When building a complex process model engineers sometimes loose track of the complete picture what they are building. The individual parts of the model may be perfectly allright, but when you put them together without considering the overall process, things can go badly wrong.
I recently saw an example of this where sour gas was stripped from a crude oil in a stabiliser tower. The model of this tower was done in more detail than usual and the overhead system was somewhat complex. The stripped gas from this first part of the model was re-compressed for reinjection. Things got problematic when the liquid formed in the aftercooler of each compression stage was knocked out in a separator and was fed back into the oil circuit that was feeding the stabiliser column. The conditions after the highest pressure stage after cooler were such that half of the sour gas was condensed and hence sent back into the oil circuit. Of course, all this condensed sour gas is subsequently stripped out again in the stabiliser column and sets up a BIG loop of sour gas. This continues until the number of iterations in the recycle is exceeded or until the column fails to solve.
To solve this you have to look at the big picture: the sour gas should not be present in the bottom of the stabiliser and the only other way out of the process is through the gas phase leaving the highest pressure compressor stage. So the conditions there need to be such that the sour gas can effectively leave.
The solution is fairly simple, make sure the outlet temperature of the last stage after cooler is such that only a small amount (or no) sour gas is condensed. Going through this experience is not necessarily bad, it may alert the engineer to the fact that the plant engineers and operators should be made aware of the risk they run if in winter time for example the cooler is performing better than expected.
Posts
No comments:
Post a Comment