Condition Monitoring and Artificial Intelligence

Do you still remember? 22 years ago, insurers in the wind energy sector were faced with a flood of claims that they could not financially cope with in the long term. An overhaul clause was drawn up, which essentially stipulated that some of the drive elements had to be replaced after 40,000 operating hours or no later than 5 years after commissioning or the last overhaul, regardless of their condition. This also included the main bearings, all roller bearings in the gearbox and the generator bearings. This financially costly and not environmentally friendly scenario was a cry for help. At the same time, however, an alternative was also sought. And this was found in the form of modern condition monitoring based on vibration diagnostics. Gothaer Versicherungen then formulated the principles for the condition-based maintenance of wind turbines, which GfM also contributed to.

In fact, GfM has been increasingly involved in mobile vibration diagnostics on wind turbines since 2002, both through its own measurements and through cooperation with appraisers. The first automatic online condition monitoring system has been in operation also since 2002.

A lot has changed in the last 20 years. The stocking of spare parts, availability of maintenance personnel and crane capacities, but also simply the better adaptation of the drives to the special application of wind energy and the tribological processes are better controlled. This has a positive effect on the damaging processes. And yet it is true for wind energy, as for all other industries, that damage can develop that leads to premature failure. Condition monitoring helps to detect this damage in good time in order to avoid consequential damage and longer downtimes.

GfM has continuously developed its systems and has already implemented an extremely high level of diagnostic reliability and automation. In the future, steps will follow that use artificial intelligence tools. At the 32nd Wind Energy Days from November 5-7, 2024 in Linstow, interested listeners will be able to hear about initial approaches in a presentation.