PV4Life – Extending the service life of power converters through optimised component design using digital AI twins

The PV4Life project is investigating the service life of power converters for photovoltaic (PV) systems, as the reliability and service life of power converter-based systems are key social and economic factors in shaping our energy system to ensure greater security of supply at the lowest possible cost.

With this in mind, the project is conducting research into an effective approach to navigating the various stages of the process for the new development and improvement of power converter-based systems and their critical components.

To investigate the service life of PV power converters, three endurance tests spanning several years are being carried out so that the ageing of a specific power converter can be observed and measured. In parallel, the power converters are being examined at component level in separate test set-ups, by subjecting the power semiconductors, capacitors and power chokes to accelerated ageing tests.

The aim of the endurance tests and component tests is to determine how power converters age and which parameters are affected. Based on these findings, service life models will first be developed for the individual components and then for the entire power converter. Furthermore, the findings will be incorporated into the development of accelerated ageing tests, which will be integrated into a hardware-in-the-loop (HiL) system.

The power converter model is also extended to create a power converter twin, the aim being for it to behave in the same way as the real power converter. 

Using the real power converter and the digital twin, a power converter analyser is ultimately developed, which, based on artificial intelligence, is designed to detect faults or technical anomalies, for example. This is intended to be incorporated into future power converters as a control strategy, thereby increasing their service life.

The focus of the H-BRS in this project is on investigating power converter components and their ageing, as well as developing accelerated ageing tests for use in the HiL system.