More Like this

1. Selective Sampling for Sensor Type Classification in Buildings

   https://doi.org/10.1109/IPSN48710.2020.00028  

   Ma, Jing; Hong, Dezhi; Wang, Hongning (April 2020, 2020 19th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN))

   A key barrier to applying any smart technology to a building is the requirement of locating and connecting to the necessary resources among the thousands of sensing and control points, i.e., the metadata mapping problem. Existing solutions depend on exhaustive manual annotation of sensor metadata - a laborious, costly, and hardly scalable process. To reduce the amount of manual effort required, this paper presents a multi-oracle selective sampling framework to leverage noisy labels from information sources with unknown reliability such as existing buildings, which we refer to as weak oracles, for metadata mapping. This framework involves an interactive process, where a small set of sensor instances are progressively selected and labeled for it to learn how to aggregate the noisy labels as well as to predict sensor types. Two key challenges arise in designing the framework, namely, weak oracle reliability estimation and instance selection for querying. To address the first challenge, we develop a clustering-based approach for weak oracle reliability estimation to capitalize on the observation that weak oracles perform differently in different groups of instances. For the second challenge, we propose a disagreement-based query selection strategy to combine the potential effect of a labeled instance on both reducing classifier uncertainty and improving the quality of label aggregation. We evaluate our solution on a large collection of real-world building sensor data from 5 buildings with more than 11, 000 sensors of 18 different types. The experiment results validate the effectiveness of our solution, which outperforms a set of state-of-the-art baselines. 

   more » « less
2. Color-based Lightweight Utility-aware Load Shedding for Real-Time Video Analytics at the Edge

   https://doi.org/10.1145/3629104.3666037  

   Gupta, Harshit; Saurez, Enrique; Röger, Henriette; Bhowmik, Sukanya; Ramachandran, Umakishore; Rothermel, Kurt (June 2024, ACM)

   Real-time video analytics typically require video frames to be processed by a query to identify objects or activities of interest while adhering to an end-to-end frame processing latency constraint. This imposes a continuous and heavy load on backend compute and network infrastructure. Video data has inherent redundancy and does not always contain an object of interest for a given query. We leverage this property of video streams to propose a lightweight Load Shedder that can be deployed on edge servers or on inexpensive edge devices co-located with cameras. The proposed Load Shedder uses pixel-level color-based features to calculate a utility score for each ingress video frame and a minimum utility threshold to select interesting frames to send for query processing. Dropping unnecessary frames enables the video analytics query in the backend to meet the end-to-end latency constraint with fewer compute and network resources. To guarantee a bounded end-to-end latency at runtime, we introduce a control loop that monitors the backend load and dynamically adjusts the utility threshold. Performance evaluations show that the proposed Load Shedder selects a large portion of frames containing each object of interest while meeting the end-to-end frame processing latency constraint. Furthermore, it does not impose a significant latency overhead when running on edge devices with modest compute resources. 

   more » « less
3. Construction Cost Decomposition of Residential Building Energy Retrofit

   https://doi.org/10.3390/buildings13061363  

   Zhao, Dong; Mo, Yunjeong (June 2023, Buildings)

   Buildings are responsible for significant energy consumption and carbon emissions. Green buildings, which incorporate advanced building technologies, offer a solution to reducing energy use. However, high costs associated with green building development present a barrier to widespread adoption. Retrofit projects, involving remodeling, renovation, and redevelopment of existing buildings, offer a viable solution. While prior studies have examined the cost analysis of green and non-green buildings, there is a lack of evidence comparing new and retrofit projects. This study aims to address this gap by providing empirical evidence for the cost decomposition and benefits of new and retrofit projects. Data on energy use, building technology, and costs from 235 certified green homes in the United States were collected, and cost benefits were evaluated. Results show that retrofit projects cost, on average, $1270.5/m2 ($118.0/ft2), which is 30% less than new projects. Land acquisition and development account for 35% of retrofit costs, six times greater than new projects. Excluding land costs, retrofit projects cost, on average, $733.88/m2 ($68.2/ft2), 49% less than new projects. Retrofit projects use similar building technologies as new projects and produce larger energy savings. The cost-benefit values generated by retrofit projects are 86% greater than new projects when considering land costs and 142% greater without considering land costs. These findings contribute to cost management for complex building projects and energy policy for sustainable development. Retrofitting offers great potential to promote the green building movement and suggests effective subsidy programs as a public policy implication. 

   more » « less
4. The Impact of Urban Heat Island on Calibrated Building Energy Model Predictions

   Jahani, Elham; Vanage, Soham; Cetin, Kristen (January 2020, ASCE Construction Research Congress)

   Among various elements of urban infrastructure, there is significant opportunity to improve existing buildings’ sustainability, considering that approximately 40% of the total primary energy consumption and 72% of electricity consumption in United States is consumed by the building sector. Many different efforts focus on reducing the energy consumption of residential buildings. Data-validated building energy modeling methods serve the role of supporting this effort, by enabling the identification of the potential savings associated with different potential retrofit strategies. However there are many uncertainties that can impact the accuracy of energy model results, one of which is the weather input data. Measured weather data inputs located at each building can help address this concern, however, weather station data collection for each building is also costly and typically not feasible. Some weather station data is already collected, however, these are generally located at airports rather than near buildings, and thus do not capture local, spatially-varying weather conditions which are documented to occur, particularly in urban areas. In this study we address the impact of spatial temperature differences on residential building energy use. An energy model was developed in EnergyPlus for a residential building located in Mueller neighborhood of Austin, TX, and was validated using actual hourly measured electricity consumption. Using the validated model, the impact of measured spatial temperature differences on building energy consumption were investigated using multiple weather stations located throughout the urban area with different urban fractions. The results indicate that energy consumption of a residential building in a city with a 10% higher urban fraction would increase by approximately 10%. This variation in energy consumption is likely due to the impact of UHI effects occurring in urban areas with high densities. 

   more » « less
5. Sensitivity of Building Energy Model Predictions to Spatial Variations in Climate under Different Levels of Urbanization

   Jahani, Elham; Vanage, Soham; Cetin, Kristen; Gallus, William; Jahn, David (January 2020, ASHRAE Annual Virtual Conference 2020)

   null (Ed.)

   In the United States, approximately 40% of the primary energy use and 72% of the electricity use belong to the building sector. This shows the significance of studying the potential for reducing the building energy consumption and buildings’ sustainability for ensuring a sustainable development. Therefore, many different efforts focus on reducing the energy consumption of residential buildings. Data-validated building energy modeling methods are among the studies for such an effort, particularly, by enabling the identification of the potential savings associated with different potential retrofit strategies. However, there are many uncertainties that can impact the accuracy of such energy model results, one of which is the weather input data. In this study, to investigate the impact of spatial temperature variation on building energy consumption, six weather stations in an urban area with various urban density were selected. A validated energy model was developed using energy audit data and high-frequency electricity consumption of a residential building in Austin, TX. The energy consumption of the modeled building was compared using the selected six weather datasets. The results show that energy use of a building in an urban area can be impacted by up to 12% due to differences in urban density. This indicates the importance of weather data in predicting energy consumption of the building. The methodology and results of this study can be used by planners and decision makers to reduce uncertainties in estimating the building energy use in urban scale. 

   more » « less