In a significant stride towards enhancing ship performance and safety, researchers have developed a novel approach to digital twinning for ship motion prediction. The study, led by Giorgio Palma from the National Research Council–Institute of Marine Engineering in Rome, Italy, introduces a Bayesian extension of the Hankel dynamic mode decomposition (HDMD) method. This advancement is poised to revolutionize how the maritime industry approaches operational efficiency and safety.
Digital twins, virtual replicas of physical systems, are increasingly seen as game-changers in various industries, including maritime. They provide real-time insights and predictions, enabling better decision-making throughout a vessel’s lifecycle. Palma’s research focuses on digital twinning for ship performance in waves, a critical factor in operational efficiency and safety.
The proposed Bayesian HDMD method stands out because it adapts based on data from the physical system, requires only a limited amount of data, and produces real-time predictions with reliability estimates. This makes it particularly suitable for digital twinning applications. The study compares this Bayesian approach with its deterministic counterpart, demonstrating improved accuracy in predictions.
The research team tested the method on the course-keeping of the 5415M model in beam-quartering sea state 7 irregular waves at a Froude number of 0.33. They used benchmark data from three different computational fluid dynamics (CFD) solvers as a proxy for real vessel operations. The results showed reasonably accurate predictions up to five wave encounters, with the Bayesian formulation outperforming the deterministic forecasts.
One of the most promising findings was the relationship between uncertainty and accuracy. As Palma explains, “The Bayesian formulation not only improves the accuracy of the predictions but also provides a measure of uncertainty, which is crucial for informed decision-making.”
This research, published in the journal Applied Ocean Research (which translates to “Applied Ocean Research” in English), opens up new avenues for the maritime industry. By leveraging digital twins, shipping companies can optimize vessel performance, reduce fuel consumption, and enhance safety. The ability to predict ship motion accurately can also lead to better route planning and risk management.
The commercial impacts are substantial. For instance, more accurate predictions can help shipping companies avoid rough seas, reducing wear and tear on vessels and minimizing the risk of accidents. This can lead to significant cost savings and improved safety records.
Moreover, the method’s ability to adapt based on real-time data means it can be used for various types of vessels and sea conditions, making it a versatile tool for the maritime industry. As Palma notes, “The flexibility of the Bayesian HDMD method makes it applicable to a wide range of scenarios, from commercial shipping to offshore operations.”
In conclusion, this research represents a significant step forward in the digital twinning of ship performance. By providing more accurate and reliable predictions, it offers the maritime industry a powerful tool to enhance operational efficiency and safety. As the industry continues to embrace digital transformation, such advancements will play a crucial role in shaping the future of maritime operations.