Data-driven scenario generation of renewable energy production based on controllable generative adversarial networks with interpretability

Dong, Wei; Chen, Xianqing; Yang, Qiang

doi:10.1016/j.apenergy.2021.118387

Cited by 86 publications

(29 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, will also consider the generated samples' fidelity and consistency to the given condition when doing discrimination. As many existing studies [13,14,15,16,17,18], cGAN becomes the downstream backbone architecture of our conditional solar irradiance time-series generation task.…”

Section: Generative Adverserial Network (Gan)mentioning

confidence: 99%

“…We sort all the values regardless of the time steps into 50 bins to form their respective histogram distributions for both generated as well as ground-truth samples. Then, similar to [17,18,31], we apply Jensen-Shannon Divergence (JSD), as shown in Eq. 10) to measure the similarity between these two histogram distributions.…”

Section: Performance Evaluation On Time Series 421 Statistical Metric...mentioning

confidence: 99%

“…It can be understood as calculating the covariance between a time series after shifting time-steps and itself. Increasing from 1 to − 1 on the year-long daily cumulative solar irradiation gives us an annual autocorrelation curve, which is a classic indicator of the long-term trend of the solar irradiation [42,18,31]. We use Mean Absolute Error (MAE) to calculate the discrepancy between the autocorrelation curves of all the generated samples and that of the ground-truth.…”

Section: Performance Evaluation On Time Series 421 Statistical Metric...mentioning

confidence: 99%

“…In the energy sector, DGNs have already been used to generate synthetic energy renewable energy yield or building energy consumption timeseries data after being trained on simulated ground-truth data. [13,14,15,16,17,18]. These energy time-series data generators first overcome the lack of time resolution, and can generate high-fidelity but stochastic samples based on the given conditions, e.g., weather, date, value range, etc.…”

Section: Introduction 1urban Solar Modelling: Challenges and Opportun...mentioning

confidence: 99%

See 3 more Smart Citations

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

Zhang¹,

Schlueter²,

Waibel³

2022

Preprint

View full text Add to dashboard Cite

A generative model using simple fisheye images as inputs to predict solar irradiance time series on building facades in urban environments• Fisheye images as categorical shading masks are captured from opensource LOD 1 urban geometry, but in principle may also stem from real photographs• Model architecture allows for generation of stochastic ensembles of annual hourly time series, as well as good consistency to the deterministic data• Developed a network architecture combining Variational Autoencoder (VAE) and Generative Adversarial Networks (GAN)• Developed model addresses inherent uncertainty of intermittent solar energy and allows parametrically altering urban environmental features

show abstract

Section: Generative Adverserial Network (Gan)mentioning

confidence: 99%

Section: Performance Evaluation On Time Series 421 Statistical Metric...mentioning

confidence: 99%

Section: Performance Evaluation On Time Series 421 Statistical Metric...mentioning

confidence: 99%

Section: Introduction 1urban Solar Modelling: Challenges and Opportun...mentioning

confidence: 99%

See 2 more Smart Citations

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

Zhang¹,

Schlueter²,

Waibel³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…e uncertainty and uctuation modeling of renewable energy (such as wind and solar energy) production based on e cient and reliable scenario generation can be used for power system planning and operation [7].…”

Section: Introductionmentioning

confidence: 99%

Generative Adversarial Network and Its Application in Energy Internet

Xiao

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Energy Internet (EI) can provide consumers with flexible energy-sharing services. Recently, artificial intelligence (AI) technology has been widely used in the field of EI. Generative adversarial network (GAN) is one of the hottest research directions in the field of AI in recent years, and its excellent data generation ability has attracted wide attention. First, this paper introduces the framework, advantages, disadvantages, and improvement of classic GAN. Then, the application of GAN in the field of EI is reviewed. Finally, the paper is summarized, and the possible application of GAN in EI is prospected.

show abstract

Discriminating Deceptive Energy Generation of Photovoltaic Systems by Deep Learning and Adversarial Networks

Polo-Rodriguez,

Almonacid-Olleros,

Almonacid

et al. 2023

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Data-driven scenario generation of renewable energy production based on controllable generative adversarial networks with interpretability

Cited by 86 publications

References 35 publications

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

Generative Adversarial Network and Its Application in Energy Internet

Discriminating Deceptive Energy Generation of Photovoltaic Systems by Deep Learning and Adversarial Networks

Contact Info

Product

Resources

About