An official website of the United States government
Buildings in the U.S. are responsible for approximately 40% of energy and 70% of the electricity consumption. To address rising greenhouse gas emissions and climate changes, various studies have explored strategies to reduce energy consumption in buildings. One opportunity to improve the building envelope performance is through improvements to fenestrations, particularly complex multi-layer fenestration systems for exterior windows. Windows are the least thermally efficient of all components in a typical building envelope. Windows also permit solar radiation into a building, which significantly increases the building energy consumption during the summer season. Meanwhile, windows are necessary to provide occupants with natural light, a view to the outside, and to support productivity. Thus, there is a need to strike a balance between energy savings, and the thermal and visual comfort impacted by windows. Traditionally, shading devices are one method used to adjust the amount of heat and light entering an interior space. However, such shading devices are typically operated manually by occupants, and are seldom used effectively over time. Currently the building energy simulation program EnergyPlus, has limited capabilities to model shading devices, and more limited abilities to model dynamic fenestrations. In this study, thus, we propose to model and validate several types of automated multi-layer fenestration elements, using co-simulation of EnergyPlus and Radiance using laboratory-collected data. EnergyPlus was used to model energy consumption and thermal comfort while Radiance was used to model lighting levels. BCVTB was used to interface between EnergyPlus and Radiance to facilitate co-simulation. To validate the models, experimental data was collected from 5 illuminance sensors in an exterior office space located in a test facility in Ankeny, IA. This model methodology can be used to improve the flexibility and modeling capabilities of dynamic fenestration elements for building energy performance evaluation methods.
more »
« less
Improving Building Energy Efficiency through Data Analysis
For many lawmakers, energy-efficient buildings have been the main focus in large cities across the United States. Buildings consume the largest amount of energy and produce the highest amounts of greenhouse emissions. This is especially true for New York City (NYC)’s public and private buildings, which alone emit more than two-thirds of the city’s total greenhouse emissions. Therefore, improvements in building energy efficiency have become an essential target to reduce the amount of greenhouse gas emissions and fossil fuel consumption. NYC’s buildings’ historical energy consumption data was used in machine learning models to determine their ENERGY STAR scores for time series analysis and future pre- diction. Machine learning models were used to predict future energy use and answer the question of how to incorporate machine learning for effective decision-making to optimize energy usage within the largest buildings in a city. The results show that grouping buildings by property type, rather than by location, provides better predictions for ENERGY STAR scores.
more »
« less
- PAR ID:
- 10440682
- Date Published:
- Journal Name:
- The 14th ACM International Conference on Future Energy Systems (e-Energy ’23 Companion)
- Page Range / eLocation ID:
- 1 to 7
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Heating buildings using fossil fuels such as natural gas, propane and oil makes up a significant proportion of the aggregate carbon emissions every year. Because of this, there is a strong interest in decarbonizing residential heating systems using new technologies such as electric heat pumps. In this paper, we conduct a data-driven optimization study to analyze the potential of replacing gas heating with electric heat pumps to reduce CO 2 emission in a city-wide distribution grid. We conduct an in-depth analysis of gas consumption in the city and the resulting carbon emissions. We then present a flexible multi-objective optimization (MOO) framework that optimizes carbon emission reduction while also maximizing other aspects of the energy transition such as carbon-efficiency, and minimizing energy inefficiency in buildings. Our results show that replacing gas with electric heat pumps has the potential to cut carbon emissions by up to 81%. We also show that optimizing for other aspects such as carbon-efficiency and energy inefficiency introduces tradeoffs with carbon emission reduction that must be considered during transition. Finally, we present a detailed analysis of the implication of proposed transition strategies on the household energy consumption and utility bills, electric grid upgrades, and decarbonization policies. We compute the additional energy demand from electric heat pumps at the household as well as the transformer level and discuss how our results can inform decarbonization policies at city scale.more » « less
-
Lighting is a major component of energy consumption in controlled environment agriculture (CEA) operations. Skyscraper farms (multilevel production in buildings with transparent glazing) have been proposed as alternatives to greenhouse or plant factories (opaque warehouses) to increase space-use efficiency while accessing some natural light. However, there are no previous models on natural light availability and distribution in skyscraper farms. This study employed climate-based daylight modeling software and the Typical Meteorological Year (TMY) dataset to investigate the effects of building geometry and context shading on the availability and spatial distribution of natural light in skyscraper farms in Los Angeles (LA) and New York City (NYC). Electric energy consumption for supplemental lighting in 20-storey skyscraper farms to reach a daily light integral target was calculated using simulation results. Natural lighting in our baseline skyscraper farms without surrounding buildings provides 13% and 15% of the light required to meet a target of 17 mol·m−2·day−1. More elongated buildings may meet up to 27% of the lighting requirements with natural light. However, shading from surrounding buildings can reduce available natural light considerably; in the worst case, natural light only supplies 5% of the lighting requirements. Overall, skyscraper farms require between 4 to 11 times more input for lighting than greenhouses per crop canopy area in the same location. We conclude that the accessibility of natural light in skyscraper farms in dense urban settings provides little advantage over plant factories.more » « less
-
Communities are considering local food production in response to the pressing need to reduce food system greenhouse gas (GHG) emissions. However, local food systems can vary considerably in design and operation, including controlled environment agriculture (CEA), which refers to agricultural production that takes place within an enclosed space where environmental conditions, such as temperature, humidity, and light, are precisely controlled. Such systems require a considerable amount of energy and thus emissions; therefore, this study seeks to quantify these environmental impacts to determine how local CEA systems compare to alternative systems. For this study’s methods, we apply life cycle assessment methodology to quantify the cradle-to-storeshelf GHG emissions and water consumption of four lettuce production systems: local indoor plant factory, local greenhouse, local seasonal soil, and conventional centralized production in California with transportation. Using geographically specific inputs, the study estimates the environmental impact of the different production systems including geospatially resolved growth modeling, emissions intensity, and transportation distances. The results include the major finding that baseline CEA systems always have higher GHG emissions (2.6–7.7 kg CO2e kg−1) than centralized production (0.3–1.0 kg CO2e kg−1), though water consumption is significantly less owing to hydroponic efficiency. In contrast, local seasonal soil production generally has a lower GHG impact than centralized production, though water consumption varies by crop yield and local precipitation during growing seasons. Scenario analyses indicate CEA facilities would need to electrify all systems and utilize low-carbon electricity sources to have equivalent or lower GHG impacts than California centralized production plus transportation. We conclude that these results can inform consumers and policy makers that local seasonal production and conventional supply chains are more sustainable than local CEA production in near-term food-energy-water sustainability nexus decision making.more » « less
-
A novel green granular neural network (GGNN) with new fast software-FPGA co-designed learning is developed to reduce both CO2 emissions and energy consumption more effectively than popular neural networks with the traditional software-CPU-GPU-based learning. Different from traditional tedious CPU-GPU-based training algorithms using gradient descent methods and other methods such as genetic algorithms , the software-FPGA co-designed training algorithm may quickly solve a system of linear equations to directly calculate optimal values of hyperparameters of the GGNN. Initial simulation results indicates that the FPGA equation solver code ran faster than the Python equation solver code. Therefore, implementing the GGNN with software-FPGA co-designed learning is feasible. In addition, the shallow high-speed GGNN is explainable because it can generate interpretable granular If-Then rules. In the future, The GGNN will be evaluated by comparing with other machine learning models with traditional software-based learning in terms of speeds, model sizes, accuracy, CO2 emissions and energy consumption by using popular datasets. New algorithms will be created to divide the inputs to different input groups that will be used to build different small-size GGNNs to solve the curse of dimensionality. Additionally, the explainable green granular convolutional neural network will be developed by using the GGNNs as basic building blocks to efficiently solve image recognition problems.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3599733.3600244
