11 replies to this topic

[alfirin]

Hello everyone,

I have an assignment to evaluate the heat recovery potentials in a crude oil distillation unit (picture attached). My tasks are to use the pinch analysis technique to identify the ideal maximum furnace inlet temperature, without taking into account costs, and calculate Qrec. Second task is to design a HEN to reach or get close to the ideal furnace inlet temp. Only duty and area must be specified.

Data: Extra heat provided by furnace, duty to be calculated
Extra cooling by cooling water, duty to be determined
DT = 20 C
Process stream Data (attached)

The basic design indicates that the crude oil enters the furnace at a Temperature of 290 C, but with a good design it is supposed to go over 300 C.

I use the above data in my simulation software and it calculates the composite curves (attached). I find the pinch, the recovery potential, the heat duty of the furnace and the cooling water. Now, I use the graph to identify the inlet and outlet Temperatures of the utilities. I get Tin = 295 C and Tout = 355 C for the furnace and Tin = 45 C and Tout = 95 C for the cooling water.

Am i missing something so far?

The problem is that I get an inlet temperature of 295 C and I know from literature that this can be way beyond 300 C. Do i have to complete another step to identify the inlet Temp in the furnace, or is it 295 C the max I can get?

As far as the second task is concerned, I guess after calculating the data of the utilities I have to put them in the stream table and proceed with the HEN design. I will post any questions that may arise during this step here, but meanwhile could somebody assist me in the first task?

Thanks in advance.

Attached Files

•  Data_Crude_oil.xls   20KB   60 downloads

Edited by alfirin, 08 April 2011 - 08:35 AM.

[pavanayi]

alfirin,

What software did you use?


How did you estimate the hot utility temperatures and cooling water temperatures?
I do not think there will be any plant in the world with cooling water return temperature at 95°C.

What you get from the composite curve is this much: You need external heating between 295°C and 355°C, and external cooling between 95°C and 45°C. Am I right?

As per the graph, yes your furnace inlet temperature is 295°C.

[Art Montemayor]

Alfirin:

You are still a student, so here is some practical advice:

If you seriously are using Celsius degrees in your cooling water outlet value, then you have made a fatal error. The ability to heat ANY process cooling water to that level of temperature does not exist in the practical, real world of engineering. The highest design cooling water outlet temperatures I have ever seen in over 50 years of engineering is 120-125 ^oF (49 -51 ^oC). 100 -110 ^oF are usually the highest actually allowed in practice. The reason for these limits is that all process water - indeed, ALL water - has dissolved solids and other contaminants associated with it and these affect the heat transfer to the point where they can make it impossible to operate if allowed to exit at higher temperatures. Solids will precipitate out in the heat exchanger and literally plug it all up - into a solid mass! You clearly can not allow that as a practical engineer.

There are other qualities of water that you can use, for example boiler feedwater which is used to generate steam (boil). But they are NOT PROCESS COOLING WATER. There is a big, big difference in the associated qualities of the different waters. Do not fall into that bear trap.

[alfirin]

Thank you both for your answers and comments.

I'm new in the pinch analysis and HEN design, that is why I'm still confused with utilities.

Obviously, I can't identify the inlet and outlet temperatures of the hot and cold utilities using the graph. My question is whether the utility dutys are enough to help me with my HEN design. Is there a way to estimate the inlet and outlet temperatures or is it possible to make assumptions?

The problem is that the software requires the utilities to be used as streams. Therefore, it needs from me duties and inlet-outlet temperatures. I'm using the SPRINT software developed by the centre of process integration developed by the Uni. of Manchester.

Thanks a lot for the comments about the cooling water. I didn't know that these temperatures are impossible.

Regards

[Art Montemayor]

Alfirin:

You don't have to offer any explanation for your ignorance in this issue. You are still a student, and you guys are not expected to dwell on or think about the practical side and what you are supposed to design and expect as soon as you step out of those hallowed halls of ivy. I had the same handicap when I was in university and I started out as a graduate. It isn't fair, but universities and engineering profs have their weaknesses. Some professors know everything there is to know about theory - but lack the practical expertise and fail to point out the real-world realities to students. That's life as it exists. And that's why I interjected my comments into this thread to make you and other students aware of why we are here and how you can gain a better insight and preparation by using this Forum.

Be kind with your profs. They can only teach you what they know - if they are good and loyal to their calling. But you must start - sooner better than later - to learn to question, analyze, reason, and challenge whatever you sense does not make for common sense or real life. That - for better or for worse - is engineering at its best. The good and dedicated profs will value your ability to challenge and question. They know that it is a sign of a good, quality future engineer and will respond accordingly. We engineers make it happen - regardless of the theory or the hard realities in life. Our job is to overcome the problems that exist because of life's realities and deviations from theory. We rely A LOT ON COMMON SENSE and hard facts to resolve those problems. The more tips and advice you can assimilate from the Forum, the easier and faster that things will start to come together and make sense in your career of solving other people's problems.

Good luck.

[pavanayi]

Alfirin,
as far as this stage of your assignment is concerned,
For cold utility, you should assume a initial and final temperatures, and then calculate the flow rate (since the enthalpy is fixed). As a suggestion, take cooling water inlet temperature as 20°C and return temperature as 30°C
For hot utility, since you are using a fired heater (furnace), you should assume suitable temperatures ( as an example, Tin as 1200°C, Tout as 200°C) This should be included in your grand composite curve, and then you can fit your hot utility requirement to this hot utility.

When using heater as utility, you have to be aware that even though the T out is 200°C, due to utility pinch, you might not be able to use the entire temperature range, and be limited by the pinch temperature. (The best person to ask, if in doubt about this, is Mr Steve Doyle. You should be knowing whom I am referring to )
SPRINT is one of the best software you can get for pinch analysis (personal experience)

The utility duties are enough to help you with HEN Design. If you are stuck somewhere, I can help as far as I can.

Edited by pavanayi, 20 March 2011 - 05:42 PM.

[alfirin]

Thanks for the tips!

SPRINT is indeed a great software and so is STAR, which I'll be using in my second assingment soon

At first, I thought to use the furnace temperature range you indicated, but I wasn't sure if I should go that 'far'. I believed it wouldn't be neccesary to do estimations.

I'll work on it tomorrow morning and I will post here my development! Thanks for the help. I try to use the forum before I ask quentions to my supervisors, as this may be helpful to other students in the future!

Regards

[alfirin]

Dear all,

I did some simulations these days and I try to figure out my mistakes. The problem is that the software designs more than 80 exchangers, which is way too much I believe. I try to put a minimum value for exchanger duty to avoid very small duties.

Anyways, I managed to find some utility costs for water and fuel oil in $/m3. since I know the utility dutys I calculate flow rates and by using Cp values and net calorific values I find out the the volume of utilities per m3. Then I can convert it to $/kW.yr as this is the input values of the software.

My supervisor told me that the HEN should not have many splits, because the plant control won't be able to manage them effectively. So far I get 4-5 splits in one of my cold streams which gives another 4 splits in each one. I don't know how to fix that yet.

I was also told that there is another way to design the HEN: Split the problem in two -above and below the pinch - and design two different HEN. Then combine them together to make the final one. If someone has a clue how to do this, it would be nice to help me!

Finally, I used some correlations for the exchangers' capital costs. However, my supervisor adviced me that the exchangers connected in the sector of the steam ejectors and surface condensers have a much higher capital cost compared to the ones in the rest of the unit. Does someone have a clue about it or knows where I can access some costs?

Regards

Edited by alfirin, 24 March 2011 - 06:25 AM.

[pavanayi]

Alfirin,
I can give you some tips. It might prove useful. Instead of putting all the streams in for the automatic heat exchanger network synthesis,
Make an assumption that the reboilers will be using steam, and cooling water in condensers. (For the side strippers, preflash column etc). This means you can remove those streams. No splits on those streams either.
With the rest of the streams, attempt the network synthesis and see what happens

Edited by pavanayi, 25 March 2011 - 08:24 AM.

[alfirin]

Indeed Mr Pavanayi, this is a good idea. I don't get more than 2 splits in the hot streams. The streams that gets the most are the cold process streams. So ,maybe I should constraint the hot streams as you mentioned to connect only with the cooling water. I will give it a try this weekend.

Thanks again!

[alfirin]

Dear all,

My supervisor adviced me to try optimizing the HEN without using the automated network desing option. This means that I have to change some probabilities and try to find the best one which will give me the desired temperatures and utility consumptions.

So far I have managed to reach an inlet temperature of 291.5 C and the utility consumptions are close to the minimum. hope I will find a better solution today, otherwise I'll seek advise from Steve Doyle, as Mr Pavanayi said!

[alfirin]

In my last optimisations I managed to reach an inlet temperature of 294.2 C, which I think is a good value with a small error. My HEN contains 29 heat exchangers with utility consumpions close to the minimum.

I have attached a file with my network data, if someone wants to have a look and see if there is something strange in the calculations!

Regards

Attached Files

•  Network report.xls   34KB   33 downloads