Showing papers on "Occupancy published in 2021"

Building occupancy and energy consumption: Case studies across building types

Abstract: Past research has shown that occupancy information can be used to reduce building energy consumption through occupant-based controls and by mitigating wasteful occupant behavior. In this study, we investigate the dynamic relationship between WiFi connection counts (as a proxy to occupancy) and building electricity consumption across four building typologies (office, lab, health center, and library). Our findings based on one year of data show a strong positive linear correlation between electricity consumption and WiFi count across all four building when the building is in operation. The data exploration also indicates higher interactions between occupants with the plug and lighting loads in office and lab space types as compared to in a health center and a library. Next, using principal component analysis (PCA) for feature extraction followed by Density-based spatial clustering of applications with noise (DBSCAN), we show that distinct clusters could be generated, characterized by an increase in the between-cluster variance and smaller within-cluster variation. Lastly, we apply linear regression to manifest how the clustering results can be used to better model the variables.

Data-Driven Occupancy Detection Hybrid Model Using Particle Swarm Optimization Based Artificial Neural Network

Abstract: The occupancy level of the room in a building is responsible for consumption of electrical power within the building. The occupancy level in a room depend on several controllable and/or uncontrollable parameters within and/or outside environment of the building. So, for the optimal demand forecasting and planning within the building, occupancy detection play an important role. In this study, the occupancy is determined by using simple measureable parameters of the inside environment of the building such as light (in lux), temperature (in Celsius), relative humidity (in %), CO2 level (in ppm) and humidity ratio (in kg water-vapor/kg-air). The data-driven occupancy detection using particle swarm optimization (PSO) based artificial neural network (ANN) is designed in R language and proposed approach is validated with different seventeen models by using the measured dataset. The occupancy detection for these models are 77.9–98.95% for ANN models and 87.8–99.5% for PSO-ANN models, which shows that PSO based ANN model’s performance is more acceptable in comparison to only ANN models.

Using the ecological significance of animal vocalizations to improve inference in acoustic monitoring programs

Abstract: Recent bioacoustic advances have facilitated large-scale population monitoring for acoustically active species. Animal sounds, however, can of information that is underutilized in typical approaches to passive acoustic monitoring (PAM) that treat sounds simply as detections. We developed 3 methods of extracting additional ecological detail from acoustic data that are applicable to a broad range of acoustically active species. We conducted landscape-scale passive acoustic surveys of a declining owl species and an invasive congeneric competitor in California. We then used sex-specific vocalization frequency to inform multistate occupancy models; call rates at occupied sites to characterize interactions with interspecific competitors and assess habitat quality; and a flexible multivariate approach to differentiate individuals based on vocal characteristics. The multistate occupancy models yielded novel estimates of breeding status occupancy rates that were more robust to false detections and captured known habitat associations more consistently than single-state occupancy models agnostic to sex. Call rate was related to the presence of a competitor but not habitat quality and thus could constitute a useful behavioral metric for interactions that are challenging to detect in an occupancy framework. Quantifying multivariate distance between groups of vocalizations provided a novel quantitative means of discriminating individuals with ≥20 vocalizations and a flexible tool for balancing type I and II errors. Therefore, it appears possible to estimate site turnover and demographic rates, rather than just occupancy metrics, in PAM programs. Our methods can be applied individually or in concert and are likely generalizable to many acoustically active species. As such, they are opportunities to improve inferences from PAM data and thus benefit conservation.

Occupancy Estimation Using Thermal Imaging Sensors and Machine Learning Algorithms

Abstract: Occupancy estimation has a broad range of applications in security, surveillance, traffic and resource management in smart building environments. Low-resolution thermal imaging sensors can be used for real-time non-intrusive occupancy estimation. Such sensors have a resolution that is too low to identify occupants, but it may provide sufficient data for real-time occupancy estimation. In this paper, we present a systematic study of three thermal imaging sensors with different resolutions, with a focus on sensor characterization, estimation algorithms, and comparative analysis of occupancy estimation performance. A unified processing algorithms pipeline for occupancy estimation is presented and the performance of three sensors are compared side-by-side. A number of specific algorithms are proposed for pre-processing of sensor data, feature extraction, and fine-tuning of the occupancy estimation algorithms. Our results show that it is possible to achieve about 99% accuracy for occupancy estimation with our proposed approach, which might be sufficient for many practical smart building applications.

Review of developments in whole-building statistical energy consumption models for commercial buildings

Abstract: A significant portion of energy consumption occurs in buildings today. Accurate and easy-to-implement methods are needed to calculate building energy consumption for a wide range of applications. These areas have attracted research interest as early as the 1980's. Among a number of approaches for building energy analysis, the statistical methods have remained popular because they are simple to use and able to provide accurate prediction of building energy consumption. As the availability and quality of building energy data continue to improve, the methodologies behind building energy calculation also evolved over time. Although relevant areas such as calibrated simulation and machine learning methods have had numerous recent literature reviews, the statistical methods have not been reviewed in depth. This work aims to fill this knowledge gap for whole-building energy consumption modelling. This work will discuss how the methodology developed through time and summarise the applications of this approach in various areas of building energy analysis. This work has identified that the statistical methods have rarely been applied to model the electric demand, power factor, or domestic water use. The use of an occupancy variable and novel model forms are also areas with limited literature.

Extracting typical occupancy schedules from social media (TOSSM) and its integration with building energy modeling

Abstract: Building occupancy, one of the most important consequences of occupant behaviors, is a driving influencer for building energy consumption and has been receiving increasing attention in the building energy modeling community. With the vast development of information technologies in the era of the internet-of-things, occupant sensing and data acquisition are not limited to a single node or traditional approaches. The prevalence of social networks provides a myriad of publically available social media data that might contain occupancy information in the space for a given time. In this paper, we explore two approaches to extract the typical occupancy schedules for the input to the building energy simulation based on the data from social networks. The first approach uses text classification algorithms to identify whether people are present in the space where they are posting on social media. On top of that, the typical building occupancy schedules are extracted with assumed people counting rules. The second approach utilizes the processed Global Positioning System (GPS) tracking data provided by social networking service companies such as Facebook and Google Maps. Web scraping techniques are used to obtain and post-process the raw data to extract the typical building occupancy schedules. The results show that the extracted building occupancy schedules from different data sources (Twitter, Facebook, and Google Maps) share a similar trend but are slightly distinct from each other and hence may require further validation and corrections. To further demonstrate the application of the extracted Typical Occupancy Schedules from Social Media (TOSSM), data-driven models for predicting hourly energy usage prediction of a university museum are developed with the integration of TOSSM. The results indicate that the incorporation of TOSSM could improve the hourly energy usage prediction accuracy to a small extent regarding the four adopted evaluation metrics for this museum building.

Modeling COVID-19 hospital admissions and occupancy in the Netherlands

Abstract: We describe the models we built for hospital admissions and occupancy of COVID-19 patients in the Netherlands. These models were used to make short-term decisions about transfers of patients between regions and for long-term policy making. We motivate and describe the model we used for predicting admissions and how we use this to make predictions on occupancy.

Transfer Learning Approach for Occupancy Prediction in Smart Buildings

Abstract: Accurate occupancy prediction in smart buildings is a key element to reduce building energy consumption and control HVAC systems (Heating – Ventilation and– Air Conditioning) efficiently, resulting in an increment of human comfort. This work focuses on the problem of occupancy prediction modelling (occupied / unoccupied) in smart buildings using environmental sensor data. A novel transfer learning approach was used to enhance occupancy prediction accuracy when the amounts of historical training data are limited. The proposed approach and models are applied to a case study of three office rooms in an educational building. The data sets used in this work are actual data collected from the Urban Sciences Building (USB) in Newcastle University. The results of the proposed transfer learning approach have been compared with the models from Support Vector Machine and Random Forest algorithms. The final results demonstrate that the most accurate model in this study to predict occupancy status was produced by stacked Long-Short-Term-Memory with a transfer learning framework.

Sequence-to-sequence deep learning model for building energy consumption prediction with dynamic simulation modeling

Abstract: This study used a more detailed dataset approach to address the limitations of existing building energy prediction methods and to predict building energy demand more accurately. The EnergyPlus dynamic simulation program closely modeled building envelope performance, zone division, and heating, ventilation, and air conditioning systems. Building energy simulation used actual weather data and generated occupancy data. The occupancy, lighting, and equipment schedules for each zone were generated in 5-min intervals using the Lawrence Berkeley National Laboratory occupancy simulator. Summer electric power consumption based on Internet of Things information from the testbed building validated the model. When applying generated occupancy, lighting, and equipment schedules, the mean bias error of simulation similarly improved to 4.73%, and the coefficient of variation of the root-mean-squared error (Cv(RMSE]) improved to 12.26%. Subsequently, a demand prediction model was constructed as a sequence-to-sequence (seq2seq) model using long short-term memory (LSTM) cells in recurrent neural network (RNN) algorithms, then its accuracy was evaluated. In the seq2seq model, the learning performance based on the EnergyPlus data exhibited an RMSE of 4.48% and a weighted average percentage error of 3.07%. As a result of applying prediction methods while changing climate scenarios, prediction performance also satisfied statistically significant levels. The occupancy information and solar radiation were determined to exert the greatest influence on the building energy demand prediction.

A Review on Numerical Approach to Achieve Building Energy Efficiency for Energy, Economy and Environment (3E) Benefit

Abstract: Increasing energy demand in buildings with a 40% global share and 30% greenhouse gas emissions has accounted for climate change and a consequent crisis encouraging improvement of building energy efficiency to achieve the combined benefit of energy, economy, and environment. For an efficient system, the optimization of different design control strategies such as building space load, occupancy, lighting, and HVAC becomes inevitable. Therefore, interdisciplinary teamwork of developers, designers, architects, and consumers to deliver a high-performance building becomes essential. This review aims to endorse the importance of Building Performance Simulation in the pre-design phase along with the challenges faced during its adaptation to implementation. A morphology chart is structured to showcase the improvement in Building Energy Efficiency by implementing Building Performance Simulation for different building energy systems and by implementing various energy efficiency strategies to achieve the 3E benefit. As a developing nation, India still lacks mass application of Building Performance Simulation tools for improving Building Energy Efficiency due to improper channelizing or implementation; thus, this framework will enable the designers, architects, researchers to contemplate variable building energy optimization scenarios.

Quantifying the impacts of bushfire on populations of wild koalas (Phascolarctos cinereus): Insights from the 2019/20 fire season

Abstract: The impact of bushfire events on wild Koala (Phascolarctos cinereus) populations is poorly understood. Following the 2019/2020 bushfire season in eastern Australia, we resurveyed 123 field sites for which contemporaneous (current koala generation) pre‐fire survey data were available. Field sites were distributed across six fire grounds between Foster and Ballina on the north coast of New South Wales. At these sites, pre‐fire naive occupancy levels by koalas ranged from 25% to 71% of the sampled habitat, while post‐fire naive occupancy levels ranged from 0% to 47%. The median reduction in the naive occupancy rate by koalas when considered across all six fire grounds was 71% when standardized against pre‐fire occupancy levels. Field data provided strong corroboration between site‐based, post‐fire foliage canopy cover estimates and modelled Google Earth Engine Burnt Area Map (GEEBAM) fire‐severity categories. In terms of GEEBAM fire‐severity categories, koala survival was five times more likely in areas where forest canopies were modelled as Unburnt or Partially burnt, compared to areas where forest canopies were Fully burnt. The capacity of bushfire‐affected koala populations to recover from the 2019/20 fire season will be conditional upon size of the original population in each fire‐affected area, the enactment and implementation of supportive, recovery‐themed management regimes, future inter‐fire intervals and associated intensities. Management actions necessary to assist recovery actions are discussed.