Solar irradiance nowcasts can be derived with sky images from all sky imagers (ASI) by detecting and analyzing transient clouds, which are the main contributor of intra-hour solar irradiance variability. The accuracy of ASI based solar irradiance nowcasting systems depends on various processing steps. Two vital steps are the cloud height detection and cloud tracking. This task is challenging, due to the atmospheric conditions that are often complex, including various cloud layers moving in different directions simultaneously. This challenge is addressed by detecting and tracking individual clouds. For this, we developed two distinct ASI nowcasting approaches with four or two cameras and a third hybridized approach. These three systems create individual 3-D cloud models with unique attributes 2 including height, position, size, optical properties and motion. This enables us to describe complex multi-layer conditions. In this paper, derived cloud height and motion vectors are compared with a reference ceilometer (height) and shadow camera system (motion) over a 30 day validation period. The validation data set includes a wide range of cloud heights, cloud motion patterns and atmospheric conditions. Furthermore, limitations of ASI based nowcasting systems due to image resolution and image perspective constrains are discussed. The most promising system is found to be the hybridized approach. This approach uses four ASIs and a voxel carving based cloud modeling combined with a cloud segmentation independent stereoscopic cloud height and tracking detection. We observed for this approach an overall mean absolute error of 648 m for the height, 1.3 m/s for the cloud speed and 16.2° for the motion direction.
International audienceBecause of the cloud-induced variability of the solar resource, the growing contributions of photovoltaic plants to the overall power generation challenges the stability of electricity grids. To avoid blackouts, administrations started to define maximum negative ramp rates. Storages can be used to reduce the occurring ramps. Their required capacity, durability, and costs can be optimized by nowcasting systems. Nowcasting systems use the input of upward-facing cameras to predict future irradiances. Previously, many nowcasting systems were developed and validated. However, these validations did not consider aggregation effects, which are present in industrial-sized power plants. In this paper, we present the validation of nowcasted global horizontal irradiance (GHI) and direct normal irradiance maps derived from an example system consisting of 4 all-sky cameras (“WobaS-4cam”). The WobaS-4cam system is operational at 2 solar energy research centers and at a commercial 50-MW solar power plant. Besides its validation on 30 days, the working principle is briefly explained. The forecasting deviations are investigated with a focus on temporal and spatial aggregation effects. The validation found that spatial and temporal aggregations significantly improve forecast accuracies: Spatial aggregation reduces the relative root mean square error (GHI) from 30.9% (considering field sizes of 25 m2) to 23.5% (considering a field size of 4 km2) on a day with variable conditions for 1 minute averages and a lead time of 15 minutes. Over 30 days of validation, a relative root mean square error (GHI) of 20.4% for the next 15 minutes is observed at pixel basis (25 m2). Although the deviations of nowcasting systems strongly depend on the validation period and the specific weather conditions, the WobaS-4cam system is considered to be at least state of the art
The demand for accurate solar irradiance nowcast increases together with the rapidly growing share of solar energy within our electricity grids. Intra-hour variabilities, mainly caused by clouds, have a significant impact on solar power plant dispatch and thus on electricity grids. All sky imager (ASI) based nowcasting systems, with a high temporal and spatial resolution, can overall mean-absolute deviation (MAD) and root-mean-square deviation (RMSD) are 0.11 and 0.16 respectively for transmittance. The deviations are significantly lower for optically thick or thin clouds and larger for clouds with moderate transmittance between 0.18 and 0.585. Furthermore we validated the overall DNI forecast quality of the entire nowcasting system, using this transmittance estimation method, over the same data set with three spatially distributed pyrheliometers. Overall deviations of 13% and 21% are reached for the relative MAD and RMSD with a lead time of 10 minutes. The effects of the chosen data set on the validation results are demonstrated by means of the skill score.
The share of distributed solar power generation is continuously growing. This increase, combined with the intermittent nature of the solar resource creates new challenges for all relevant stakeholders, from generation to transmission and demand. Insufficient consideration of intra‐minute and intra‐hour variabilities might lead to grid instabilities. Therefore, the relevance of nowcasts (shortest‐term forecasts) is steadily increasing. Nowcasts are suitable for fine‐grained control applications to operate solar power plants in a grid‐friendly way and to secure stable operations of electrical grids. In space and time, highly resolved nowcasts can be obtained by all sky imager (ASI) systems. ASI systems create hemispherical sky images. The associated software analyzes the sky conditions and derives solar irradiance nowcasts. Accuracy is the decisive factor for the effective use of nowcasts. Therefore, the goal of this work is to increase the nowcast accuracy by combining ASI nowcasts and persistence nowcasts, which persist with the prevailing irradiance conditions, while maintaining the spatial coverage and resolution obtained by the ASI system. This hybrid approach combines the strengths while reducing the respective weaknesses of both approaches. Results of a validation show reductions of the root mean square deviation of up to 12% due to the hybrid approach.
Abstract. 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 %.
Accurate nowcasts of the direct normal irradiance (DNI) for the next 15 min ahead can enhance the overall efficiency of concentrating solar power (CSP) plants. Such predictions can be derived from ground based all sky imagers (ASI). The main challenge for existing ASI based nowcasting systems is to provide spatially distributed solar irradiance information for the near future, which considers clouds with varying optical properties distributed over multiple heights. In this work, a novel object oriented approach with four spatially distributed ASIs is presented. One major novelty of the system is the application of an individual 3D model of each detected cloud as a cloud object with distinct attributes (height, position, surface area, volume, transmittance, motion vector etc.). Frequent but complex multilayer cloud movements are taken into account by tracking each cloud object separately. An extended validation period at the Plataforma Solar de Almería (PSA) on 30 days showing diverse weather conditions resulted in an average relative mean absolute error (relMAE) of around 15 % for a medium lead time of 7.5 minutes and a temporal average of 15 minutes. Further reductions of the relMAE were achieved by spatial aggregation, with a relMAE of 10.7 % for a lead time of 7.5 minutes, a field size of 4 km² and a temporal average of 1 minute (during one day). Nowcasting systems described in the literature reach similar deviations but were often validated only for a few days based on a single ground measurement station, which confirms the good performance and the high applicability of the presented system. Three implementations of the system exist already demonstrating the market maturity of the system.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.