The global shift towards a fully decarbonised energy supply requires the installation of a large amount of renewable energy systems, with great hopes being invested in photovoltaic technology. In order to further improve the competitiveness of photovoltaic systems that are operated continuously, a high energy yield at high levels of efficiency are required.
By exploring possibilities for efficient energy conversion in photovoltaic power plants, the OptPV4.0 research project aims to optimize the operation of solar power plants in order to increase and guarantee the energy yield and profitability of these systems. An optimum operation of photovoltaic installations can only by guaranteed by a long-term performance and a continuous system optimization. Therefore, a loss of performance caused by abrupt failures or gradual degradations must be prevented by identifying the causes of failure in a quick and reliable manner. In order to provide extensive knowledge of the plant performance, an intuitive sensor system, which is capable of extending missing measured values in existing photovoltaic power plants, will be developed within the OptPV4.0 project.
In order to identify failures, a reference image of a photovoltaic power plant in a fault-free condition is needed. Such a condition will be simulated as a type of “digital twin” in the OptPV4.0 project by developing algorithms based on physical models (whitebox modelling with integrated failure thresholds or degradation curves), statistical models (e.g. neural networks and deep learning) or a combination of both. The models should be able to simulate the state and operation of the system in a holistic manner and predict the time-related developments of the energy yield.
The aim is to simulate and analyse long-term trends on the basis of power plant data, covering longer periods of time. By integrating the evaluation of such data and by identifying trends, new possibilities for a predictive and preventive maintenance of solar power systems are created that reach far beyond the current maintenance intervals and breakdown services in the photovoltaics industry. Overall, the models developed in the project, hope to ensure long-term and efficient operation of photovoltaic power plants to achieve a maximum energy yield.