If I understand correctly, we have to solve the electricity supply of 3 systems, one of which is mobile, from a generator operating cyclically.
That is to say three stations connected in parallel (generator in central position) and a mobile circulating on either side of this line.
It is assumed that the water and oxygen stations are connected in parallel to the generator by copper cables buried below the surface (-170 ° C). (provide this function on the mobile).
One station operates permanently (oxygen), one only during the "day" (water) and the mobile is considered to be functioning potentially permanently (except recharge cycle).
The cyclic generator is supposed to provide enough energy (10800 Megajoules) to cover the needs of the different systems over the 708 hour cycle. One or more energy accumulators are therefore necessary to ensure the continuity of operation of the deoxidizer and of the charging station (s) of the mobile (s) (see end of text: one or two mobile).
For the mobile, we first consider that its charging station will be the "central" position (generator) and that it will be there during the period of sunshine producing the energy.
It works mainly in the crater near the intermittent station in order to supply it with "wet" regolith, and on the other hand on the edge of the crater in order to deliver "dry" regolith to the continuously operating site located near the generator.
For the two fixed consuming stations, one is therefore at a distance equal to 1 unit (permanent operation) and the other at a distance equal to 10 units (operation in "day" mode).
The consuming station which operates continuously and which is located one unit away is subject to variable thermal conditions and is the one which consumes the most (process = +/- 3800W continuously). +/- 3050 Megajoules must be stored in order to be restored when the generator is on standby. It will consume +/- 5000kg of "dry" regolith and reject +/- 4250kg of deoxidized waste at 1100 ° C for every 709 hours of operation.
The consuming station operating intermittently and which is located ten units away is supposedly subjected to little thermal variation: permanent low temperatures, 4 ° K, mainly radiative exchanges (ideal place for an inertial accumulator with HTSC technologies). It comes in second order of consumption with 800 Megajoules per cycle (300h / 709h) or an average power of +/- 750W. It will consume 1500kg of "wet" regolith and reject 1470kg of "dry" regolith at 150 ° C for every 709 hours of operation.
The consumption of the mobile is therefore the main variable depending on the order in which it will perform the various operations and where it can draw the energy necessary for its operation and at what period (day / night). Indeed, one parameter is: is it desirable to abandon 1470kg of "dry" regolith at 150 ° C for every 709 hours of operation within the crater? This could constitute a handicap in the search for "wet" regolith in the perimeter around the water production station, the mobile having to increase its search perimeter for "wet" raw material.
If we consider that the mobile must go up with the dry matter, it can thus supply the deoxidizer with +/- 1/3 of the "hot" mass necessary (200 ° C input ???) to perform a cycle. the mobile will also take advantage of the residual thermal energy to maintain the temperature necessary for its operation. In return, the mobile will consume more power to perform the 10h ascent.
One can also imagine that all or part of the "hot" dry matter could be used to maintain the temperature of the water-producing station during its night standby. The dry matter may or may not be collected afterwards.
If the mobile does not need to come out of the crater, another mobile which will remain at the top of the crater would be useful in order to supply the desoxydeur with "dry" matter taken outside the crater.
Useful data would also be the times required to carry out the various collections of materials in order to determine the recharge times and the accumulation capacity.
Until now and, considering the different hypotheses, all the residual energy of the waste from the deoxidizer still remains to be exploited. Accumulators and peltier elements used as generators could therefore still provide additional power.
It would in fact be necessary to establish a matrix taking into account all these different elements (and technological data) in order to determine which is the best solution in terms of yield and compared it to other solutions in terms of flexibility. Unfortunately, I don't have the time or the skills ...
If this summary is correct (to check!) And that it can be useful for some, I am happy and ask no more!
Thanks for your comments.