I can see why you would want a pressure compensated flow control valve now – presumably you want the speed to remain constant regardless of the force needed to push the sample. You might, however, be disappointed with the performance of the simple pressure compensated flow control valves which are readily available. The speed regulation with varying pressure is OK for general use but I wouldn’t have a great deal of faith in it for materials testing. Similarly the speed regulation with varying temperature is also likely to be a bit underwhelming (and your oil is going to get quite hot with this method of control).
Basic pressure compensated flow controls also have a nasty characteristic where the cylinder always starts with a jerk. Explanation: while the cylinder is depressurised the compensator in the flow control valve opens fully, when you switch your DCV to pressurise the circuit the compensator needs a finite amount of time to get back to the proper regulating position – during this period the flow is too high hence the jerk. The jerkiness of the initial movement might spoil your test results.
You could go quite high tech and fit some proportional (or servo) valves, then use a stroke transducer on the cylinder and a closed loop speed control system (analogue or digital). But you might find that a simpler (yet still adequate) solution would be to fit a basic electronic speed controller to the electric motor driving your pump.