One thing you may not be aware of yet is that small scale test data does not scale linearly up to large motors. There are multiple reasons for this, outside the scope of this thread. But for the moment you can take my word for it if you choose.
Some practical consequences:
1) The formula you develop for a small motor is likely deficient in some possibly important ways for a larger motor.
2) Scaling too much too fast leads to results deviating from expectation. I recommend scaling no more than 4x at a time.
3) Larger motors have problems that in a practical sense don't really exist in smaller ones. The required level of engineering is not a constant!
4) Motors under acceleration have additional issues compared to static tests.
5) Because of the scaling problems, and flight related problems, you should expect to have to perform at least one if not many more tests at each level before moving to the next level. That is, at least one static test and one flight test. And if your results are not within a pretty narrow tolerance of what you expect, test again until you can hit your tolerance range before scaling up again.
So you are going to go through many times the propellant mass you are anticipating. Getting through the process will take years. It will take you years to get the experience to be able to reliably pull it off.
My last several EX motors - the flight ones in the N and O range - burned to within 0.1 second of planned duration, and the rockets they were flown in came in at under 3% altitude error, usually closer to 1%. These are APCP motors, first burns. Testing was with smaller motors but within acceptible (to me based on my experience) scaling factors. And I do make corrections during the scaling...
In my EX hybrid I failed to compute one detail - just one - and had a cold flow start and delayed ignition. I didn't realize until I computed it, that I needed roughly 1MW of power for the ignition profile I was aiming for. I assumed instead of computed, and was off a couple digits on my WAG. It worked, but not fast enough. At least it was a static test, and no harm done.
The larger the motor, the more complex. The more complex, the more you have to compute EVERYTHING, or else perform lots of static and flight tests to determine what works in practice. Either approach is ok. Combined they are better.
I think I have enough experience to tackle making a first attempt at a motor perhaps a quarter the size of the one you are contemplating. I'd expect to take a couple years to get it done to be flight worthy to where I know exactly how it is going to perform. And it would be expensive... and I'd want a team of helpers at that scale... and a dedicated fabrication facility... and a test facility.
There are others here who are commenting on this thread who have more experience than I have. In some cases, a lot more!
IMHO, I'd expect you to take at least 5 years to get to where I am now, assuming you have at least a degree in a relevant field such as physics, chemistry, or some hard engineering discipline, and dedicate yourself to learning everything. It might take a little longer otherwise.
I might be off somewhere, but from ground zero - where you essentially are now - I'd expect this to be a 10 year project if you do it right and don't get distracted. If you half-ass it, it will just be an expensive CATO or worse.
I'm not trying to discourage you. Quite the contrary! Just trying to provide one opinion on a bit of calibration of the scale of what you want to attempt.
Cheers!
Gerald