Let’s take a look at the results .The wellbores are spread along the arbitrary dimensions in the time-depth clustering space, with regards to their similarity. Considering the heterogeneity among wellbores, we are displaying the selected similar wellbores in two graphs with range limits corresponding to :
- The maximum values of time and depth of the entire wellbores (first graph)
- The maximum values of time and depth of the selected wellbores (second graph)
The results suggest that the model is able to capture the pattern of the whole time-depth curve rather than addressing only the top and bottom depth of operations. This way we can easily translate complex feature relationships into similarity coordinates for our wellbores. All you have to do is select a bunch of wellbores in our time-depth clustering map and get insight and explore the endless machine driven opportunities.