COMLAT Adaptive Solar Tracking System, Istanbul, Turkey
收藏Zenodo2025-09-01 更新2026-05-26 收录
下载链接:
https://zenodo.org/doi/10.5281/zenodo.17015641
下载链接
链接失效反馈官方服务:
资源简介:
The experimental studies in this study deal with solar energy panels from Istanbul, Turkey, which is a place that can be described using the geographical coordinates 41.015137° N and 28.979530° E. Istanbul, a city that has a strategic location, lying on the railways that connect two continents, Europe and Asia, was chosen on the basis of its variability in temperature regimes owing to its climatic variability and geographical location as a city being the climate-resilient solar energy hotspot. Climate change in the city is caused by seasonal changes, that is, heat in summer, cold in winter, and moderate weather in spring and fall, making these seasons ideal for thoroughly examining the adaptability and effectiveness of the COMLAT solar tracking system. In addition, Istanbul has high solar energy levels, with an average production capacity of 1200-1400 kWh/m² per year; thus, it is of practical and economic interest for photovoltaic power applications. This climatic confusion and the possibility of more energy than was previously the case, as the major factors for these cities are supposed to have the capacity of adaptive tracking systems that can perform effectively under dynamic environmental conditions.
COMLAT Adaptive Solar Tracking System, a CNN-LSTM-based AI framework that involves climate forecasting, reinforcement learning, and low-power edge computing. The foremost goals are to strengthen energy yield, downsize computational costs, increase climate resilience, and ensure large-scale deployment in smart grids. The COMLAT architecture allows us to anticipate changes in solar radiation, dynamically reposition panels, and make real-time tracking decisions based on the environmental data. Solar forecasting is performed using CNN-LSTM deep learning models, whereas reinforcement learning is used to optimize panel movement based on irradiance variations that are learned continuously. The application of the system to edge AI hardware enables rapid decision-making and reduced energy consumption. Different tracking setups were examined using fixed-tilt, single-axis, dual-axis, and COMLAT systems under different climatic conditions.
提供机构:
Zenodo
创建时间:
2025-09-01



