Digital Energy Journal- Akselos – engineering simulation digital twins

Join | Existing Member Login

You are Home » News » View Article

Akselos – engineering simulation digital twins

Monday, December 2, 2024

Akselos makes 'engineering simulation digital twins' - high resolution digital models of assets subject to stresses - which can be used to optimise designs and assess remaining operational life

Akselos, an engineering simulation technology company, has a technology to quickly develop 'structural digital twins' to assess strength of structures, optimise designs and calculate remaining operational life.

It commercialises a method developed at Massachusetts Institute of Technology to conduct Finite Element Analysis calculations much faster, by creating a reduced representation of the analysis that 'filters out' irrelevant data.

Akselos has offices in Lausanne (Switzerland), London, Houston, Boston and Ho Chi Minh.

Finite Element Analysis (FEA) is a standard technique used in engineering, based around splitting a structure into tiny 'finite elements' and analysing them individually, then putting the analysis together to get an understanding of the strength of the overall structure. This is a widely used approach, but it is also well-known that FEA is highly computationally intensive which puts major limitations on the size and/or detail level of analysis that can be performed.

Akselos' product is based around what it calls 'Reduced Basis FEA', which uses a reduced order modelling approach to break through the computational limitations of FEA and to enable highly detailed modelling of even the largest assets.

As an example, the company presented a project to assess whether a pressure vessel is safe, based on having new 'C-SPAN' ultrasonic inspection data, measuring wall thickness.

Its model of the pressure vessel is highly detailed, with over 25 million 'degrees of freedom', and hence cannot be solved with conventional FEA in a practical manner. This model is detailed enough to incorporate all of the C-SPAN inspection data, and due to the acceleration provided by RB-FEA the digital twin with inspection data incorporated can be analysed in a few seconds.

The software runs on the cloud.

When you add in new thickness scan data, you can see where material is thinning due to corrosion. You can do a quick calculation to identify high stress locations. The software can make a report 'assessment of metal loss'. You can prove you can still operate the pressure vessel safely.

Because models can be assessed so quickly, it is possible to try out many different models, when designing a new asset - such as with different material properties, thickness, geometry and loads.

This means you can reduce the amount of 'over conservatism' you need, where you specify material to be thicker than the calculation says, to compensate for elements you are unsure about.

The software can use the model to calculate different failure modes resulting from cracks and corrosion.

Engineers can focus on models, while management can focus on the reports.

It has an 'API 579' plugin which can do 'fitness for service' assessment and auto-generate a report that summarizes key findings. This standards-based analysis includes both linear elastic and nonlinear (e.g. elastoplasticity), and thermal (e.g. thermal stress) analysis.

The process could be considered making a 'structural digital twin' of the asset, incorporating the original design together with any updates, such as recent inspection data.

By observing trends in the structure over time, such as how fast metal is corroding, together with the structural calculations, it is possible to estimate how much longer the asset can be safely used.

Akselos has one project with a FPSO operator, to build a high fidelity model, which will show a representation of the whole FPSO, not separate localised models.

It has also been used on wind turbine designs. On an 8 MW wind turbine, one blade weighs 29 tonnes. Akselos can be used to optimise the structure of the blade, drive train and support column, trying out multiple designs and improving them. You can check any deviations to your plan to see what impact they would have.

Ryan Perdue, senior director of Akselos, and a mechanical engineer with a background in oil and gas, sees digital twins as a 'missing link between data and information'.

Many asset monitoring systems have very low accuracy, because people struggle to get any meaning or structure out of the data, he said. In contrast, Akselos's RB-FEA is based on first principles of physics and hence provides robust analysis and reporting for critical assets.

The software can absorb all of the asset data through design, commissioning, operations and decommissioning.

Associated Companies
» Akselos

CREATE A FREE MEMBERSHIP

To attend our free events, receive our newsletter, and receive the free colour Digital Energy Journal.