Vibration Monitoring in Ships

PoC/Research owned by Viking Analytics English
2y ago update
Read more at https://vikinganalytics.se/user-cases/vibration-monitoring-in-ships/
Summary Objectives Solution Outcome
Based on vibration data collectedin a group of fans in a ship, MultiViz Vibration detected the ones that should be prioritized for inspection. 

Being able to properly  monitor machine health allows for better maintenance scheduling to reduce  unnecessary interventions in the machines,prevent accidents and increase the resting time of the vessel’s crew

This use case was developed under AI4EU, an European Union Artificial Intelligence project, which aims to bring together the AI community and facilitate knowledge transfer from research to business application. It provides a platform for experiments that promote the sharing and deployment of AI models.

Monitoring the health of the machine is a complex problem, given that many faults can be combined and affect the typically monitored parameters in different ways. In ships, the main challenge is requiring a vibration monitoring expert on board, which may not be feasible given the increase in overhead costs.  Furthermore, data transmission is expensive and difficult due to synchronization issues and the limited internet coverage at sea.
The assets monitored were a group of fans. Ships normally do not have many different types of fans, so the equipment was classified in three categories mainly. In total, data was received from 1160 assets. Each asset had two or three tri-axial vibration sensors on them. Given the constraints on data transmission, only 1-second-long measurements per axis and measuring point were available.

Given the large number of similar assets and the fact that they remain constant across all ships, the first approach was to find the assets with most atypical behavior based on vibration data. This is done through our feature Blacksheep Detection, which identifies machines that behave differently from the rest of the population. It then displays a list of the assets according to how atypical each of them is likely to become.

As the measurements were recorded with a high resolution, it was possible to split the 1-second-long measurements into smaller signals defined by the number of rotations on the asset. Then, all the signals from all measurement points and axis were concatenated to represent a single source.

Then, each individual asset was subjected to Automatic Mode Identification, which automatically identifies all the operational modes in a machine and selects the most relevant time stamps in each mode for analysis. As all measurement points and axis were transformed into a single source, Mode Identification enables the expert to identify which measurement points and axis may be worth inspecting first.

Being able to properly monitor machine health allows for better maintenance scheduling to reduce unnecessary interventions in the machines, prevent accidents and increase the resting time of the vessel’s crew.

Attributes

Transportation & Logistics
Operations
More Efficient
Professional
Optimization
Machine Learning
Sensor Data