Proposal of new algorithm nd workflow for handling Uadd's

[scheidan]

Idea to tackle Uadd's (#50)

The problem with Uadd's is the combinatorial explosion. Image that for source U1 we find N and for source U2 M systems if both are used independently (i.e. not in combination). The systems of this sources could be combined in N*M ways!

However, the systems we care most are the once with some synergies. That means systems that are partly similar for U1 and U2 so that they can share techs. With the following algorithm we would check for every combination if synergies are possible (in most cases this will not be the case) and only if this is the case construct a new, combined system.

Let say sys1 is a system of U1 and sys2 from U2

1. Do sys1 and sys2 share some techs? If not, break
2. Are the shared techs connected in the same order? If not, break
3. Build a combined system
   3.1. combine techs and connections (without dublications)
   3.2. add new connections
   3.3 update SAS and other properties

Workflow

systems1 = build_systems(U1, techs)
systems2 = build_systems(U1, techs)
syn_sys = build_synergy_systems(systems1, systems2)
all_sys = hcat(systems1, systems2, syn_sys)

[spuhlerd]

First step to understand possible synergies and how many to expect:

1. Calculate as suggested all Sys for U1 and all Sys for U2 (which is a Uadd).
2. Find synergies manually for U1 systems and U2 systems of preference.
   To be tested with Master thesis starting May.