The goal of the AID4PV project is to support the improvement of the financial efficiency of PV plants by increasing their operational availability time and reduction of their maintenance costs. For that purpose, a UAV-based detection and decision-making modular system will be developed and tested under real-life operating controlled conditions. The UAV platform will capture EL, RGB & IRT imagery that when integrated with electrical data analysis, near real-time fault detection will occur, leading to time and cost-efficient PV plant diagnosis. The system will be controlled by a user-friendly geo-visualization interface with advanced reporting and operational functionalities. Therefore, the main objectives are to:
- Develop novel EL and well-established IRT image processing algorithms complemented with RGB imagery to support PV panel characterization and early fault detection and localization
- Develop Artificial Intelligence (AI) methods to combine and complement image and electrical data analyses results, in order to maximize the classification rate of PV module failures
- Design, implement and operate an advanced UAV platform with on-board processing capabilities that will support sensor fusion, autonomous operation and decision making as well as precise absolute geo-location of faulty PV modules
- Develop a user-friendly visualization platform and system-user interface that will provide advanced geospatial and visualization tools and functionalities for near real-time fault-detection and fault-localization reporting
The expected project results with the specific impact and KPIs to measure the success of the project are outlined below:
- To improve the Performance Ratio of PV plants above 85%
- To improve individual PV module (panel) identification error below 10%, which are state of the art results, and trying to achieve 5% error obtained by means of a manual camera
- To improve both defect detection and classification (e.g. improving 24% error, which are state-of-the-art results, in detecting and classifying hot spot defects)
- To implement 100% autonomous UAV missions
- To provide optimized architectures for UAV performance in terms of data collection, real-time processing, safety and data transfer
- Perform economic analysis and evaluate the impact on levelized cost of energy (LCOE) (e.g. target a relative reduction of LCOE by 6%)
