Title- Ship Propulsion System on Reliability and Condition
Monitoring and Digital Twins
Introduction:
The paper discusses the importance of the trade chain through ships and the frequent damage
of propulsion systems on various occasions. It has shown the study of world trade through
shipment and the causes of accidents related to the propulsion system. The reports are
submitted related to Maritime transportation plays a critical role in global trade, responsible
for transporting approximately 80-90% of goods worldwide. A key component of maritime
operations is the ship propulsion system, which has been identified as the most vulnerable
part of the vessel, with failures accounting for over a third of shipping incidents between
2010 and 2019. Failures of propulsion systems are particularly problematic in extreme
environments, such as Arctic waters, where nearly half of all incidents are caused by
propulsion issues.
Report related to casualty:
The Lloyd’s List Intelligence Casualty Statistics reported 26,071 shipping incidents between
2010-2019. More than a third (35 %) of these incidents were caused by damage or failure of
the propulsion system, which is over twice as many as the next highest cause, collision.
Thus, it becomes clear that the maintenance of a ship’s propulsion system is relevant for the
value chain and therefore it is stated that the improvement of the maintenance system leads
to a
reduction of incidents. The developing trend is increasingly progressing from reactive
maintenance to scheduled maintenance and predictive maintenance. Predictive maintenance,
enabled through condition monitoring, is an alternative to scheduled maintenance which
offers effective early detection of failure. This translates to increased reliability and decreased
maintenance costs and reduces unexpected shut-downs.
Figure 1 provides a stacked plot of the total shipping losses worldwide over the past decade.
The number of total losses has been declining over the past decade, with foundering
accounting for the majority of losses (75 % in 2019). Factors contributing to the foundering
included bad weather, flooding, engine problems, and capsizing. Figure 2 provides a
breakdown of the incidents that occurred worldwide in shipping from 2015 to 2019. The
main cause of reported incidents has consistently been damage to, or failure of, the
propulsion machinery.
Propulsion Maintenace and measurement:
Maintenance strategies for ship propulsion systems include routine or planned
maintenance, condition-based maintenance, and unplanned maintenance. Routine
maintenance follows a fixed schedule, condition-based maintenance relies on real-time data
to determine when
repairs are needed, and unplanned maintenance occurs when unexpected failures happen.
The methods like prognostics, help predict the remaining useful life (RUL) of components,
allowing maintenance to be scheduled proactively to prevent failures. Key monitoring
methods used in the industry include vibration monitoring to detect machinery issues, oil
analysis to assess lubrication health, and thermography to identify abnormal heat patterns
that signal potential problems. Together, these standards and practices emphasize the shift
toward condition-based maintenance, which enhances system reliability, reduces the risk of
unexpected failures, and ultimately improves maritime safety and efficiency.
Problem statement:
Propulsion system failures are a leading cause of incidents in the maritime industry, with
significant financial and operational implications. The traditional approach of reactive or
scheduled maintenance is often inadequate for preventing these failures. Therefore, there is a
pressing need to develop advanced condition monitoring and predictive maintenance
strategies that can detect potential issues early and prevent unexpected shutdowns.
Conditioning monitoring:
The shipping industry is transitioning from time-based to preventative and condition-based
maintenance strategies to improve reliability and reduce failures in marine propulsion
systems. Preventative maintenance follows schedules recommended by component
manufacturers, as seen in the example of the S.A. Agulhas II, which mainly employs
scheduled maintenance but also monitors key parameters like oil, bearing, and
motor temperatures.
Condition monitoring methods for propulsion components include vibration analysis,
temperature measurement, oil quality assessment, and advanced techniques like Acoustic
Emission (AE). Vibration analysis is commonly used for detecting faults in electrical
components, propulsion shafts, and plain bearings, identifying issues like misalignment,
unbalance, and wear. For electrical machines, Motor Current Signature Analysis (MCSA) and
temperature monitoring are key in fault detection.
Role of Digital Twins in Condition Monitoring
Digital twins are virtual entities connected to physical assets, allowing them to replicate the
behavior of these assets through measurements, models, and decision-making tools. They can
be physics-based, data-driven, or hybrid, with hybrid models combining the strengths of
high-fidelity simulations and machine learning predictions. One key advantage of digital
twins is their ability to create virtual sensors that can predict parameters that cannot be
directly measured, enabling analysis of conditions without historical data. This technology's
adoption for predictive maintenance and condition monitoring is expected to enhance the
efficiency and reliability of marine propulsion systems and other critical applications.
Digital twin challenges:
Digital twins face several challenges, particularly when implemented in the maritime
industry. Data-driven digital twins require large amounts of data from similar or identical
assets, while physics-based twins are constrained by model detail and can become
computationally expensive. Hybrid digital twins, which combine data-driven and physics-
based approaches, offer a balanced solution by leveraging the strengths of both methods to
predict parameters like Remaining Useful Life (RUL) under various conditions. Successful
implementation of digital twins requires several elements: dynamic and degradation
models of components, system models, measurement infrastructure, models that correlate
environmental conditions with loads, predictive models of future operations, and feedback
mechanisms to inform engineers of changes in RUL. These elements ensure the digital twin
accurately mirrors the physical asset’s behavior, improving predictive maintenance and
operational efficiency.
Conclusion
The integration of condition monitoring and digital twins into maritime operations presents a
transformative approach to propulsion system maintenance. By shifting from reactive to
predictive maintenance, the shipping industry can significantly reduce incidents, improve
reliability, and enhance overall operational efficiency. The continued development of digital
twins, combined with advanced data analytics and simulation capabilities, will be essential in
addressing the challenges of monitoring ship propulsion systems, especially in extreme
environmental conditions.
