Single VFD control two motors of compressor?

Q: We have two compressors with each motor 10HP, can we control all of them by a single variable frequency drive, like 20HP?

What you are attempting to do is very difficult. It is OK to run multiple simple drive motors on a single variable frequency drive (VFD) because you add up the motor loads and then you must use a Eutectic (bi-metallic) Overload for each motor after the VFD. The larger VFD for multiple motors together can not clearly detect only one motor overloading so a separate motor protection is required for each motor after the VFD. They should be the bi-metallic type because a VFD usually has very good short circuit protection or phase loss protection, better than an electronic OL. An electronic OL is usually never used after a VFD because of the Pulse-Width-Modulation output made by the VFD. It is not a standard sine wave through the electronic OL and does not work well. A bi-metallic OL heats up in the same way the motor coils heat up so they will detect each motor OL or fault and better protect the motors individually.

Now comes the second problem that I could not solve on this blog. You are powering COMPRESSORS. I am assuming that they are reciprocating compressors. When the compressor strokes forward to compress the gas, the motor is doing WORK. When the compressor is stroking back, there is a period when the remaining gas under pressure is expanding and the compressor is pushing back against to motor, creating a GENERATOR. This work against the motor is generating power and it has no where to go. The motor and VFD (and bi-metallic OLs) will get HOT. You might need a heater coil on the cables at the motor to release the heat that is being electrically generated. This is a solution for a SINGLE MOTOR compressor. But you have 2 compressors. they will "mechanically" get out of phase with each other, meaning that the mechanical phasing (compression WORK stroke versus gas expansion back stroke) will not all be at the same time from the two parallel motors and the Electrical generation wave will cross-talk across the motors. This will cause many other electrical and heating problems. The cross-talk is passing back through the bi-metallic OLs and happens at the common branch point. The variable frequency drive will see all sorts of strange voltage and current spikes, I am not sure it can even be done. It might "work OK" for a while, but it is not the way to go.

If you look at all the things you need to do to save a little money on one larger 20 HP VFD versus 2 smaller 10 HP VFDs in parallel, and add in all the proper protections, you are probably spending more money and will have more problems trying to control multiple motors on one VFD.
Also, separating the two systems and running them in parallel separates the single-point failure risk so you can still run one unit if the other one fails. You can control on a single loop to run them together or you can adjust the different compressors to blend your rates if you use two variable frequency drives.
I would seriously consider doing this with 2 separate VFDs of the proper size for each single motor, you will not need eutectic OLs, you have no single point failure risk, and all you need is a heater coil to release the heat from the power generated by the single compressor expansion stroke.