In the water cycle power from the sun causes water to evaporate and then disperses that water in the form of precipitation over a varied topography; some high spots and some not so high. It's not humanly possible to calculate how much energy is used from the sun to get water at what ever point in a stream it finally ends up at. I have no idea where the water comes from that feed mountain springs that are at a very high elevation that constantly flow and feed streams. I don't think it's humanly possible to calculate the energy used to cause these springs to have a source either.
What is possible to calculate is the force of a stream at a given point as gravity pulls that water down hill. That would be the starting point of defining perpetual motion. I don't think that the force of the stream over time (gallons per minute, etc.) can drive a water wheel such that the wheel produces more force over time than it receives. It only captures a part of that force; somewhere less than 100% of the force.
The analogy of a water to a gravity wheel breaks down in the idea of a 'stream'. As far as we know there's no 'stream' of gravity. We have no idea what the source of its force or attraction is. What we know about gravity is that it exerts with monotonous regularity a downward force on a weight as it swings thru its various moments on an arm around an axis. These moments of force are depicted by a sine wave. I'm pointing out the obvious to make this point:
With all the forces (gravity, centrifigual, momentum, etc.) acting on a mass thru it's moments on a moment arm you will not get the weight to swing and arrive back at (much less thru) zero degrees.
If that could happen then instead of the mass starting from a dead stop it would have a little head start. As the velocity accumulated the mass would eventually be spinning so fast it would rip free of its axis. It would look like one of those working model designs. :)
As a sidebar I'd like to say that since there are people here with experience with working model this is a good place to post your problems with it. Someone just might have an answer for you. That seems to be the case a lot.
The point of a mass spinning thru its moments is this:
If you can cause a mass to start from zero plus 1/10 of a degree (heading downhill) dropping it and allowing it to fall then if it arrives back at zero with some degree of velocity so that it can round that hill and head back down with some speed you will cause that mass to accelerate to its terminal velocity and you will have achieved perpetual motion.
No one as far as we know (except Bessler) has managed to do that. It's considered impossible. The forces on a mass as it swings around an axis are conserved. It's impossible (so they say) to arrive back at zero degrees with more velocity than you left there with.
My point of all this and my concluding point is this:
The randi organization will never agree to a sensible definition of perpetual motion. You will never collect a nickle from them. On top of that they will give you a bill for their expenses. They'll publish your idea and use it as they see fit and never fail to mention you're the sort of person that won't pay your bills.
A. Gene Young