A new method was developed to estimate the cloud optical depth (τ_c) from photovoltaic (PV) power measurements under overcast sky conditions. It is a fully physical method utilizing directly PV power measurements. It exploits the recent advances and real-time availability at global scale of aerosol properties, downwelling shortwave irradiance and its direct and diffuse components received at ground level under clear-sky conditions and ground albedo, altogether provided by the Copernicus Atmosphere Monitoring Service (CAMS) radiation service. In addition to CAMS data, wind speed and air temperature from European Centre for Medium-Range Weather Forecasts twentieth century reanalysis ERA5 products are also used as inputs. The τ_c estimates have been compared to different data sources of τ_c retrievals at four experimental PV sites located in various climates. When compared to τ_c retrieved from groundbased pyranometer measurements serving as reference, the correlation coefficient is greater than 0.97. The bias ranges between -3 and 4, i.e., -8 % and 14 % in relative value. The root mean square error (RMSE) lies in the interval [3,8] ([9,21] % in relative value). When compared to satellitebased retrievals from Meteosat Second Generation and Moderate Resolution Imaging Spectroradiometer, both relative errors become comprehensively greater. Nevertheless, our method remarkably reduces the relative bias and RMSE, by up to 10 % and 20 % respectively, compared to the existing state-of-the-art approach. This work demonstrates the accuracy of the method and clearly shows its great potential use whenever PV power measurements are available.

Intelligent condition monitoring of wind turbines is essential for reducing downtimes. Machine learning models trained on wind turbine operation data are commonly used to detect anomalies and, eventually, operation faults. However, data-driven normal behavior models (NBMs) require a substantial amount of training data, as NBMs trained with scarce data may result in unreliable fault detection. To overcome this limitation, we present a novel generative deep transfer learning approach to make SCADA samples from one wind turbine lacking training data resemble SCADA data from wind turbines with representative training data. Through CycleGAN-based domain mapping, our method enables the application of an NBM trained on an existing wind turbine to a new one with severely limited data. We demonstrate our approach on field data mapping SCADA samples across 7 substantially different WTs. Our findings show significantly improved fault detection in wind turbines with scarce data. Our method achieves the most similar anomaly scores to an NBM trained with abundant data, outperforming NBMs trained on scarce training data with improvements of +10.3% in F1-score when 1 month of training data is available and +16.8% when 2 weeks are available. The domain mapping approach outperforms conventional fine-tuning at all considered degrees of data scarcity, ranging from 1 to 8 weeks of training data. The proposed technique enables earlier and more reliable fault detection in newly installed wind farms, demonstrating a novel and promising research direction to improve anomaly detection when faced with training data scarcity.

As wind energy adoption is growing, ensuring the efficient operation and maintenance of wind turbines becomes essential for maximizing energy production and minimizing costs and downtime. Many AI applications in wind energy, such as in condition monitoring and power forecasting, may benefit from using operational data not only from individual wind turbines but from multiple turbines and multiple wind farms. Collaborative distributed AI which preserves data privacy holds a strong potential for these applications. Federated learning has emerged as a privacy-preserving distributed machine learning approach in this context. We explore federated learning in wind turbine condition monitoring, specifically for fault detection using normal behaviour models. We investigate various federated learning strategies, including collaboration across different wind farms and turbine models, as well as collaboration restricted to the same wind farm and turbine model. Our case study results indicate that federated learning across multiple wind turbines consistently outperforms models trained on a single turbine, especially when training data is scarce. Moreover, the amount of historical data necessary to train an effective model can be significantly reduced by employing a collaborative federated learning strategy. Finally, our findings show that extending the collaboration to multiple wind farms may result in inferior performance compared to restricting learning within a farm, specifically when faced with statistical heterogeneity and imbalanced datasets.

Wind energy plays a critical role in the transition towards renewable energy sources. However, the uncertainty and variability of wind can impede its full potential and the necessary growth of wind power capacity. To mitigate these challenges, wind power forecasting methods are employed for applications in power management, electricity trading, or maintenance scheduling. In this work, we present, evaluate, and compare four machine learning-based wind power forecasting models. Our models correct and improve 48-hour forecasts extracted from a numerical weather prediction (NWP) model. The models are evaluated on datasets from a wind park comprising 65 wind turbines. The best improvement in forecasting error and mean bias was achieved by a convolutional neural network, reducing the average NRMSE down to 22%, coupled with a significant reduction in mean bias, compared to a NRMSE of 35% from the strongly biased baseline model using uncorrected NWP forecasts. Our findings further indicate that changes to neural network architectures play a minor role in affecting the forecasting performance, and that future research should rather investigate changes in the model pipeline. Moreover, we introduce a continuous learning strategy, which is shown to achieve the highest forecasting performance improvements when new data is made available.

Federated learning has recently emerged as a privacy-preserving distributed machine learning approach. Federated learning enables collaborative training of multiple clients and entire fleets without sharing the involved training datasets. By preserving data privacy, federated learning has the potential to overcome the lack of data sharing in the renewable energy sector which is inhibiting innovation, research and development. Our paper provides an overview of federated learning in renewable energy applications. We discuss federated learning algorithms and survey their applications and case studies in renewable energy generation and consumption. We also evaluate the potential and the challenges associated with federated learning applied in power and energy contexts. Finally, we outline promising future research directions in federated learning for applications in renewable energy.

Surface solar irradiance (SSI) plays a crucial role in tackling climate change — as an abundant, non-fossil energy source, exploited primarily via photovoltaic (PV) energy production. With the growing contribution of SSI to total energy production, the stability of the latter is challenged by the intermittent character of the former, arising primarily from cloud effects. Mitigating this stability challenge requires accurate, uncertainty-aware, near real-time, regional-scale SSI forecasts with lead times of minutes to a few hours, enabling robust real-time energy grid management. State-of-the-art nowcasting methods typically meet only some of these requirements. Here we present SHADECast, a deep generative diffusion model for the probabilistic spatiotemporal nowcasting of SSI, conditioned on deterministic aspects of cloud evolution to guide the probabilistic ensemble forecast, and based on near real-time satellite data. We demonstrate that SHADECast provides improved forecast quality, reliability, and accuracy in all weather scenarios. Our model produces realistic and spatiotemporally consistent predictions extending the state-of-the-art forecast horizon by 26 min over different regions with lead times of 15-120 min. Our physics-informed generative approach leads to up to 60% performance improvement in extreme value prediction over the state-of-the-art deterministic models, showcasing the advantage of probabilistic modeling of cloudiness over the classical deterministic approach. It also surpasses the probabilistic benchmarks in predicting extreme values. Finally, SHADECast empowers grid operators and energy traders to make informed decisions, ensuring stability and facilitating the seamless integration of PV energy across multiple locations simultaneously.