In photovoltaic power plants, early information on meteorological influences and automated, digitized and optical inspection procedures can further improve plant performance
Photovoltaics (PV) have been steadily expanded in the last years and now contribute significantly to the power supply in Germany. In 2021, photovoltaic systems produced 9.1 percent of the electricity generated in Germany. In the PVOptDigital joint project, five companies, the Fraunhofer Center for Silicon Photovoltaics and the DLR Institute for Solar Research are developing solutions to increase the energy performance of PV power plants and reduce operation costs so that PV plants can make their best possible contribution to the energy transition in Germany and worldwide in the future.
The goal of the PVOptDigital project is to increase the quality of monitoring of power plant operation and performance and thus achieve better operation of photovoltaic plants. Automated monitoring and the resulting optimized and more efficient operational management will open up an additional yield potential of up to 5 percent. In addition, automated, digitized data processing reduces the maintenance effort of PV systems. As a result, the cost of electricity decreases and the certainty of achieving the planned production increases at the same time. A limiting factor is currently the insufficient availability of information on the condition of the PV system down to module level. This is necessary to deduce concrete recommendations for action.
Researchers investigate options for quality improvement
In order to increase the performance and economic efficiency of PV systems, researchers are continuously investigating options for quality management. Faults are to be detected at an early stage so that suitable measures can be developed. The focus is on the early detection of temporal trends based on performance parameters, instead of reacting to threshold values that are too simple and sometimes set randomly. Performance parameters provide conclusions about how well the PV system is functioning and where the causes of performance drops lie. If this is observed over a longer period of time, PV system operators can derive a downward trend from this and react earlier instead of waiting for a threshold value, for example a lower limit of produced electricity, to be undershot.
In addition, the scientists are researching a demand-oriented, economically sensible individual module evaluation by means of optical inspection using drone flights and surveillance cameras. Existing field measurement methods are being improved and used to achieve a greater level of detail in the yield analysis of a park. Instead of prescribing maintenance intervals at fixed times, the PV power plant can use the measurement results to take demand-specific measures when they are actually needed. The cost of such a predictive approach is lower than the cost of preventive maintenance.
For example, processes such as image evaluation are automated and digitized. Early detection of certain performance parameters through automated and digitized processes enables better profitability and minimizes the costs to run the operation. It also increases overall yield, production quality, equipment reliability, and monitoring quality, and allows repairs to be completed in less time. To implement this, it is first necessary to record yield target values as accurately as possible, using newly developed yield sensors for various module technologies or image processing techniques with existing monitoring cameras. With the evaluation of the operating data, a comparison is made with the actual yield values. A detected deviation trend triggers a drone deployment. The use of novel measurement techniques, such as hyperspectral imaging or near-field and reverse current IR imaging, will be explored and their cost-effectiveness investigated.
Ultimately, options for action in operation will be derived, such as mowing, cleaning, replacement of modules and components, or warranty claims. The innovations should enable more cost-effective and efficient operation of large and difficult-to-access PV systems.
Task of the DLR
The primary task of DLR's working group in the PVOptDigital project is to develop an image analysis method to reliably and cost-effectively determine soiling, vegetation, module orientation and snow cover. In addition, researchers are measuring soiling on modules to further improve infrared (IR) measurement methods. A feasibility study will examine how the method of image analysis can be applied to satellite data. Incorporating project results into new or enhanced standards is another DLR goal in PVOptDigital.
In the PVOptDigital project, partners from the companies DiSUN Deutsche Solarservice GmbH, Aquila Capital Management GmbH, Aerial PV Inspection GmbH, CSP Services GmbH, Papendorf IT Services are working together with scientists from the DLR Institute for Solar Research.
Supported by Federal Ministry for Economic Affairs and Climate Action