The scientists who invent time travel presumably know how it works, and it is made clear that what already happened is always going to happen.
Well, that's not necessarily true. Science is a process of successive approximations. Just because you know how to do something, that doesn't mean you know everything about it; you have a model good enough to work, but there's always more to learn, stuff you won't discover until you put it into practice and gather firsthand data.
Our current real-world theories tell us that altering history would be impossible unless
nonlinear quantum mechanics exists, which as far as we know, it doesn't. But maybe some future discovery will show that nonlinear QM does exist, at least in some atypical circumstances. Sometimes it's just a matter of coming up with the right way to tweak the equations. Scientists used to insist that faster-than-light drive is impossible. Now NASA's already doing a proof-of-concept experiment for warping space. We've understood relativity and the lightspeed limit for over a century, but that understanding has undergone continued refinement, and some very smart people have thought of ways to bend the rules, so that what was once assumed absolutely impossible is now just considered extremely unlikely to be practical, but worth testing out in principle.