This project investigated the feasibility of automated collection and analysis of visual data for predictive inspection and evaluation of the condition of concentrating solar power (CSP) and solar photovoltaic (PV) plants.
Following the rapid expansion of the renewables industry sector in Australia over the last ten years, there is a growing focus in the industry on operation and maintenance (O&M) of infrastructure. A key component of effective O&M is the availability of detailed and timely data on infrastructure to enable the scheduling of maintenance and cleaning activities to economically maximise operational efficiency. For large distributed infrastructure, such as is common in renewable energy systems, efficient and economical acquisition and analysis of such data is in itself a substantive challenge.
This project investigated the feasibility of automated collection and analysis of visual data for predictive inspection and evaluation of the condition of concentrating solar power (CSP) and solar photovoltaic (PV) plants. Large-scale imaging of solar plant facilities from moving aerial, ground vehicles, and fixed view-points has the potential to reduce the time, cost, and labour spent on the monitoring of facility conditions and operating efficiency through early detection of equipment failures, accurate diagnosis of performance losses, and identification of maintenance and cleaning priorities. Aerial vehicles in particular have the potential to operate autonomously over fields of photovoltaic panels or concentrating solar power heliostat mirrors, minimizing damage and soiling caused by vehicle traffic through a solar field, and allowing direct transit from point to point leading to efficiency gains. The project evaluated aerial robotic technology for acquisition of visual data and automated processing of visual data to provide analytics on fault detection and soiling levels in solar photovoltaic and concentrating solar power plants.