What:
Predict high flow events in a oil and gas trunk line from platforms to on-shore.  Avoid fouling gas system.

How:
Use platform pressures, temperatures and flows to predict the likelihood of high liquid events.

Results:
Significant event (large slug) predictions are 99-100% accurate.

Take Action

	Call 1-952-746-5761
	Request Information
	Tech Requirements
	Get a Quotation

Recommended Products
Process Modeler
Process Intellect

Not Finding What You are Looking For?
   Tell Us!
(that's how we improve)

Separating the Oil from the Gas
Most every multi-phase oil and gas pipeline has a "slug catcher" at the end.  Slug catchers handle unusually high liquid flows that discharge from the pipeline from time-to-time.  These transient liquid flows occur because oil accumulates within the line and is forced out at seemingly sporadic intervals.  Slug occurrence, frequency and magnitude has a lot to do with the transported gas-liquid characteristics, the gas and oil flow rates and the pipeline geometry.

Unexpectedly Large Slugs Can Foul Up the Works
If an unexpectedly large liquid slug enters the slug catcher, it can be overloaded allowing liquid to get into the gas system.  This can create a serious problem due to the gas system "drowning".  Lost production from the unnecessary shutdowns and difficult and expensive cleanings may be required.

Complex Dynamics Easily Predicted.
We have found that significantly high flows (big slugs) can be predicted very accurately, even at fairly long time intervals in advance.  Our non-linear dynamic modeling technology makes is quite easy.  Large slug alerts enable operators to prepare, driving down the liquid level in the slug catcher to avoid overflows, lost production and difficult cleanings.

The image below shows predicted vs. actual liquid flow to a slug catcher 3 hours in advance based on off-shore platform conditions:

The above results are "out-of-sample", meaning the model was not constructed using these conditions.

The image below shows predicted vs. actual significantly high flow events to a slug catcher 3 hours in advance, where 1.0 is a statistically large flow event and 0.0 is non-eventful flow.  The model was created using historical data from the feeding off-shore platforms' conditions:

Purple is the model's prediction and black is the actual event occurrence.  You can see in nearly every case the model alerted operations to a high flow event 3 hours in advance, allowing for operations to run down the slug catcher to avoid fouling the gas system.  These results are "out-of-sample", meaning the model was not constructed using these events.