More Like this

1. Spatiotemporal Variations in Ozone and Carbon Dioxide Concentrations in an HVAC System of a LEED-Certified Office Building

   https://doi.org/10.1016/j.buildenv.2025.112651  

   Jiang, Jinglin; Huang, Junkai; Jung, Nusrat; Boor, Brandon E (April 2025, Building and Environment)

   Indoor air quality (IAQ) is crucial for the health, well-being, and productivity of office occupants. IAQ is strongly influenced by occupancy and the operational mode of the heating, ventilation, and air conditioning (HVAC) system. This study investigates the spatiotemporal variations in ozone (O3) and carbon dioxide (CO2) concentrations throughout the HVAC system of a LEED-certified office building. A four-month field measurement campaign was conducted at the Ray W. Herrick Laboratories, employing an automated multi-point sampling system to monitor O3 and CO2 at eight locations throughout the HVAC system. The objectives of this study are to characterize the spatiotemporal distribution of these gases under different ventilation modes and occupancy levels, and to identify O3 loss mechanisms in the office and its HVAC system. Spatiotemporal variations in O3 and CO2 concentrations were observed throughout the HVAC system. Results indicate that outdoor air exchange rates (AERs) significantly impact indoor O3 levels, with higher AERs resulting in increased indoor O3 but reduced CO2 concentrations. Measurements reveal that HVAC filters and ducts contribute to O3 loss, with up to 18% O3 removal observed in the longest HVAC duct segment. Additionally, occupancy influences O3 deposition onto human skin and clothing surfaces. This research underscores the limitations of ventilation standards that focus only on CO2, highlighting the need for ventilation strategies that consider the effects of occupancy and outdoor AERs on different gases. By integrating multi-point gas sampling into building automation systems, more effective control strategies can be developed to enhance IAQ and occupant health while reducing energy consumption. 

   more » « less
2. Novel Physics-Based Machine-Learning Models for Indoor Air Quality Approximations

   Mohammadshirazi, Ahmad; Nadafian, Aida; Monsefi, Amin Karimi; Rafiei, Mohammad; Ramnath, Rajiv (August 2023, https://doi.org/10.48550/arXiv.2308.01438)

   Cost-effective sensors are capable of real-time capturing a variety of air quality-related modalities from different pollutant concentrations to indoor/outdoor humidity and temperature. Machine learning (ML) models are capable of performing air-quality "ahead-of-time" approximations. Undoubtedly, accurate indoor air quality approximation significantly helps provide a healthy indoor environment, optimize associated energy consumption, and offer human comfort. However, it is crucial to design an ML architecture to capture the domain knowledge, so-called problem physics. In this study, we propose six novel physics-based ML models for accurate indoor pollutant concentration approximations. The proposed models include an adroit combination of state-space concepts in physics, Gated Recurrent Units, and Decomposition techniques. The proposed models were illustrated using data collected from five offices in a commercial building in California. The proposed models are shown to be less complex, computationally more efficient, and more accurate than similar state-of-the-art transformer-based models. The superiority of the proposed models is due to their relatively light architecture (computational efficiency) and, more importantly, their ability to capture the underlying highly nonlinear patterns embedded in the often contaminated sensor-collected indoor air quality temporal data. 

   more » « less
3. A Tree-Structured Neural Network Model for Household Energy Breakdown

   https://doi.org/10.1145/3308558.3313405  

   Jia, Yiling; Batra, Nipun; Wang, Hongning; Whitehouse, Kamin (May 2019, The World Wide Web Conference)

   Residential buildings constitute roughly one-fourth of the total energy use across the globe. Numerous studies have shown that providing an energy breakdown increases residents' awareness of energy use and can help save up to 15% energy. A significant amount of prior work has looked into source-separation techniques collectively called non-intrusive load monitoring (NILM), and most prior NILM research has leveraged high-frequency household aggregate data for energy breakdown. However, in practice most smart meters only sample hourly or once every 15 minutes, and existing NILM techniques show poor performance at such a low sampling rate. In this paper, we propose a TreeCNN model for energy breakdown on low frequency data. There are three key insights behind the design of our model: i) households consume energy with regular temporal patterns, which can be well captured by filters learned in CNNs; ii) tree structure isolates the pattern learning of each appliance that helps avoid magnitude variance problem, while preserves relationship among appliances; iii) tree structure enables the separation of known appliance from unknown ones, which de-noises the input time series for better appliance-level reconstruction. Our TreeCNN model outperformed seven existing baselines on a public benchmark dataset with lower estimation error and higher accuracy on detecting the active states of appliances. 

   more » « less
4. Learning from Non-Experts: An Interactive and Adaptive Learning Approach for Appliance Recognition in Smart Homes

   https://doi.org/10.1145/3491241  

   Codispoti, Jackson; Khamesi, Atieh R.; Penn, Nelson; Silvestri, Simone; Shin, Eura (February 2022, ACM Transactions on Cyber-Physical Systems)

   With the acceleration of ICT technologies and the Internet of Things (IoT) paradigm, smart residential environments , also known as smart homes are becoming increasingly common. These environments have significant potential for the development of intelligent energy management systems, and have therefore attracted significant attention from both academia and industry. An enabling building block for these systems is the ability of obtaining energy consumption at the appliance-level. This information is usually inferred from electric signals data (e.g., current) collected by a smart meter or a smart outlet, a problem known as appliance recognition . Several previous approaches for appliance recognition have proposed load disaggregation techniques for smart meter data. However, these approaches are often very inaccurate for low consumption and multi-state appliances. Recently, Machine Learning (ML) techniques have been proposed for appliance recognition. These approaches are mainly based on passive MLs, thus requiring pre-labeled data to be trained. This makes such approaches unable to rapidly adapt to the constantly changing availability and heterogeneity of appliances on the market. In a home setting scenario, it is natural to consider the involvement of users in the labeling process, as appliances’ electric signatures are collected. This type of learning falls into the category of Stream-based Active Learning (SAL). SAL has been mainly investigated assuming the presence of an expert , always available and willing to label the collected samples. Nevertheless, a home user may lack such availability, and in general present a more erratic and user-dependent behavior. In this paper, we develop a SAL algorithm, called K -Active-Neighbors (KAN), for the problem of household appliance recognition. Differently from previous approaches, KAN jointly learns the user behavior and the appliance signatures. KAN dynamically adjusts the querying strategy to increase accuracy by considering the user availability as well as the quality of the collected signatures. Such quality is defined as a combination of informativeness , representativeness , and confidence score of the signature compared to the current knowledge. To test KAN versus state-of-the-art approaches, we use real appliance data collected by a low-cost Arduino-based smart outlet as well as the ECO smart home dataset. Furthermore, we use a real dataset to model user behavior. Results show that KAN is able to achieve high accuracy with minimal data, i.e., signatures of short length and collected at low frequency. 

   more » « less
5. Data-driven modeling of dynamic occupant thermostat override behavior for demand response applications

   Sharma, Kunind (April 2025, Northeastern University)

   Buildings consume nearly 40% of global energy and produce similar emissions. Whiletechnological advances address efficiency, occupant behavior causes energy use variations up to 300% between identical buildings. This gap between predicted and actual building performance impacts building design, operations, and grid demand management programs. Through analyses of smart thermostat data from 1,400 single-occupant homes, the researchdemonstrates that occupants respond to 8°F thermostat setpoint changes within a median of 15 minutes, while 2°F changes trigger responses within a median of 30 minutes. This highlights an understudied temporal relationship between thermostat setbacks and response time of occupant behaviors. Models of such behavior dynamics are required to incorporate occupant impacts into building performance simulation. A key contribution of this dissertation is the Thermal Frustration Theory (TFT), which positsthat thermal discomfort driven behaviors are caused by the time-accumulation of discomfort, not simply a temperature deviation threshold or a delay from an initiating event. Using a dataset of 634 thermostats, each with 25+ manual setpoint changes, a comparative analysis of TFT and comfort zone and a delayed response theories demonstrated that personalized TFT models better predict when manual setpoint change occur. This was measured by the area under the curve statistical measure (AUC); all three models perform similarly by a Matthews Correlation Coefficient measure. Higher AUC performance is especially important for modeling occupant behavior in demand response programs where false negatives of rare occupant interactions could adversely affect grid stability. EnergyPlus based simulations were conducted with TFT-derived occupant models, demonstrating the ability to identify parameters of known TFT models from only data observable with smart thermostats, even under the presence of noise from routine overrides. Overall, the dissertation highlights that thermostat interactions are neither static,instantaneous, nor driven solely by the environment. Instead, temporal accumulation of discomfort and routine-based behavior play important roles. The methodology and results offer a pathway towards more accurate modeling of human-building interactions for policy assessment, building design, and demand response programs. 

   more » « less