OptOM – Cost-optimised operational management of PV systems over their economic lifetime

Duration: 04/2020 - 03/2023
Contracting Authority/ Sponsors: Federal Ministry of Economic Affairs and Climate Action (BMWK, Förderkennzeichen: 03EE1058)
Project Partners: Pohlen Solar GmbH / Centroplan GmbH (Konsortialführer); Mondas GmbH
Project Focus:        
The creeping degradation of aging PV power plants can be detected by AI methods. 
© Pohlen Solar GmbH
The creeping degradation of aging PV power plants can be detected by AI methods. 

More than 1.8 million photovoltaic plants with a total capacity of 52 GW are now in operation in Germany and have supplied 11.4 percent of net electricity generation so far in 2020. In the "OptOM" project, Pohlen Solar GmbH, Centroplan GmbH, Mondas GmbH and the Fraunhofer Institute for Solar Energy Systems ISE are developing new methods for intelligent monitoring of PV power plants. The goal of the project, which is funded by the German Federal Ministry for Economic Affairs and Energy, is the early detection of degradation symptoms in PV plants with the help of artificial intelligence. This should avoid yield losses in older plants and save maintenance costs through predictive maintenance. 

With the operating age of photovoltaic systems, the signs of wear and tear increase continuously. Increasing maintenance requirements and decreasing solar yields are the result. Because the degradation phenomena take place insidiously, conventional monitoring systems have not been able to detect them until now. The OptOM project consortium now wants to develop rule- and AI-based methods that make it possible to predict the plant failures associated with operating age. 

Intelligent monitoring of PV power plants 

As part of the project, 200 solar power plants of Pohlen Solar GmbH will be connected to the Internet-of-Things platform of the Freiburg IT company Mondas GmbH and their operating states will be analyzed. In parallel, Fraunhofer ISE scientists are developing corresponding algorithms that evaluate correlations between unusual operating states and plant failures. This will enable operators of photovoltaic power plants to prioritize suitable service work before malfunctions occur and yields are lost (predictive maintenance). 

The web platform to be developed in the project is intended to use intelligent, digital methods to help   

  • efficiently detect errors, 
  • plan preventive maintenance operations and 
  • prioritize trouble calls. 

The goal is to develop an ecologically and economically sustainable strategy for the operation of the more than 2000 PV power plants of Pohlen-Solar. In the future, large numbers of irrelevant fault messages, imprecise fault assignments, and unnecessary on-site visits are to be significantly minimized. 

High relevance for profitability and climate protection 

A look at the PV expansion figures in Germany shows that the topic of improving the performance of photovoltaic systems is certainly relevant for the energy transition and climate protection: according to the draft bill for the EEG amendment 2021, solar capacity is expected to almost double by another 50 GW by 2030. Intelligent monitoring systems can help increase the yield of many existing plants and thus contribute to climate protection.