Yann Fabel

Yann Fabel

Auto-generated profile, based on publicly available information

Yann Fabel

Institute for Solar Research German Aerospace Center (DLR) Calle Doctor Carracido N°44 04005 Almería Spain

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 February 2022 in Atmospheric Measurement Techniques
Verified
Journal Article
Open access
PDF file
YFYann Fabel
Yann Fabel
BNBijan Nouri
Bijan Nouri
SWStefan Wilbert
Stefan Wilbert
NBNiklas Benedikt Blum
Niklas Benedikt Blum
RTRudolph Triebel
Rudolph Triebel
RPRobert Pitz-Paal
Robert Pitz-Paal
Semantic segmentation of ground-based all-sky images (ASIs) can provide high-resolution cloud coverage information of distinct cloud types, applicable for meteorology-, climatology- and solar-energy-related applications. Since the shape and appearance of clouds is variable, and there is high similarity between cloud types, a clear classification is difficult. Therefore, most state-of-the-art methods focus on the distinction between cloudy and cloud-free pixels without taking into account the cloud type. On the other hand, cloud classification is typically determined separately at the image level, neglecting the cloud's position and only considering the prevailing cloud type. Deep neural networks have proven to be very effective and robust for segmentation tasks; however they require large training datasets to learn complex visual features. In this work, we present a self-supervised learning approach to exploit many more data than in purely supervised training and thus increase the model's performance. In the first step, we use about 300 000 ASIs in two different pretext tasks for pretraining. One of them pursues an image reconstruction approach. The other one is based on the DeepCluster model, an iterative procedure of clustering and classifying the neural network output. In the second step, our model is fine-tuned on a small labeled dataset of 770 ASIs, of which 616 are used for training and 154 for validation. For each of them, a ground truth mask was created that classifies each pixel into clear sky or a low-layer, mid-layer or high-layer cloud. To analyze the effectiveness of self-supervised pretraining, we compare our approach to randomly initialized and pretrained ImageNet weights using the same training and validation sets. Achieving 85.8 % pixel accuracy on average, our best self-supervised model outperforms the conventional approaches of random (78.3 %) and pretrained ImageNet initialization (82.1 %). The benefits become even more evident when regarding precision, recall and intersection over union (IoU) of the respective cloud classes, where the improvement is between 5 and 20 percentage points. Furthermore, we compare the performance of our best model with regards to binary segmentation with a clear-sky library (CSL) from the literature. Our model outperforms the CSL by over 7 percentage points, reaching a pixel accuracy of 95 %.

19 October 2024 in Solar RRL
Verified
Journal Article
Open access
BNBijan Nouri
Bijan Nouri
YFYann Fabel
Yann Fabel
NBNiklas Blum
Niklas Blum
DSDominik Schnaus
Dominik Schnaus
LZLuis F. Zarzalejo
Luis F. Zarzalejo
Stefan WilbertStefan Wilbert
Stefan Wilbert
Solar irradiance forecasting plays a crucial role in integrating large quantities of intermittent solar power. Forecasting systems are commonly evaluated using metrics like root‐mean‐ square error (RMSE) and skill scores. However, these metrics aggregated over larger data sets do not adequately assess the prediction of ramp events, which are critical for many applications. This article introduces a novel, simple, and adaptable ramp rate metric that analyzes ramp events between successive lead times within forecasts. A case study on ramp rate mitigation in PV systems benchmarks suitable ramp thresholds for various solar irradiance components. The capabilities and limitations of deterministic and probabilistic forecasts from two all‐sky imager‐based models are evaluated for ramp prediction. A state‐of‐the‐art data‐driven vision transformer End2End model excels in RMSE and skill scores but performs poorly in ramp prediction. Conversely, a novel generative forecasting model combined with a convolutional neural network‐based irradiance model shows superior ramp prediction, achieving an F1 score of ≥0.7 for critical ramp events. This study underscores the importance of suitable ramp rate metrics and highlights the potential of generative models for enhancing ramp forecasting.

1 March 2023 in Solar Energy
Verified
Journal Article
Closed access
BNBijan Nouri
Bijan Nouri
Stefan WilbertStefan Wilbert
Stefan Wilbert
NBNiklas Benedikt Blum
Niklas Benedikt Blum
YFYann Fabel
Yann Fabel
ELElke Lorenz
Elke Lorenz
RPRobert Zentrum für Luft- und Raumfahrt e.V. Pitz-Paal
Robert Zentrum für Luft- und Raumfahrt e.V. Pitz-Paal

21 January 2024 in Solar RRL
Verified
Journal Article
Open access
YFYann Fabel
Yann Fabel
BNBijan Nouri
Bijan Nouri
Stefan WilbertStefan Wilbert
Stefan Wilbert
NBNiklas Benedikt Blum
Niklas Benedikt Blum
DSDominik Schnaus
Dominik Schnaus
RPRobert Zentrum für Luft- und Raumfahrt e.V. Pitz-Paal
Robert Zentrum für Luft- und Raumfahrt e.V. Pitz-Paal
Intermittent solar irradiance due to passing clouds poses challenges for integrating solar energy into existing infrastructure. By making use of intrahour nowcasts (very short‐term forecasts), changing conditions of solar irradiance can be anticipated. All‐sky imagers, capturing sky conditions at high spatial and temporal resolution, can be the basis of such nowcasting systems. Herein, a deep learning (DL) model for solar irradiance nowcasts based on the transformer architecture is presented. The model is trained end‐to‐end using sequences of sky images and irradiance measurements as input to generate point‐forecasts up to 20 min ahead. Further, the effect of integrating this model into a hybrid system, consisting of a physics‐based model and smart persistence, is examined. A comparison between the DL and two hybrid models (with and without the DL model) is conducted on a benchmark dataset. Forecast accuracy for deterministic point‐forecasts is analyzed under different conditions using standard error metrics like root‐mean‐square error and forecast skill. Furthermore, spatial and temporal aggregation effects are investigated. In addition, probabilistic nowcasts for each model are computed via a quantile approach. Overall, the DL model outperforms both hybrid models under the majority of conditions and aggregation effects.

18 August 2025 in Solar Energy
Verified
Journal Article
Open access
DMDavid Magiera
David Magiera
YFYann Fabel
Yann Fabel
BNBijan Nouri
Bijan Nouri
NBNiklas Benedikt Blum
Niklas Benedikt Blum
DSDominik Schnaus
Dominik Schnaus
LZLuis F. Zarzalejo
Luis F. Zarzalejo

24 April 2025 in Solar Energy
Verified
Journal Article
Open access
NBNiklas Benedikt Blum
Niklas Benedikt Blum
PMPaul Matteschk
Paul Matteschk
YFYann Fabel
Yann Fabel
BNBijan Nouri
Bijan Nouri
RRRoberto Román
Roberto Román
Stefan WilbertStefan Wilbert
Stefan Wilbert