Agreed, there is no absolute frame of reference, but regardless of initial boundary conditions the rate of change of kinetic energy changes with time if acceleration is constant?
Which is irrelevant because the mistake you are still making is assuming that the kinetic energy change of the craft is somewhat proportional to the energy that is provided by the power plant. They are not directly linked.
See my example. And note it was only a single example, same craft in the same place with the same vector in the same universe doing the same acceleration. All my example does is view this from a number of different frames of reference, there are no variations in "initial boundary conditions" but the
change in kinetic energy calculated from each of those equally valid frames of reference is radically different: 50 KJ, 150 KJ, 10 MJ, -50 KJ. All of these are the same example just measured from a different viewpoint.
And also there is no reason why the frame of reference needs to be constant. From the frame of the craft's passengers the acceleration is always "from zero" instant by instant.
My point is that you cannot say : "the power plant generates 50 KJ of energy (after losses) to the thruster and the craft gains 50 KJ of Kinetic energy". The change in kinetic energy is different in different frames of reference. There is always one frame of reference where the craft does gain 50 KJ of KE but there is another equally valid frame of reference where the craft
loses 10 GJ of kinetic energy when you give its thruster 50 KJ of energy. The energy provided by a power plant is invariant (in a Galilean sense) but the Kinetic Energy is not.
For the same reason you cannot generally say "it takes more power plant energy to accelerate at a fixed rate at high velocities because the change in kinetic energy is greater" (see later for Thruster plates). Power plant energy (invariant) and kinetic energy (frame of reference variant) are not the same. The higher/increasing velocity in one frame of reference
could be is a lower/decreasing velocity in many other frames of reference.
For example let's take your original example:
Example: Accelerate a 1 Dton ≈ 10 tonne craft by 1 G ≈ 10 m/s² for an hour = 3600 s:
Velocity achieved is v = at = 10 m/s² × 3600 s = 36000 m/s.
E = mv²/2 = 10000 kg × (36000 m/s)² / 2 = 6 480 000 000 000 J = 6480 GJ.
Average power is 6480 GJ / 3600 s = 1.8 GJ/s = 1.8 GW = 1800 MW = 1 800 000 kW.
Accelerate for two hours and the result is:
Velocity achieved is v = at = 10 m/s² × 7200 s = 72000 m/s.
E = mv²/2 = 10000 kg × (72000 m/s)² / 2 = 25 920 000 000 000 J = 25 920 GJ.
Average power is 25920 GJ / 7200 s = 3.6 GJ/s = 3.6 GW = 3600 MW = 3 600 000 kW.
But we will start at a frame of reference with a vector of 50000m/s in the direction of acceleration without actually changing anything real.
From this viewpoint the 10 tonne craft starts with a velocity of -50000 m/s, kinetic energy of 12500 GJ.
After one hour of acceleration it has reduced its velocity to 14000m/s, kinetic energy: 980 GJ (
reduced by 11520 GJ), average Power -3.2 GJ (minus).
After two hours the craft is travelling at 22000 m/s in our frame of reference. Kinetic energy is 2420 GJ (10 GJ less than the starting KE and at one point it actually dropped to zero before rising again).
Nothing real has changed in this example from your earlier calculations but your general supposition that power plant energy increases (feeds into) kinetic energy is false. The power plant has not transmuted into an energy sink in this example. It is the same example as you provided earlier when you calculated that the kinetic energy would increase. I made no changes.
The crux of this is that we always need to consider the whole mass system not just the primary (craft) but the reaction mass as well. The reaction mass has its own, mass, velocity, momentum and kinetic energy. When we do this the invariant Power Plant energy is distributed between both the primary mass and the reaction mass (both receive equal and opposing momentum). The total change in kinetic energy of
both the primary and the reaction mass equals the total propulsion (power plant) energy but the proportional distribution depends on the frame of reference.
When we have a frame of reference where the kinetic energy
seems to be is reducing it because in that frame of reference the reaction mass is gaining more energy than the power plant is providing (the combined primary loss and reaction excess cancel out).
Conclusion: The power consumed by the M-drive does not generate the acceleration of the craft directly, presumably it is used to trick the outside of the craft (system) to accelerate the craft. Hence the achieved acceleration cannot be used to estimate the power required according to our current technology.
Again not necessarily true - the primary craft can gain more kinetic energy that the power plant provides so long as this is balanced by the reaction mass losing kinetic energy - e.g when a hurtling rocket ejects a reaction mass at an ejection velocity close to or slower than its travel velocity in a specific frame - so the ejected reaction mass seems to have reduced velocity and has lost kinetic energy relative to its velocity/KE when it was a component of the rocket mass.
But I want to circle back. In Traveller there are two basic drive models: Reaction mass drives and Thruster Plates.
Reaction Mass Drives (CT Fusion drives, HEPlaR) emit mass carried from the ship so they constitute a closed system. Propulsion energy basically goes into accelerating and ejecting the reaction mass and this can be modelled from an accelerating frame of reference anchored to the rocket. If the propulsion energy put into the reaction mass is constant for the same mass emission rate then its acceleration, exit velocity, kinetic energy are constant and the momentum it generates for the craft's acceleration is also constant, regardless of the external frame of reference or apparent velocity in any classic frame of reference at least until the rocket equation starts to kick in.
Thruster Plates (a.k.a Space Bicycle Drive) pushes against nearby Planetary/Stellar objects through some field. The local Planetary System in this case is the reaction mass. In this case the relative velocity of the craft and the reaction mass (i.e the average motion of the surrounding environment) does impact the kinetic energy and the propulsion input will have an impact on velocity generated. If the ship and the reaction mass have a positive relative velocity (ship generally leaving the system) then the acceleration will reduce as the relative velocity increases. If the ship is moving into the system mass then the acceleration per power provided should increase. This is also complicated by the localisation of the field which may be decoupling from some mass as it moves away and finding new mass as the craft moves through the system. Also if the thruster tends to decouple from mass that has a large relative velocity as it accelerates (it cannot "grip" it) it may also recouple to nearby mass that has a similar velocity to the craft's vector and therefore gives it more grip.
There was also an abomination of a reactionless thrust mechanism in MegaTraveller but I am not going to address that and all its infinite energy/perpetual motion implications.