Continuing the series of posts on the analysis of offshore wind turbine installation methods, this post looks at the improvements, if any, which can be gained by increasing the carrying capacity of the installation vessel. It is thought that time spent transiting between the vessels home port and the wind farm could be better spent on performing the installation operations, particularly in the summer months when favourable weather occurs. To do this we’re looking at increasing the on station time by allowing the vessel to carry more monopiles and transition pieces.
The Simulation
The station keeping transit and access/depart capabilities of the vessel have not changed since we first established them in Part 1 of this series. In fact, the assumption we’re making here is that the vessel is the same one, rather than a new larger vessel, and that the project engineers have invested time in designing grillages which allow the effective storage of more components.
To model this in Mermaid we are assuming that the new grillage has no impact on the loading or at site operations, and we’re assuming that it is as easy to load and rig components as it was previously. The image below shows the changes to the flow diagram that have been made, which are:
- The loading cycle contains a new fully suspendable group “Perform Load”. The tasks in this group are as previously, but are now repeated 3 times, i.e. we’re assuming we can carry 3 monopiles and 3 transition piece. Having this group be fully suspendable preserves the previous behaviour. It’s also worth noting that we still only have one loading of j-tubes and grout. We’re designing the operation so that a container of these items is loaded and that single load contains sufficient for one excursion.
- We’ve set the “Installation Cycle” to repeat 3 times, this means that we perform three lots of these operations on each trip out from port, working through the locations specified.
- We’ve reduced the number of repeats on the “Foundation Installation” group from 72 to 24. The nested repeating groups, with 3 installations per loop, mean that the overall plan only need to be done 24 times to install all the foundations.
Everything else in the analysis is the same, so with these basic changes performed the analysis can be run.
The Results
In the post “Project Progress Burn Up Charts” I discussed a new feature we’re adding to the results side of Mermaid, this work was on going at the time of this analysis so I’ve processed my data outside Mermaid. It’s worth noting that the outputs from simulations are available to users in their raw format, this being a zip file of csv’s, and that these are not hidden or blocked in any way. This means that they can be integrated into other simulation software, modelling or scripted post processing. To produce the chart presented here I wrote a short Python script to get the data into the format I needed and used Excel as my charting tool.
Firstly I’ll describe the content of the charts, using the image of the base case which is shown below. These results were presented as a box-whisker plot in the last post. The chart shows the number of days after the project start (x-axis) at which each foundation was completed (y-axis). This gives us an idea of both when work was performed and of how quickly we’re getting the foundations in. A number of lines are shown: The blue lines represent “winter” months with the bold line displaying the median; the green lines represent “summer” months. The dashed lines represent each individual month (with the data grouped in the same manner as in the previously presented box-whiskers).
This base case shows us the duration of the installation in each case and we can see, as we already know, that winter months are quicker. We can see the progress change as work moves between poor and good weather as the line becomes steeper when installations are completed more swiftly. The steeper this line, and the further to the left the end point is, the better, and this is what we are targeting as we perform these optimisation simulations.
Presented below are the results for the simulation of a larger carrying capacity. There are a few points to note:
- In all cases the installation process is quicker. Each line terminates further to the left than the corresponding line on the base case chart.
- The progress lines are much less smooth, there is a definite wave to them. We can see steep sections where 3 foundations are installed fairly quickly, followed by flatter sections where we are transiting and re-loading.
- Starting the installation in summer is still faster at the start of the operation, but as the summer ends and the winter weather arrived the progress slows. The inverse is true of the winter installations. These start slowly as we are affected by the weather before speeding up, ultimately to finish faster than the summer installations.
This last point is worth some additional consideration, we’ll do this below the chart.
The Next Steps
Being able to carry more components on the deck of the vessel improves the performance noticeably, however:
- This is dependent on being able to engineer suitable grillage for this storage and on the vessel being physically capable of fitting and carrying this additional load.
- There is a cost associated with this design and mobilisation, although this may be less than the reduction in cost seen by reducing the hire time (in fact this is likely to be the case). So far we haven’t considered the cost associated with the work, this is a like for like comparison on the vessel and we’ve been happy to assume that shorter is better.
- We are still being affected by the poorer winter weather, and installation processes which start in winter are achieving a higher level of performance as they can access more of the summer season.
This last point suggests that we can improve our operation further if we can make more effective use of favourable weather, this is something we haven’t fully managed so far and that we’ll look at next time.