4 replies to this topic

[tafolabi]

Hi everyone,

 

I need to perform a plant test, how do I go about doing that? I know I need the temperature, pressure, composition and flowrate at strategic points. I also have created a simulation for the plant so as to compare how the real plant is operating as opposed to the simulated plant. I am a chemical engineer and just graduated from college, my job experience is 0-1. If you have ever had to perform a plant test, what are the other basic things that need to be considered other than material and energy balance?

Edited by tafolabi, 28 March 2013 - 08:09 AM.

[Erwin APRIANDI]

_{You may required more than material and energy balance, you will required lots of other stuff to try to mimic the plant running.
You also should know what process that happen on each equipment, and to make a control condirion to have the same operating conditions
 
So please give a background on why you are doing it, and where are you going to do it, lab or others and who will be guiding you along the process, since doing a plant test can be dangerous.}

Edited by erwin.apriandi, 28 March 2013 - 03:55 AM.

[thorium90]

I assume your plant test means some performance test? If so, you should compare it with what the designers have guaranteed it should do and better still if they have provided a plant simulation or data from similar plants and not from your own self made untested simulation

Edited by thorium90, 28 March 2013 - 08:17 AM.

[tafolabi]

Thanks,

 

The plant is an ngl processing plant (amine sweetening, dehydration,all the works that go into processing ngl etc). I am tasked with doing the plant/ performance test so that I can gain better understanding of how the plant works as well as report to my supervisor on whether or not the plant is running as it should. This project is solely mine and it is up to me to come up with all the things I would need/ do to get accurate readings of the plant. I will enroll the help of the operators at the plant and I also called the labs to come in during the plant test to gather samples. Also, if anyone has a sample of the plant test they performed as a guide that will also go a long way in helping.

[paulhorth]

tafolabi,

I know this type of plant, I have designed several such plants. You have a difficult assignment.

First I would say that it is surprisingly difficult to take an accurate mass balance from an operating plant.  To do this you first need to find where all the various vent and drain streams and other offtakes are and ensure that they are closed or else have a means of measuring them. This includes the acid gas from the amine unit, the fuel gas, the waste water etc. Then you need to get good calibration for the flowmeters and learn how these instruments are corrected for temerature, pressure and density to get a reading in mass flow units (not volume flow). Then you need to get a good composition analysis of the feed and product streams.

Calibrating the temp and pressure instruments is important, if possible. You can easily find two temperature instruments on the same stream with different readings - which one do you believe?

Level instrument output is always given as Percent which is useless until you have the Range, and the zero setting. If you want to know the flow of a stream from a vessel,such as a compressor KO drum, which has no flow measurement, you can do it by closing the valve and timing the rise in level between two points, if the operators will let you do this. You could cause a high level trip which will shut down the plant.

Make sure you can get printed historic trends from the control system of as many variables as you can manage. Then if there is an unexpected spike somewhere you could try to correlate it with readings of other variables at the same time.

Measure the flows of utility streams to the reboilers and the supply and return temperatures, so you can do a heat balance on the columns. Doing a heat balance on an aircooler is going to be difficult.

Good luck

 

Paul