definitions and ranges of acoustic comfort, calculating the necessary acoustic indicators can be challenging for designers. Speech intelligibility is a crucial component of indoor acoustics and acoustic comfort and can be calculated using formulas, simulation software, and data-based web tools. While formulas are fast, they lack details; acoustic simulation software is highly accurate but time-consuming and expensive. Data-based web tools, including machine learning algorithms, offer both speed and accuracy and are widely accessible. In this study, we present a system utilizing machine learning techniques to predict acoustic indicators, numeric and heatmap, in an educational building. The Pachyderm plugin in the Grasshopper was utilized to conduct extensive simulations in a single educational space, focusing on acoustic indicators in six different frequencies and general modes. Using Catboost and the pix2pix algorithm, the prediction models provide numerical and image indices on the developed dataset. Also, SHAP values were employed to interpret the Catboost model, analyzing the significance of each feature. The results showed remarkable accuracy, (i.e., between 89 % and 99 %) in the numerical portion, and PSNR index ranging from 0.817 to 0.970, and an SSIM index ranging from 15.56 to 31.57 in the image section. By utilizing data-driven methods, the system provides an efficient and accurate approach to calculating acoustic indicators, helping to ensure optimal acoustic environment in educational buildings. • It addressed challenges in calculating indoor acoustic indicators in buildings. • Providing a solution for calculating acoustic comfort in educational buildings. • It introduced a system for predicting both numerical and image-based indicators. • Using CatBoost and pix2pix algorithms, the system achieves remarkable accuracy. • Employed SHAP values to analyze the significance of each feature. [ABSTRACT FROM AUTHOR]

The thermal comfort models defined in existing standards are usually ineffective in predicting students' mean thermal sensation. Accordingly, overheating and overcooling are frequently reported, which affect students' well-being and cause unnecessary energy waste. Existing studies have not further explored the model improvement strategies for educational buildings. Hence, this research aims to improve the mean thermal sensation prediction model for students in naturally ventilated schools by incorporating correlated factors, based on a holistic field survey in the Mediterranean climate. The results indicate that the modified extended predicted mean vote of thermal adaptations reinforced around thermal neutrality has the highest accuracy among all validated adaptive predicted mean vote models. However, the performance of the adaptive predicted mean vote models is limited by the deficiencies of the original predicted mean vote model. The overall performance of the improved models was doubled after introducing the identified correlated factors, but the intrinsic error of the predicted mean vote index can slightly affect the models. As a result, the improved model built directly with the identified correlated factors achieved the best performance. Moreover, the potential of the machine learning models is limited by the volume of the mean thermal sensation vote data that can often be obtained, therefore conventional models such as multiple linear regression can be considered an effective and practical tool for developing the mean thermal sensation vote models. The findings of this study benefit the well-being of students and the energy conservation of educational buildings. • Improvement strategies were investigated for the predicted mean vote model. • Adaptive predicted mean vote models were validated in Mediterranean schools. • The operable window and door area was found to be correlated with thermal comfort. • Predictive models of students' mean thermal sensation vote were proposed. [ABSTRACT FROM AUTHOR]

Abstract Effects of indoor environmental quality of office buildings on occupants have been widely studied. However, relatively fewer studies focus on that of educational buildings, through the perspective of building life cycle costing assessment in particular. In this study, effects of indoor thermal condition and air quality on students' wellbeing and performance are investigated with other major building metrics through a life cycle costing case study of two university tutorial rooms. Metrics for students' wellbeing and performance are monetized, and different weighting schemes for the metrics are compared with sensitivity analysis. Results show that consideration of students' wellbeing and performance can lead to significant total net benefits, which highlights that other traditional metrics focusing on building energy and resource efficiency should be balanced with human factors. For more sustainable building design and operation, balanced weighting scheme should be adopted, in contrary to the current practice where weights are mainly given to capital and energy while wellbeing and performance are often ignored. The results underscore the importance of incorporating students' wellbeing and performance into educational building design and operation, as thermal or air quality conditions in school classrooms are often worse than stipulations in standards and building codes. [ABSTRACT FROM AUTHOR]

Floor dust samples were collected from Jordanian indoor environments (eight dwellings and an educational building) in Amman. Quantitative PCR (qPCR) analyses of selected fungal and bacterial groups were performed. The bacterial and fungal concentrations were also correlated with PAHs concentrations, which were previously measured in the same samples by using GC–MS. The bacterial and fungal concentrations varied significantly among and within the tested indoor environments. Based on the collected samples in the entrance area of the dwellings, the largest variation was found in Gram-negative bacteria and total fungi concentration. The lowest bacterial and fungal concentrations were found in the dwelling that was least occupied and the most recently built. At the educational building, the Gram-positive bacteria concentrations were lower than those observed in the dwellings. Unlike for bacteria, we observed significant negative correlation with some polycyclic aromatic hydrocarbons (PAHs). This calls for further studies investigating biodegradation of PAHs in house dust and presence of potentially health hazardous PAH metabolites. Since biocontamination in floor dust has been given relatively little to no attention in the MENA region we recommend that more extensive measurements be conducted in the future with chemical and biological analysis of floor dust contaminants and their exposure indoors. [ABSTRACT FROM AUTHOR]

The characteristic that describes the extent to which buildings and their systems maintain their performance during shocks is called resilience. Building policies in the EU have already addressed the resilience of buildings against possible hazards (i.e., natural disasters, extreme weathers, fires). However, with increasing overheating risks (e.g., climate change) accompanied by their detrimental health and economic impacts, the thermal performance of nearly zero energy buildings (nZEB) is not guaranteed. This study aims to assess the impact of shocks and combinations on the thermal resilience of educational nZEB against heatwaves (HW) and system shocks (SS) including failure of indirect evaporative cooling (IEC), natural night ventilation (NNV) and solar shading failure (SF). A Modelica model of the building was developed and experimentally validated. Shocks were classified and quantified using the novel normalized degree of shock (d o S) index. Heatwaves (HWs) had 20 × to 93 × more critical impact than the worst SS (NNV failure). Additionally SS occurring at the start of the operational period is 1.2 × more critical than SS occurring later in the day as it allowed for significant heat build-up in both classrooms. In future climate scenarios a combination of HWs and power outages will become frequent. This study showed that a combination of a full day of cooling strategy and shading failure occurring on the hottest day of a 6-day long HW , is 10% more critical on both lecture rooms than an individual 10-day long HW. Classroom with heavy thermal mass prolongs the absorptivity of the shocks but delays recovery. [Display omitted] • Impact of external & system shocks on resilience to overheating of educational nZEB. • External & system shocks (SS) quantified with the novel normalized degree of shock. • Heatwaves (HW) are 20–93 × more critical than IEC, NNV & shading failure. • System shock at the start of operation is 1.2× more critical than later in the day. • Combined 6 day HW & SS is 10% more critical than a single 10 day HW. [ABSTRACT FROM AUTHOR]

Lately EU has promoted several policies with the aim of reducing buildings energy impact. Despite such policies have successfully contributed to reduce residential buildings (RBs) energy consumption, non-residential buildings (NRBs) have shown an increasing of operational energy demand by 15.7%, during last decade. On the one hand, energy impacts are underestimated since only primary energy consumption (PEC) is considered while other energies, such as those related to the construction phase, replacements, or end-use phase, are missed. On the other hand, policies commonly lack of specific standards regarding NRBs since NRBs heterogeneity makes its standardization difficult. Therefore, with the aim of contributing to literature and in order to show the true impact of the overall energy consumption, this paper assesses a new school building. This was erected in partnership with constructors and energy consumption has been measured for two years. Calculations have been based on the Life cycle energy assessment methodology by taking into account the pre-use phase, the use phase and the estimated post-use phase. It has been concluded that, despite the building was designed for a very low PEC (<92 kWh m −2 year −1 ) this consumption only represents 56% of the total demanded energy along the building life cycle. [ABSTRACT FROM AUTHOR]

The existing stock of institutional buildings constructed before current thermal regulation codes is known to be high-energy consuming. To make energy savings, retrofitting solutions have to deal with important transformations of those buildings (e.g. envelope, energy systems) and with better-suited management solutions. Such technical solutions quite often neglect occupants' comfort. The present work aims to develop and implement an energy audit protocol to tackle simultaneously the questions of thermal comfort and energy efficiency for higher education buildings. Our transverse approach allowed us to achieve a complementary view of the building under examination, including its operating conditions. At any rate, capturing the full complexity of a building-system (building energy devices, management strategies, and occupancy and behaviours impacts) requires a broad perspective and points to the limits of key-in-hand audits and solutions. [ABSTRACT FROM AUTHOR]

The paper discusses cost-efficient energy performance renovation measures for typical educational buildings built in the 1960s and 1970s in cold climate regions. The study analyzes the impact of different energy renovation measures on the energy efficiency and economic viability in a Finnish case study educational building located in Lappeenranta University of Technology (LUT) campus area. The main objective of the study was to determine the cost-optimal energy performance renovation measures to meet the proposed national nearly zero-energy building (nZEB) requirements, which are defined according to the primary energy consumption of buildings. The main research method of the study was simulation-based optimization (SBO) analysis, which was used to determine the cost-optimal renovation solutions. The results of the study indicate that the minimum national energy performance requirement of new educational buildings (E primary ⩽ 170 kWh/(m 2 ,a)) can be cost-effectively achieved in deep renovations of educational buildings. In addition, the proposed national nZEB-targets are also well achievable, while improving the indoor climate (thermal comfort and indoor air quality) conditions significantly at the same time. Cost-effective solutions included renovation of the original ventilation system, a ground source heat pump system with relatively small dimensioning power output, new energy efficient windows and a relatively large area of PV-panels for solar-based electricity production. The results and conclusions of this study can be generalized to similar climates and techno-economic environments. [ABSTRACT FROM AUTHOR]

Building retrofitting is the most feasible and cost-effective method to improve building energy efficiency. The paper presents a comparative analysis of two different strategies to enhance the energy performance of an existing building, trying to enhance indoor comfort conditions too. On one hand, the strategy aimed at reducing the heat transfer by transmission, on the other one at decreasing the heat transfer by ventilation. The study has been applied to a school, located in the surroundings of Bologna, Italy. Potential energy savings were calculated by dynamic simulation using Trnsys © software. [ABSTRACT FROM AUTHOR]

• To encourage students to perform different activities and interact in open spaces. • To evaluate indoor and outdoor environment in the Faculty of Physical Education. • Significant temperature reductions of 1.5–4.9 K are achieved for indoor spaces. • Significant reduction in the PET for inner courtyard temperatures with an average temperature difference of 6.56 °C. • To increase tree density and courtyard shading (50% shading) inside courtyards. Thermal comfort in university buildings is a critical issue since a high level of indoor and outdoor thermal comfort is required to encourage students to perform different activities and interact. This study aims to evaluate and improve student thermal comfort inside educational building spaces. First, we evaluated the indoor and outdoor environment in the Faculty of Physical Education, Assiut University, located in a hot, arid climate, based on a field measurement and questionnaire survey. Then, the ENVI-met software was used to employ three design scenarios to achieve student thermal comfort enhancement inside the inner courtyard. The building used a passive strategy (trees, semi-shading 50%, hybrid between trees and semi-shading 50%) with courtyard ratios (H/W) equal to 1.6, and the measurements were conducted during the post-occupancy stage in the academic year. Furthermore, the questionnaire survey was studied using 116 responses from a thermal sensation vote (TSV) evaluation. The results suggested increasing vegetation and providing more shading increased thermal comfort by 34% and 28% in the inner courtyard, respectively. Significant temperature reductions of 1.5–4.9 K are achieved for indoor spaces due to the interaction between buildings and the climate factors such as air temperature (Ta) and relative humidity by using a passive strategy of hybrid between trees and semi-shading 50%. The base case concluded that high heat stress occurred inside the courtyard space with a physiological equivalent temperature (PET) above the acceptable range with similar results from TSV. In addition, there was a significant reduction in the PET for inner courtyard temperatures based on the ENVI-met model with an average temperature difference of 6.56 °C compared to the base case due to increased tree density and courtyard shading (50% shading). Thus, increasing shading and trees density inside courtyards of education buildings is recommended, especially with an H/W ratio of 1.6 on north-south axis orientation that could be applied in hot arid climate universities of Upper Egypt. [ABSTRACT FROM AUTHOR]

The necessity to reduce energy requirements of lighting systems should bring among engineers a more mature and conscious vision while planning and this means that their main goals should be the visual comfort and a decrease in energy and maintenance costs. Therefore this paper examines the case study of a classroom located in the Faculty of Engineering of Sapienza University in Rome and, to evaluate the possibility to combine natural and artificial light, it focuses on the realization of a new lighting system. This new solution is formed by LED sources and control systems able to modulate the power absorbed by every single lamp (to adapt, according to the hour and day, to the conditions of the natural light). In order to meet the EN 12464-1, the process of incorporating natural and artificial light must occur while guaranteeing both the minimum levels expected for the average illuminance E m [lx] and the uniformity coefficient ( E min / E m ) and this is the reason why a simultaneous exertion of artificial lights and a variable shielding system of glass surfaces activated through a motorized electronic control unit is required. Thanks to the software DIALux Evo 5.1 a 3D model of the classroom was reproduced and validated in order to simulate the combination of natural and artificial lighting and to verify if the results complied with the regulations. This was followed by an analysis concerning: the reliability of the system (through the examination of the MTBF – Mean Time Between Failure) and energetic and economic aspects (through the software ecoCALC 4.5.4). The novelty of this study is represented by the fact that in order to obtain the highest results in each of the three fields examined, it is possible to avoid the exertion of those negative feedback control-type systems (requiring high installation and maintenance costs due to sensor devices), thanks to pre-programmed logic control systems based on the data obtained with the simulations by prediction softwares. In this case study the solution suggested, while using a pre-programmed control logic, presents a MTBF of 1205 h (about twice of a negative feedback system solution) with payback periods that justify the higher costs presented by the electronic characterizing the control logic system with respect to traditional plant solutions. [ABSTRACT FROM AUTHOR]

This work is the result of a field study about overall comfort aspects performed in a secondary-school building during the winter season. The campaign aimed at describing the conditions of the school both from an objective and a subjective point of view, thus a questionnaire was administered to pupils during ongoing lessons. The monitored attributes concerned typical indoor quality aspects: acoustical, thermal, indoor air and visual quality. Weak points emerged from the data analysis and possible solutions are illustrated, focusing in particular on the acoustic aspects. [ABSTRACT FROM AUTHOR]

Ultra-fine dust refers to particulate matter from external sources, and modernization contributes toward increasing the presence of ultra-fine dust. Young children are particularly vulnerable to the ill effects of ultra-fine dust. Educational buildings, where young children spend the longest duration after their houses, are typically difficult to retrofit. Consequently, they are often used for a long time in the same state as they were when first completed. The buildings deteriorate due to long-term use, particularly because the openings are opened and closed frequently by occupants. Hence, architectural retrofits were performed during vacation, and the effects were evaluated. The evaluation factors include the temperature, relative humidity, and presence of ultra-fine dust. It was confirmed that the temperature and humidity inside the room decreased after the retrofit, while the airtightness performance was strengthened, thereby reducing the I/O ratio. To evaluate the sustainability of architectural remodeling with regard to not only the indoor air environment but also the enhancement of airtightness and insulation performance through the retrofit, a representative scenario was selected with reference to the Intergovernmental Panel on Climate Change's Future Climate Report. Although it was found that improving both the main entrance and the outdoor window was appropriate, replacing only the outdoor window was the adjudged the optimal retrofit scenario in consideration of the recovery of the investment cost. • Retrofit openings was carried out to control the I/O ratio. • Through retrofit I/O ratio has been reduced by up to 48.49%. • It is predicted that energy consumption will be reduced by 9.67% due to retrofit. • It takes maximum 21 years to payback the investment cost. [ABSTRACT FROM AUTHOR]

Nowadays, with the advancement of construction technology, saving energy use of the building has been considered as one of the fields of environmental sustainability. The use of green cover and plants on the buildings is a suitable passive method that leads to the replace reduced urban green spaces and reduce the problems caused by the effect of the urban heat island (UHI) and improve the energy performance of buildings by reducing the indoor temperature, regulating humidity and creating thermal insulation. The objective of this study was to evaluate the reduction of energy consumption through the green walls of an educational building in Gorgan (Csa climate). For this purpose, first, a building designed on the university site, and the effects of two methods of constructing green walls by using three different plant species were simulated. For comparison, thermal data of a building with a direct green facade in the same climate was measured as an experimental sample, and the results were used in the simulation. The effects of using two methods of a green wall with three different plant species (Sedum, Parthenocissus tricuspidata, Hedera helix) were simulated, which Sedum has had better results than other species. The final results showed that the latent heat of the building is increased by green walls and less significant heat is generated in the environment than buildings without green cover. Also, in buildings with living green walls, compared to buildings with direct green facades, the thermal transmittance of the wall is reduced due to more material layers. The results show that some types of green walls reduce heat loss, and finally the total energy use of the building. The green wall has a more efficient effect on the buildings with a simple form and the limited number of interior spaces. In general, depending on the variables of each building, the use of green walls with plant species is different. It is better for a building before design, to determine the most appropriate method of constructing a green wall and plant species to reduce the energy usage according to project factors and climate type of the region. [ABSTRACT FROM AUTHOR]

• A local energy benchmarking model was proposed. • National and local benchmarking models were applied in 12 actual buildings. • Similar classification was obtained for both national and local benchmarking models. • A process to support decision-making concerning energy benchmarking models is proposed. • It is suggested that local benchmarking models were applied only for detailed analysis. Energy benchmarking models assist the decision-making process for urban planners, architects, engineers and stakeholders. However, specialists diverge about the accuracy of benchmarking models used to represent a location. The aim of this study is to compare two benchmarking models of nursery schools: first, considering the national building stock for Brazil; second, using local data for a Southern Brazilian city. The national benchmarking model was obtained from a country-level benchmarking policy. The local benchmarking model was developed in four steps: data collection, archetype modeling, energy simulations and multiple linear regression to formulate the benchmark equation. Then, both national and local models were compared using data of 12 actual nursery schools, through their energy performance classification and an ANOVA test. Results showed that there are no significant differences between the models. Conclusion points out that a country-level benchmarking can adequately represent a city-level benchmarking; however, a late clustering process might be indicated for cases in locations with particularities apart from the national building stock. This paper provides an opportunity to discuss if benchmarking models, created with large stock-level analysis, indeed can evaluate the energy performance of specific buildings. The authors recommend future research comprising other regions to enhance applicability of the results. [ABSTRACT FROM AUTHOR]

Modern people spend most of their time indoors. Therefore, controlling indoor air quality is one of the most important factors for health. The indoor fine dust concentration is affected by the outdoor fine dust concentration. When the latter is high, it increases due to immersion. Therefore, improving the sealing performance of a building is an effective strategy to reduce indoor fine dust concentration during periods of severe outdoor fine dust without considering indoor fine dust generating factors. Traditional methods of improving the airtightness of a building have focused on replacing windows or doors. However, for reasons such as constructability and economic feasibility, more diverse technologies need to be considered. Due to this necessity, this study applied technologies such as sealing film, sealing lid, and padding to the educational building, and then derived the airtight performance through the blower door experiment, and measured the fine dust concentration to evaluate the effect. As a result of the experiment, it was analyzed that air leakage was reduced by up to 37% and fine dust by 22%. [Display omitted] • Factors of decreasing air-tightness of educational building in Korea were reviewed. • The building airtight performance of five classrooms was measured. • Technologies to improve airtightness were applied in the target building. • Airtightness performance increased and ACH 50 was reduced by up to 37%. • As a result of particle size analysis, the inflow of PM 2.5 was reduced by up to 22%. [ABSTRACT FROM AUTHOR]

Abstract: Calculating thermal load of building is important to find exact A/C equipment and air handling unit, to achieve comfort operation and good air distribution in the air-conditioned zone. It should be taking into account the highest temperature in summer and lower of winters that occurs in the region. Building's location, construction's materials other interior's load must be considered for estimation of accurate thermal load. All equations needed for heat transfer through the building and for the interior's load are used to get the thermal load. Then, all equations can be inserted in a personal program like M.S. Excel, to get the results. An alternative way is to use available software, like HAP 4.2 and Elit “RHVAC”. In this paper hand calculation is used to estimate the thermal load for existing building and the results were compared by the outcomes from HAP 4.2 program. It is shown that there is little difference between the two results due to defining the thermal resistance for the used materials of the wall, roof, and windows. Also the sources of constants used for the equations are not the same. [Copyright &y& Elsevier]

Abstract: Electric energy consumption is becoming an important issue for individual as well as for governmental and private organizations. Electric companies in Saudi Arabia face hard times in meeting the increasing demand for electric power. This research investigation is trying to study the electric energy consumption for an educational building located in Rabigh city, 150km North of Jeddah city, Saudi Arabia. Detailed auditing was done to the building including constructing materials used, energy consumption, cooling load and lighting. Temperature and relative humidity were recorded for several places inside the building. Thermal images for the interior zones were generated to provide information about the temperature distribution and give an idea about air or heat leak from or into the building. Based on the analysis of auditing exercise, some recommendations were suggested to reduce the electric energy consumptions which can reach up to 35.3%. The A/C units’ efficiency can also be increased by 31%. The results will be beneficial for the educational organizations and operation and maintenance staff to manage the electric usage and reduce the electric bills. [Copyright &y& Elsevier]

Sequential spaces are a spatial system comprising a series of single spaces connected in sequence by openings. They are not uncommon in the built environment and have similar acoustic characteristics with long and coupled spaces. To explore the sound attenuation and reverberation in such spaces, nine in situ measurements were conducted to examine the effects imposed by four factors, namely, opening on separating partition, number of separating partitions, source position, and acoustic absorption. Results revealed that if more rooms were connected by openings, although there could be significant changes in the distribution of sound pressure level, the resulting difference on the average level of each room is almost unobservable in the connecting rooms. In addition, T 20 of each room decreases in the source and receiving rooms, especially in the rooms with a larger source distance. According to the source position, the level difference between adjacent rooms by the first separating partition is larger than those in the successive ones when the space is divided by different number of separating partitions in the same construction. A larger source–opening distance results in a larger level difference across the spaces unless the source is placed along the opening. Increasing acoustic absorption in some rooms did not affect the level difference between those unincreased rooms. [ABSTRACT FROM AUTHOR]

• Incremental learning is rarely studied in load prediction articles. • The potential of incremental learning is evaluated. • A novel unsupervised incremental prediction model is proposed. • The proposed model is compared with five models. • The proposed model enriches the diversity of load prediction models. Load prediction is a crucial component for optimal building energy management. The challenge with buildings' load prediction is the lack of historical data since not all sites have a large amount of collected data. One of the solutions is incremental learning that updates the trained model with the most recent data. It allows the model to be deployed as soon as possible and improving its prediction accuracy as time progress. This study studies the effect of incremental learning. This study also proposes a novel DB-SOINN-R model that is based on the enhanced self-organizing incremental neural network (ESOINN), which is one of the incremental learning models. The problems with ESOINN are the inappropriate node removal by the original denoising of ESOINN, the inappropriate Euclidean distance for training data with imbalanced dimensions that usually consist of less discrete timestamp data compared to time-series historical data, and the incapability to obtain unique predicted outputs. To tackle these problems, the DB-SOINN-R incorporates a new density-based denoising that replaces the original denoising, a new mean Euclidean distance as the distance metric to handle training data with imbalanced dimensions, and k-nearest-neighbor inverse distance weighting (kNN-IDW) regression to obtain unique predicted output for every different input. The proposed DB-SOINN-R is compared with five models: feedforward neural network, deep neural network with long-short-term memory, support vector regression, ESOINN, and kNN regression. They are tested on day-ahead and one-hour-ahead load predictions, using two different datasets. The proposed DB-SOINN-R has the highest prediction accuracy among all models with incremental learning in both datasets. [ABSTRACT FROM AUTHOR]

• PCM application as a passive retrofit system on an educational building was analyzed. • PCM application to the shading system resulted in a cooling energy savings of 44%. • Different PCM melting temperatures did not affect the energy savings. • PCM shading applications are more economical than conventional shading. • Further research to reduce heating energy consumption is needed for shading. To improve the energy performance of existing buildings, retrofits may improve the energy efficiency or decrease energy demand. Many studies have been conducted on passive retrofit systems using phase change materials (PCMs). In this study, the objective is to analyze the energy performance and thermal comfort of an educational building with a shading system in South Korea. When PCMs were applied to the shading system, the cooling energy consumption decreased by 44%, and the number of hours of thermal comfort improved by 34%. The melting temperature of the PCM did not play a significant role in reducing the building energy consumption when the PCM was applied to the shading system. In the economic aspect, although the PCM shading application requires more additional investment, the simple payback period (SPP) of the PCM shading application is lower than the building without PCM. Therefore, the PCM shading application to A Hall is a successful retrofit that has more energy efficiency and thermal comfort, and has a lower SPP. However, the increase in heating energy consumption due to the shading application in winter could not be overcome. This implies that further research on the shading system in cold climates are required. [ABSTRACT FROM AUTHOR]