Researchers at Sweden's Uppsala University have applied deep machine learning to automatically identify photovoltaic and solar-thermal systems in aerial imagery and said their work yielded mixed results.
They looked specifically into ariel imagery from Sweden by using a deep learning framework called DeepSolar CNN, which was developed by Stanford scientists. The framework uses a convolutional neural network (CNN), which enables extracting and learning features from visual data.
According to the research group, the proposed framework achieved an accuracy of 63.9% when used over a Swedish data set. That is lower than previous research conducted with the same framework in other countries. For example, a US group of researchers attained an accuracy of 91%, and a study done in Germany achieved 87.3%.
However, the Swedish-trained CNN has achieved more competitive results regarding the recall rate. While the precision rate refers to the method's ability not to make mistakes, the recall rate refers to its ability not to let positive information slip through. In that recall metric, the Swedish DeepSolar has achieved 81.8%, compared to 98.1% in the US and 87.5% in Germany.
“Regarding the lower precision achieved in this study compared to the previous publications, one explanation is that our scans of complete municipalities in the sparsely populated Sweden contain a much larger share of negative images than the mentioned studies,” the scientists explained. “As the main goal is to evaluate how useful detection of decentralized solar energy systems (SESs) by aerial images and a CNN classification algorithm are for creating as comprehensive a database as possible, a high recall is more important than a high precision.”
The scientists said the algorithm was first trained with a data set from North-Rhine Westphalia state in Germany and was then fine-tuned to Sweden with pictures from eight municipalities. It was then used to scan the whole spatial area of three Swedish municipalities – Uppvidinge, Falun, and Knivsta. These data were compared with other data collected with onsite inspections.
This iterative process involved multiple scans, with the CNN algorithm being retrained after each municipality scan, resulting in progressively enhanced accuracy. In the initial scan, the algorithm detected 89% of the detectable PV systems (excluding BIPV and vertical installations) and 59% of the ST systems,” the scientists emphasized. “Remarkably, by the fourth and final scan, these detection rates improved to 95% for PV systems and 80% for ST systems.”
They also specified that most undetected PV systems were frameless modules, typically installed on darker-colored roofs. In addition, shading from trees or structures, image reflections, and systems with high tilt angles impeded the classification algorithm's detection efficacy.
“Accuracy underscores the model's ability as both an inventory tool and a mechanism for constructing comprehensive databases of existing SESs,” the Swedish team concluded. “Connecting such a database, where the exact locations of the SESs are known, to existing building and property inventories, facilitates the generation of remarkably detailed SES market segment statistics.”
Its findings can be found in the paper “Mapping of decentralized photovoltaic and solar thermal systems by remote sensing aerial imagery and deep machine learning for statistic generation,” published in Energy and AI.
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: email@example.com.