Maritime vessels and offshore platforms must endure constant waves and currents. They also must deal, often without warning, with rogue waves, freak storms and other extreme events that potentially damage structures or ships.

MIT has developed a new algorithm that uses machine learning to predict and pinpoint when these extreme events will occur, including waves of varying magnitudes, lengths and heights that can create stress and pressure on a ship or offshore platform.

The algorithm provides a faster, more accurate risk assessment for systems that are likely to endure an extreme event at some point during their expected lifetime. It takes into account not only the statistical nature of the rogue wave or storm, but also the underlying dynamics.

“With our approach, you can assess, from the preliminary design phase, how a structure will behave not to one wave but to the overall collection or family of waves that can hit this structure,” said Themistoklis Sapsis, associate professor of mechanical and ocean engineering at MIT. “You can better design your structure so that you don’t have structural problems or stresses that surpass a certain limit.”

The algorithm predicting how rogue waves impact offshore rigs. Source: MITThe technique is not limited to ships or ocean platforms but can be applied to any complex system that might be vulnerable to extreme events such as severe flooding in a city or electrical overloads that could cause blackouts.

Typically, engineers gauge extreme events through computationally intensive simulations to structure a wave coming from a particular direction. However, this is very complex as it requires engineers to simulate millions of waves with different parameters such as height and length scale. This could take months to compute.

“That’s an insanely expensive problem,” Sapsis said. “To simulate one possible wave that can occur over 100 seconds, it takes a modern graphic processor unit, which is very fast, about 24 hours. We’re interested to understand what is the probability of an extreme event over 100 years.”

As a shortcut, MIT used these simulators to run just a few scenarios, choosing to simulate several random wave types that might cause maximum damage. If a structure survives the extreme waves, engineers assume the design will stand up to similar extreme events in the ocean.

MIT’s algorithm identifies the most important wave to run through a simulation based on the idea that each wave has a certain probability of contributing to an extreme event on the structure. While there is some uncertainty, the algorithm can quickly feed in various types of waves and their physical properties along with the known effects on offshore platforms.

Researchers tested the algorithm on a theoretical scenario involving an offshore platform subjected to incoming waves. The team started by plugging four typical waves into the machine learning algorithm including the waves’ known effects on an offshore platform. The algorithm was able to identify the dimensions of the new wave that has a high probability of occurring and maximizes the reduction of errors for the probability of an extreme event, MIT said.

MIT then plugged this wave into more computationally intensive, open-source simulation to model the response of a simplified offshore platform. The results were fed back into the algorithm to identify the next best wave to simulate. They then repeated the entire process. In all, the team ran 16 simulations over several days to model a platform’s behavior under extreme events.

The results of the tests allowed the team to hone in on the waves that are most certain to be involved in an extreme event allowing designers to have more informed, realistic scenarios to simulate in order to test the endurance of not just offshore platforms but also power grids and flood-prone regions.

The full research can be found in the journal Proceedings of the National Academy of Sciences.