skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Data-Driven Digital Twins for Power Estimations of a Solar Photovoltaic Plant
Renewable energy generation sources (RESs) are gaining increased popularity due to global efforts to reduce carbon emissions and mitigate effects of climate change. Planning and managing increasing levels of RESs, specifically solar photovoltaic (PV) generation sources is becoming increasingly challenging. Estimations of solar PV power generations provide situational awareness in distribution system operations. A digital twin (DT) can replicate PV plant behaviors and characteristics in a virtual platform, providing realistic solar PV estimations. Furthermore, neural networks, a popular paradigm of artificial intelligence may be used to adequately learn and replicate the relationship between input and output variables for data-driven DTs (DD-DTs). In this paper, DD-DTs are developed for Clemson University’s 1 MW solar PV plant located in South Carolina, USA to perform realistic solar PV power estimations. The DD-DTs are implemented utilizing multilayer perceptron (MLP) and Elman neural networks. Typical practical results for two DD-DT architectures are presented and validated.  more » « less
Award ID(s):
2234032
PAR ID:
10513115
Author(s) / Creator(s):
; ;
Publisher / Repository:
IEEExplore
Date Published:
Journal Name:
IEEE SSCI
ISSN:
1558-058X
ISBN:
979-8-3503-3211-7
Subject(s) / Keyword(s):
Digital twin, neural network, power estimation, solar photovoltaic
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; (, IEEE SSCI)
    Amidst the challenges posed by the high penetration of distributed energy resources (DERs), particularly a number of distributed photovoltaic plants (DPVs), in modern electric power distribution systems (MEPDS), the integration of new technologies and frameworks become crucial for addressing operation, management, and planning challenges. Situational awareness (SA) and situational intelligence (SI) over multi-time scales is essential for enhanced and reliable PV power generation in MEPDS. In this paper, data-driven digital twins (DTs) are developed using AI paradigms to develop actual and/or virtual models of DPVs, These DTs are then applied for estimating and forecasting the power outputs of physical and virtual PV plants. Virtual weather stations are used to estimate solar irradiance and temperature at user-selected locations in a localized region, using inferences from physical weather stations. Three case studies are examined based on data availability: physical PV plant, hybrid PV plants, and virtual PV plants, generating realtime estimations and short-term forecasts of PV power production that can support distribution system studies and decision-making. 
    more » « less
  2. ; ; (, Journal of Magnetic Resonance)
    In this paper, we numerically optimize broadband pulse shapes that maximize Hahn echo amplitudes. Pulses are parameterized as neural networks (NN), nonlinear amplitude limited Fourier series (FS), and discrete time series (DT). These are compared to an optimized choice of the conventional hyperbolic secant (HS) pulse shape. A power constraint is included, as are realistic shape distortions due to power amplifier nonlinearity and the transfer function of the microwave resonator. We find that the NN, FS, and DT parameterizations perform equivalently, offer improvements over the best HS pulses, and contain a large number of equivalent optimal maxima, implying the flexibility to include further constraints or optimization goals in future designs. 
    more » « less
  3. ; ; ; ; (, 2020 Resilience Week (RWS))
    null (Ed.)
    Motivated by decreased cost and climate change concerns, the penetration of solar photovoltaic (PV) energy generation and battery energy storage has been continually increasing. The variability in solar PV power generation has led to many new challenges for utilities and researchers. One challenge is the quantification of the resilience contribution to the grid from its assets and is the topic of this paper. In this work, we propose a framework for evaluating the resilience contribution of solar generation and battery storage assets on the grid. The metric provides a quantifiable adaptive capacity measure in terms of real and reactive power and includes uncertainty for solar PV assets. A case study using very short-term and short-term solar generation forecast demonstrates the framework and provides useful insight to the resilience solar and battery storage assets can contribute to the grid. 
    more » « less
  4. ; ; ; ; ; (, Emerging Science Journal)
    Solar energy is a widely accessible, clean, and sustainable energy source. Solar power harvesting in order to generate electricity on smart grids is essential in light of the present global energy crisis. However, the highly variable nature of solar radiation poses unique challenges for accurately predicting solar photovoltaic (PV) power generation. Factors such as cloud cover, atmospheric conditions, and seasonal variations significantly impact the amount of solar energy available for conversion into electricity. Therefore, it is essential to precisely estimate the output of solar power in order to assess the potential of smart grids. This paper presents a study that utilizes various machine learning models to predict solar photovoltaic (PV) power generation in Lubbock, Texas. Mean Squared Error (MSE) and R² metrics are utilized to demonstrate the performance of each model. The results show that the Random Forest Regression (RFR) and Long Short-Term Memory (LSTM) models outperformed the other models, with a MSE of 2.06% and 2.23% and R² values of 0.977 and 0.975, respectively. In addition, RFR and LSTM demonstrate their capability to capture the intricate patterns and complex relationships inherent in solar power generation data. The developed machine learning models can aid solar PV investors in streamlining their processes and improving their planning for the production of solar energy. Doi: 10.28991/ESJ-2023-07-04-02 Full Text: PDF 
    more » « less
  5. ; ; ; (, American Society of Mechanical Engineers)
    Abstract As the United States phases out traditional fossil fuels in favor of renewable energy sources, it is important to capitalize on all available avenues to increase renewable penetration. In the last decade, the costs associated with residential solar photovoltaic (PV) installations have decreased significantly, providing more homeowners with the opportunity to generate their own clean electricity. Research has found that the decision to invest in a residential solar PV system is guided by economic, social, and personal factors. Accounting for such complexities, the joint power of agent-based modeling and social network analysis is leveraged in this study to evaluate the effect of social influence on solar PV adoption. Featuring residential consumer agents with data-driven attributes, a logistic regression function to predict solar adoption, and random and small-world social network implementations, this work simulates residential solar PV adoption in New Jersey. Results indicate that including social influence in an agent-based electricity system model leads to increased installed residential solar capacity, but not necessarily higher adoption rates. These findings suggest that, with an understanding of the intricacies of consumer social networks, there are potential opportunities to bolster residential solar installations through low-cost social campaigns that motivate individuals to adopt home solar through their social ties. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email