I've seen a few times when a differential protection relay tripped out a breaker during the initial 'core charge' inrush. Rare, but it does happen.
The older style 'spinning disc' type of relays had some sort of dampening of the disc, sometimes adjustable, usually not, to allow for the one-sided inrush. The newer electronic relays all have time delays, usually related to current (I/T). If they're set for too short of time, they'll see a core charge as a fault.
Here's the problem; a differential protection scheme doesn't actually protect anything within the protected zone. It mainly protects whatever feeds the protected zone, and certainly can minimize the magnitude of a fault. (A smaller explosion, sort of).
Here's why; in order for a differential relay to operate, a fault needs to have already occurred. It can't see a problem before it happens. For this reason, any time delay needs to be as short as possible. If, for example, the relay is protecting a bus-duct, you'd set it to trip instantaneously. There isn't much of a core charge in a set of busses. A transformer or a motor however, must magnetize it's core before it can induce current on the other side. If the relay is set to no time delay, or too short of a delay, it'll see this magnetizing (core charge) current as a fault. On the other hand, if the delay is set too long, and an actual fault occurs, There'll be a much brighter flash, and a much bigger bang!
In my experience, setting up a differential protection relay for an inductive device is more of an art than a science. Just because it allowed the breaker to stay closed for the last 50 times doesn't mean it won't trip this time. It all depends on where the 3 phase voltages are when the breaker contacts actually close vs. any residual magnetism within the core. Generally speaking (from experience) if the core is completely de-magnetized, the chance of the breaker tripping out on differential is much greater. I might get beat to death for this opinion, but this is one of the VERY few times I'd ever advise closing a breaker in a second time without extensive testing, especially if ground fault and overcurrent protection also exists, and neither of these caused the trip.