An official website of the United States government
The state of California is the foremost leader in solar photovoltaics (PV) installations in the United States. With 1,390,240 installations and 24.76% of the state's energy coming from solar, the demand for PV modules is steadily increasing. Most PV modules have an expected lifetime of 25-30 years. However, due to repowering or early module failure, module lifetime can often be shorter than anticipated. Current studies calculate the environmental impact of PV systems based on ideal installation conditions and a full 25-year module lifetime. This study considers the impact on the life cycle of PV systems from early PV module retirement and actual system installation in California. Using the life cycle cumulative energy demand, electricity data from the Energy Information Administration (EIA), and greenhouse gases, carbon payback time (CPBT) was evaluated. Data from various PV module rooftop residential installations in 2019 were collected from the California NEM database. Information on the system design (tilt, azimuth, module model) and module specification sheets were used to calculate the cumulative electricity generated in kilowatt-hours (kWh) over the system' lifetime. The calculated average CPBT was 2.8 years, shorter than most of the system lifetimes, and the mean number of zero carbon years experienced by earlier retired systems was about 5 years. Although the rapid movement towards solar energy is promising and essential as reliance on greener energy increases, attention must be paid to the diverse lifespans of PV modules, system design, and performance to substantiate or reject the assumption that PV always have a positive impact on the environment.
more »
« less
This content will become publicly available on June 30, 2026
Enabling Automatic Solar PV Array Identification using Big Satellite Imagery
Recently, there has been a growing interest in automatically collecting distributed solar photovoltaic (PV) installation information in smart grid systems, including the quantity and locations of solar PV deployments, as well as their profiling information across a given geospatial region. Most recent approaches are still suffering low detection accuracy due to insufficient sample and principal feature learning when building their models and also separation of rooftop object segmentation and identification during their detection processes. In addition, they cannot report accurate multi-deployment results. To address these problems, we design a new system-SolarDetector+, which can automatically and accurately detect and profile distributed solar PV arrays without any extra cost. In essence, SolarDetector+first leverages multiple data augmentation techniques, including Cycle-Consistent Adversarial Networks, Latent Diffusion Models, and Generative Adversarial networks, to build a large rooftop satellite imagery dataset (RSID). Then, SolarDetector+employs Mask R-convolutional neural networks algorithm to accurately identify rooftop solar PV arrays and learn the detailed installation information for each solar PV array simultaneously. We find that pre-trained SolarDetector+yields an average Matthews correlation coefficient of 0.862 to detect solar PV arrays over RSID, which is ∼50% better than the most recent open source detection system—SolarFinder.
more »
« less
- Award ID(s):
- 2238701
- PAR ID:
- 10596248
- Publisher / Repository:
- ACM
- Date Published:
- Journal Name:
- ACM Journal on Computing and Sustainable Societies
- Volume:
- 3
- Issue:
- 2
- ISSN:
- 2834-5533
- Page Range / eLocation ID:
- 1 to 28
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Much has been written on the rooftop solar photovoltaic (PV) adoption in the U.S., but granular economic assessment at large scale is missing. We provide household level PV economic assessment for a medium size city in North Central Florida, and analyze the economic viability of these installations. Results show that a large number of households will not benefit from solar installations. Further, economic viability is heavily reliant on incentives whose future is uncertain at best. Our analysis did not reveal significant variations in economic viability across different household values --- a proxy we used to differentiate household wealth. Yet, building permits and installation locations indicate economically disadvantaged communities have much lower installation rates as has been the main conclusion in the earlier literature. We argue economic assessment for PV should extend beyond simple benefit--cost analysis. A more nuanced approach should be taken in PV feasibility assessment, and structuring incentive schemes.more » « less
-
The use of photovoltaic (PV) arrays as a source of renewable energy has become increasingly popular in the USA. Despite their wide usage, rooftop PV arrays are vulnerable to damage under strong winds. This can be attributed to the underestimation of peak wind loads on these systems where dynamic effects are unaccounted for. This study consists of investigating the wind-induced dynamic effects on rooftop PV arrays based on an experimental-numerical program and field calibration. Field measurements were conducted on a rooftop PV array at Central Washington University. Finite Element Modeling was performed to design the PV array model for experimental testing such that its dynamic properties are comparable to the in-situ array. Impact hammer and wind loading tests were carried out at the NHERI Wall of Wind Experimental Facility at Florida International University. The experimental tests were calibrated and validated based on the field measurements. Significant wind-induced vibrations were observed and their effect on the structure’s response was shown to increase with increasing wind speed.more » « less
-
This work presents a decision making approach for selecting an optimal placement of the grid-connected solar generation using Geographical Information System (GIS) as the decision making tool. A terrain analysis for solar radiation assessment, as well as buildings and vegetation spatial data are analyzed in order to determine the shadow impact that can be anticipated for medium or large-scale PV installation. In addition, different historical weather conditions are considered and integrated into the model to show the impact of this variable on the solar generation output. Some details of the methodology, testbed development and results related to the selection of potential sites for PV installation are presented. To illustrate the process and proposed methodology, an example using large scale synthetic networks is implemented.more » « less
-
null (Ed.)Image-based process monitoring has recently attracted increasing attention due to the advancement of the sensing technologies. However, existing process monitoring methods fail to fully utilize the spatial information of images due to their complex characteristics including the high-dimensionality and complex spatial structures. Recent advancements in unsupervised deep models such as generative adversarial networks (GAN) and adversarial autoencoders (AAE) has enabled to learn the complex spatial structures automatically. Inspired by this advancement, we propose an anomaly detection framework based on the AAE for unsupervised anomaly detection for images. AAE combines the power of GAN with the variational autoencoder, which serves as a nonlinear dimension reduction technique. Based on this, we propose a monitoring statistic efficiently capturing the change of the data. The performance of the proposed AAE-based anomaly detection algorithm is validated through a simulation study and real case study for rolling defect detection.more » « less
