3 replies to this topic

[Rahimzadeh]

Dears:

Our gas export compressor station consist 7 compressor 5 running 2 stand by (5+2),we want change it now

case 6 running to 1 stand by (6+1).

Please let us ,what will be consequence?

Besr

M.R.Rahimzadeh

[Art Montemayor]

M.R.Rahimzadeh:

 

The question being asked, I believe, is one of strategically determining what is the "best" (reliable service+economic+contractual+operational) compressor configuration you can have to conform efficiently and reliably to your sales contracts and obligations.

 

If that - in effect - is your query, then my reply is that it is a question that only you can respond to given that you - and not us - have all the required and necessary scope data inputs required to address that question.  For example:

• What is the "export" condition?  Are you responsible for meeting gas sales contract conditions?  If so, what are ALL the conditions?
• What is the volume rate of gas being sold?
• What are the penalties linked to gas delivery failure?
• What size (capacity & Hp) and type of gas compressors are you using?
• What size and type of drivers are employed?
• What is the age and mechanical condition of each compressor and driver?
• What is the maintenance and upkeep records on each of the compressors in question?
• Do you do your own maintenance or is it out sourced?
• What critical spare parts are kept for immediate availability in your facilities?
• What is the level of gas compressor expertise on your staff and operating personnel?

There are many more questions that will normally arise as a result of how some of the above questions are answered.  So, my response is that this query is so broad and undefined with basic data that it can't be addressed specifically by persons who are not stakeholders in the operation.

[Bobby Strain]

Search for RAM analysis. You may find what you need with this analysis. Along with all that Art suggests.

 

Bobby

[pavanayi]

Art and Bobby have already given you enough topics to think about. If I was looking at this topic, my first point of investigation (data collection rather) is to see how many times in the past (5 years, 10 years etc) have you had a scenario where two compressors have been down simultaneously? It may have happened when one compressor was out for maintenance (preventive or corrective) and before it could be made ready for service, another compressor failed and the remaining stand-by compressor had to be taken online? If it did happen, what was the reason for the failure?

 

Even though the answers to that question is in no-way an indication of future performance (see Art's questions like age and maintenance regime etc), at least you'll have a direction to start thinking about it.

 

Reliability Availability Maintainability (RAM) analysis is a statistical method to perform the same analysis, based on historical data of equipment failure rate to predict the probability of plant up-time based on factors like maintenance and/or redundancy, and will present the answer to your question satisfactorily.

 

You or anyone interested in RAM should have a few things in mind. The numbers coming out of the RAM analysis is (as with everything else) based on the quality of the data used and validity of the assumptions. If you have historical information regarding the failure probabilities of equipment in your plant, then the results will be mostly acceptable. But in reality, most organisations (even though they may record everything in SAP or a similar tool) say they don't have historical data in a usable form and opt to use the default database available in all RAM tools. This data is almost always from the OREDA database, which is mostly based on failure history of equipment in offshore platforms.

 

You may be predicting the probability of failure of a pump in a chemicals plant in a desert location with ambient temperatures >40°C based on data of pumps failing on an offshore platform in the North Sea. 

Edited by pavanayi, 13 January 2017 - 08:58 AM.