Assessment of ship hull Integrity based on a Digital Twin

The maritime and offshore industries have called for a more structured method for registering and accessing information about the vessel hull condition, taking advantage of today’s technological potential. The development of a digital twin for assessing the vessel hull condition is a response to this with several benefits. Previous work has demonstrated that drones allow safe, remote, automated, and more flexible inspections. To give drone inspections an additional edge, the technology to do automated flights while the vessel is sailing, and the ability to automatically do thickness measurements, should be developed. Registering inspection data from drones into a digital twin is a natural next step. Given sufficient quality of drone-collected data and effective registration of human-collected data, the merger of such complementing data into a digital twin can allow more efficient inspections and reduce the number of times that humans are exposed to the dangers of confined space inspection. By acquiring relevant hull data, there is a potential for utilizing historical information from a multitude of vessels to create a statistical model for predicting the hull condition. This may allow the estimation of the progressing hull condition based on the operational conditions a vessel has been exposed to. In combination with structural models, this can also be used to automatically create hull inspection plans that prioritize areas with an increased risk of having defects. The project is led by DNV, a leading classification society, in partnership with ship owners Altera Infrastructure and Klaveness, who are respectively shuttle oil tanker and combination carrier operators in this project, ScoutDI, an inspection drone supplier, and NTNU through the Autonomous Robots Lab at the department for cybernetics.

The project will develop the concept of a digital twin of the hull condition for assessing the structural integrity and compliance of the physical hull. The digital twin is kept up to date with the physical hull over its lifetime by frequently collecting large amounts of data, in non-intrusive ways that do not interrupt business operations. This enables continuous monitoring and compliance assessment. Autonomous drones collect the data instead of humans to improve personnel safety. Automated damage detection and coating breakdown assessment assists the human expert, leading to improved quality of the structural integrity assessment. The project will enable digital twin-based surveys as industrial standard and realize economic and safety benefits as well as reducing environmental footprint due to reduced need for travels. The innovation will lead to improved quality and safety of vessel inspection services, in particular: 1. Hull condition assessment and continuous compliance decision in a 3D digital twin, a multi-data digital representation of the physical hull based on data such as high-precision point clouds, videos, AIS and weather data. The digital twin also holds location-tagged thickness measurements, history of damages and repairs, and data of sister vessels. 2. High coverage, high-quality, location-tagged data of the hull compartment, collected frequently by autonomous/automated drones with semantic knowledge of the physical structure. Automated data collection ensures standardized data which improves comparison with historical data. Drone-based collection during sailing or between discharge and loading of cargo causes minimal interruption of business operations. 3. Automated damage detection in point clouds and videos, automated damage probability predictions based on data in the digital twin, and an automatically generated targeted inspection plan for the drones as well when physical inspections are deemed necessary.