I N NOVATI N G WH I LE DR I LLI N G
MindMesh has added real-time prediction of downhole dysfunctions and drilling dynamics to its digital twin. In the simulations
shown here, an operator used the platform to identify ways it could drill with higher ROP without inducing lateral shock.
we’re drilling. Can we take a snapshot and use that to explain a
movie that’s two hours long, or 10 days long? That’s what we’ve
been trying to do.”
Last year, MindMesh updated its RiMo digital twin platform to
incorporate real-time prediction of drilling mechanics and down-
hole dysfunctions. RiMo is a time-domain model, meaning that it
analyzes the behavior of the BHA and the drill string as it moves
downhole. It integrates predictive models with a near-real-time
data stream through a WITSML feed, as well as parallel comput-
ing capabilities, so it can update the digital twin during drilling
while predicting potential dysfunctions downhole.
Users can build a workflow that automatically recognizes
drilling rig states and records quantifiable drilling parameters
like shock and vibration, downhole MSE and ROP in near-real
time. These events then act as cues for the predictive model. For
instance, if predicted bit RPM, downhole MSE and ROP reach a
certain threshold indicative of stick/slip, the predictive model
will alert users of a potential incident. The model then runs a
simulation, known as a “drill-off test,” where users can adjust
various operating parameters to see how it affects performance
indicators and adjust them at the rig.
Mr Gandikota touted the system’s ease of use, noting that few
inputs from the user are required. “As soon as you input the BHA,
all you have to do is set up a dashboard. We’ve tried to make this
more visual and easier to map with the real-time surface data. At
the end of the day, once you’ve installed it, it becomes a routine
process. All the information is standardized. All you have to input
is what type of bit you’re using and what type of formation you’re
looking at, and then you’re ready to go.”
In 2020, an operator in the Permian used the software to
improve its drilling operations. The operator was experiencing
high frequencies of lateral shock, where the lateral motion of the
BHA reaches a high enough amplitude to force the BHA to bounce
against the wellbore. This was occurring whenever ROP reached
100 ft/hr, so the operator used the digital twin to examine how it
could increase ROP while keeping lateral vibration low.
The predictive model analyzed instances of lateral shock
26 as ROP approached 100 ft/hr, and the operator chose to test the
impact of WOB and top-drive RPM on ROP and lateral vibration.
The model then ran two optimization scenarios in a drill-off test,
one in which the WOB was increased in 10% intervals from the
initial WOB measured at the 100 ft/hr ROP, and top-drive RPM was
decreased, also in 10% intervals.
As WOB was increased, the ROP increased from 100 ft/hr in
the initial scenario to 160 ft/hr in the first optimization and 190
ft/hr in the second optimization. The operator used lateral shock
counts above a threshold to indicate severity in vibration over
a time interval. In the initial measurement at 100 ft/hr ROP, the
model measured 40 lateral shocks. It then predicted 37 lateral
shocks in the first optimization scenario and 24 lateral shocks
in the second scenario. The operator was persuaded to increase
WOB and decrease top-drive RPM to help increase ROP and
reduce lateral shocks.
The digital twin platform is currently being used by two opera-
tors and two directional drilling companies, all in the Permian
Basin. Two drilling contractors are also trialing the platform
ahead of potential full-time deployment on their rig fleets later
this year, according to Mr Gandikota. DC
Click here to watch a video with
MindMesh’s Raju Gandikota discussing
the company’s digital twin.
M A R C H/A P R I L 202 2 • D R I L L I N G C O N T R AC T O R