Flare capacity problem
Flare capacity problem
(OP)
Dear all,
In our refinery we have some PSV designed to discharge to atmosphere, most of them are on top of ADU, FCCU and visbreaking main fractionators. Because we consider that as safety and environmental issue we want to connect them to flare. This is problem because when we calculated effluent quantities we got result that we need to build new flare.
Is there any ideas how to connect this PSV to flare and avoid building new flare?
Regards,
Milutin
Talk To Other Members
RE: Flare capacity problem
RE: Flare capacity problem
RE: Flare capacity problem
RE: Flare capacity problem
I would look at all the scenarios which result in relief from CDU fractionator overhead, and then evaluate if applying Instrumented Protective Function will result in decreased relief load (I would expect so). The key concept of IPF/IPS is to have an independent system in place to protect the Unit from any unwanted scenarios, by triggering certain actions (e.g. position of valves etc).
Usually, the highest loads result from either loss of cooling or excessive/light/water feed to the unit. Therefore an IPS would act by stopping the feed to the unit, and shutting down the heater. This is just a simplified example of course - all the scenarios and corresponding protective functions would need to be evaluated in detail.
The second concern I have is whether your flare network system can tolerate liquid phase (if discharged during reduced relief conditions), and what further measures/actions need to be taken in that direction.
http://antwrp.gsfc.nasa.gov/apod/astropix.html
RE: Flare capacity problem
For FCCU main fractionator loss of electricity going to cause loss overhead cooling and for sure relief to flare, is it acceptable to backup power source with diesel generator to prevent loss of cooling?
For ADU unit it is same scenario for power loss, and you are right excessive water in crude oil can trigger increase of pressure and popping up PSV.
Regards,
Milutin
RE: Flare capacity problem
Dynamic simulations are required to determine how quick the effect of restored power supply will be on preventing a major relief, i.e. if the EDG can cut in fast enough after electrical power loss. Dynamic simulations will also be required for evaluating system behavior in case the IPF's are implemented.
Still (or in addition to EDG), I would look for implementing IPF's since their integrity and reliability is very high and you want a system that will do what it is supposed to do, in times when you desperately need it. If the relief from CDU or FCCU can rupture the Flare header and release a huge vapor cloud that will possibly ignite, I would look for several layers of protection - starting the EDG, closing the feed valve, shutting down the heaters, etc.
http://antwrp.gsfc.nasa.gov/apod/astropix.html
RE: Flare capacity problem
I've made several refinery flare studies with the purpose of solving problems like yours.
There is a huge amount of work to do in re-analyzing all the contributions to flare loads in the various conditions, including but not limited to reliability/integrity and dynamic simulations.
Very good results can be obtained but this is very depending by the current conditions of the refinery plants (and the flare system. of course).
You can obtain various papers relevant to this subject by an internet search.
However, the help of experts is needed to follow this kind of projects.
If you want to go more deeply in your investigation, please see my profile.
Good luck.