11 replies to this topic

[mamad]

Hi evey body,
I am going to design the new compressor station for a gas plant. I want to compare the availability of different compressor arrangement but I do not know how exactly I should do. For example how can I calculate the availability of 2+0 arrangement or 2+1 in parallel and series.

[cuchuoi2005]

Hi Mamad

The availability of your system can be calculated using the database given in OREDA. From that you can select the availability of the basic component then you can build a simple logic like (CompA AND CompC) for 2x50% config, for 3x50% config the logic will be ((Comp A AND (Comp B or Comp C)) OR ((Comp A or CompC) AND (CompC)). If you don't understand this logic I suggest you to read the relevant books about Availability and Reliability.

Good luck
cuchuoi

Edited by cuchuoi2005, 17 March 2012 - 08:50 AM.

[Art Montemayor]

Mamad:

If you have to design something for the first time, you probably are a student. If you are a student, this thread should be in the Student Forum. Are you a student? No one in industry is assigned a project such as yours for the first time.

You have failed to tell us all the Basic Data required to address your query. Some of the data is:

• WHAT are you compressing? Give us the name, composition, pressure, & temperature.
• WHAT KIND of compressor are you writing about? This is very basic data and no comment can be generated without knowing this.
• WHAT SIZE (CAPACITY) of compressors are you writing about?
• WHAT IS YOUR SCOPE OF WORK? You cannot make a decision on compressor arrangement without knowing what is expected of the compressor station. For example, is one of the compressors a stand-by? Is it a spare? What kind of spare? Is it expected to remove one machine out of production while the others operate? Etc., etc., etc...........

No one can address - and much less comment - on the sparse information you have furnished in 3 simple sentences.

[mamad]

cuchuoi:

Thanks a lot.

But how can i provide this software?
and also what is the main aspect of this software?

Could you please explain a little more for me.

[mamad]

Art Montemayor:

Thanks a lot for your precise answer.

You are right. I have not given details about my plan due to I thought that my main question is vivid. Again, I state that I would like to calculate the availability of different arrangements for compressors. However, Here I answer your questions respectively:

1. Obsolutely the fluid is gas. This gas is processed in the plant and then is feed to the reservoir with inlet T=40 C and P=120 bar along with outlet condition T=60 C and P=170 bar. The type of gas, as I mentioned earlier, is processed and therefore it is lean gas with MW~19.
2. The compressor type, based on previous experience, preferably is considered to be centrifugal, having said that, there is no limitation for the type of compressor.
3. The total available gas is 950 MMSCFD. However, because the arrangement is not chosen yet, the capacity of each uint is not clear.
4. I think your question is related directly to my main question, on the grounds that after understanding of availablity for each arrangement and thus economical issues, we be able to choose the best arrangement and other related aspects.

I hope that I could specify my plan; however, in case of any question do not be hesitated.

[kkala]

Probably example of thread http://www.cheresour...ty-calculation/ '> http://www.cheresour...ty-calculation/ can help. But you have to estimate operational availability of each compressor individually. Suppliers can give operational availability; even if it may be on the high side, you can get an idea of the size. Plant experience can be a better indication, if available.
Suppose you have 3 compressors, each one of availability p. This means that probability to be able to operate (or to be in operation) is p. If the needs are covered by two compressors, operational availability of two compressors to be operating (or operable) is
C1, C2 working, C3 out of order probability = p*p*(1-p)
C1, C3 working, C2 out of order probability as above
C2, C3 working, C1 out of order probability as above
C1, C2, C3 operable probability p*p*p
Total probability p*p*p+3*(1-p)*p*p = 3*p^2-2*p^3 = operational availability.
More examples can be seen in mentioned thread

[cuchuoi2005]

Mamad

OREDA is a database of operating equipment. There are two version hardcopy and softcopy. The softcopy is more flexible and allows you to filter at more accurate level and you can also check the population of data (normally people uses data with population >30). In a word this will provide you the fundamental basis for your calculation.

As you can see from kkala response, the calculation is simple and doesn't need any software to perform this task. For more complicated system I used to use software called P-Choke.

Regards
cuchuoi

[mamad]

kkala:

Thanks for your response,

I totally agree with you in calculating the availability. I study a book which is named "The Universal Generating Function in Reliability Analysis and Optimization", in this book the writer generates a universal function that can be used for calculating of reliability in any engineering discipline.

By using this function I got the same results as you. Also I read the thread that you mentioned and it was so useful. Again, Thank you for your reply

[kkala]

I think the most significant task is to estimate individual (operational) availability p of (say) one 50% compressor, or one 100% compressor. In many cases operational availability of the compressor of half capacity is considered same as that of the full capacity. In theory availability increases somehow with capacity, but this also depends on individual quality of the equipment.
Could you advise the author of the mentioned book, and if possible publisher? In this way a member interesting in such matters could find the book.

Edited by kkala, 07 April 2012 - 02:48 AM.

[mamad]

I think the most significant task is to estimate individual (operational) availability p of (say) one 50% compressor, or one 100% compressor. In many cases operational availability of the compressor of half capacity is considered same as that of the full capacity. In theory availability increases somehow with capacity, but this also depends on individual quality of the equipment.
Could you advise the author of the mentioned book, and if possible publisher? In this way a member interesting in such matters could find the book.

Of course I do,
The author of this book is "Gregory Levitin" and the publisher is "Springer".

[kkala]

Thank you, mamad, referred book is available in current market, quite varying price http://www.bookfinde...mization&ps=bp'> http://www.bookfinder.com/search/?ac=sl&st=sl&ref=bf_s2_a1_t1_1&qi=yEgLEsC5MviQdl0829xdc7cm8ec_0782448332_1:34:709&bq=author%3Dgregory%2520levitin%26title%3Duniversal%2520generating%2520function%2520in%2520reliability%2520analysis%2520and%2520optimization&ps=bp.

[mamad]

Mamad OREDA is a database of operating equipment. There are two version hardcopy and softcopy. The softcopy is more flexible and allows you to filter at more accurate level and you can also check the population of data (normally people uses data with population >30). In a word this will provide you the fundamental basis for your calculation. As you can see from kkala response, the calculation is simple and doesn't need any software to perform this task. For more complicated system I used to use software called P-Choke. Regards cuchuoi

I reviewed the mentioned book, I mean Oreda Handbook, but I could not understand the method for calculating availability in this book. Would you mind explaning to me about the method of use of this handbook?