The answer from the mid 1700s has been that the tides are caused by the Moon,
and the tides move the Moon away.
Well maybe, but, probably not.
Both the Moon and the Earth gain mass by accretion, so both have increasing
cummulative gravitational force. The Earth is just over 81.3 times the mass
of the Moon, so it gains more mass at a faster rate. This gradually changes
the balance of power. That is, the Earth applies a gradually increasing force
on the Moon that is far larger than the gradually increasing force that the
Moon applies on the Earth. There would be four equations to solve, but,
two (Lunar) are very small, and 2 (Earth) are large.
The result is that the Moon moves away from the Earth at a net increase in
distance of 38.2 +/- 0.7 mm per year. A simplified equation that uses the
81.3 / 82.3 and the 1 / 82.3 would be Earth moves about 0.5 mm down plus
the Moon moves about 38.7 up, so the net change is the 38.2 up (away).
When using the concepts of conservation of (anything and everything), we always forget to use the limitation "in a closed system".
The Earth and the Moon, as well as all other objects orbiting a larger
sphere anywhere in the Universe, are all exposed to an "open system".
The exposure to "open system" conditions to both materials, and energy
means that conditions, and "ratios" of things change. The mass ratio changes,
the applied gravitational force ratio changes, the orbital distances change,
the angular momentums change, the kinetic energies change. The potential
energies change.
What must remain relatively constant is that the alignment must remain
co-linear. That is Earth, E-M Barycenter, Moon must remain co-linear,
while at the same time the orbital distances and velocities are changing due to
the changing mass and gravitationally applied force ratios.
Back to the original question, we can use the Solar and Lunar applied forces
as they create winds, and waves, and thermal heating, but there is an upper
limit to how much we can use. We could never reach the point of
using all that is available, so it would never reach the definition of
perpetual motion machines. There are always inefficiencies and losses
in any and every system.