if you have fatty acid, you need to convert to ester so it will be volatile in the injector port and go into the gc column. Normally, I use FID and calculate all  peak as 100%. I determine percentage of each with the total area.

I mean the standards should be FA then when converted to FAME it will compensate the recovery of the hydrolysis and derivatization step

What is the sample you're trying to analyse? Hydrolysis to the fatty acids, concentration, and derivatization should be uneventful and demonstrated by a method validation. Internal standards would be overkill, most likely. Converting your FA standards to a mixed FAME standard (or concentration series) shouldn't be an issue; anyway, as noted above, you could probably routinely use a single one, and rely on historical relative response factors to quantitate the rest. Be vigilant for non-volatiles crudding up your injector liner, though!

Standrad should be  Fatty Acid Methyl Ester 

This is a complex question. If you are analysing oil or fat samples, the main components are not fatty acids. Triglycerides and other lipid classes are the main compounds, mainly triglycerides in general. This means that you are doing a transesterfication by different procedures to obtain FAMEs. In this case the use of a triglyceride as an internal standard could be useful if it is used from the beginning of the procedure. Of course to validate the method from the point of view of linearity, repeatability and uncertainty is not correct. In this case we need to use a FAMEs standard containing all the fatty acids of interest or to prepare a mixture with non derivatized fatty acids (containing all of interest) to obtain the FAMEs and after to analyse by GC/FID at different concentrations and with the typical quimometrics procedures. On the other hand, the different methods to obtain FAMEs  do not produce the same result with a non finished debate about what is the best or more appropriated to each kind of sample. Also, with the possibility to use a GC/MS this analysis could be improved because the problems to identify minor fatty acids that is not possible with FID. 

Standards for fatty acid profile are methyl-ester's derived (FAME)

that means using triglyceride as internal stander to compensate for the recovery(hydrolysis,derivatization)and FA standards which will be converted to FAME to make calibration is the best way to go

thank you all for helping me

You  must to analyze methyl esters of fatty acids and determinate quantitatively with any of the methods described by Mario Vincenzo Russo. Methyl ester of fatty acid 23:0  is other posibility of internal standard.

 i extract an oil and i want to analyse its fatty acids by hplc, what kind of standards i need ? and which is preferred single fatty acid standard ? or a kit?