On boats this works as they are floating in this nice conductive stuff called water (with dissolved salts/ions..) to close the circuit so the sacrificial metal gets 'eaten' before the other metals so to speak.. 
On cars the best you can do is as described above.. Clean/remove the corrosion, apply a galvanic corrosion barrier/inhibitor like DuraLac or TefGel, re-assemble and then after it has all hardened/dried, for good measure, spray the exposed joints with a wax sealing compound like waxoyl or tectyl.
The ali corrosion is partly galvanic corrosion caused by the dis-similar metal contact, but part of it is also crevice corrosion where water trapped against an ali surface and in an oxygen-deprived environment starts to basically 'etch' the metal and create an ongoing reaction and stinky white oozing residue.
The last one (crevice corrosion) does NOT need any contact with other metals, just 'trapped' water on an ali surface. The floor corrosion issue on S1 (and early S2) elises because of water trapped between rubber floor mats and the ali floor is a prime example.
Of course once a galvanic corrosion reaction (usually this is sped up by contact with water or brine/road salt) has created small voids in the contact area between the metals, water can get in and the crevice corrosion kicks in as well.
The chassis - subframe interface can be prone to both types of corrosion and the double-hit of duralac on the mating surfaces and then, once assembled, a wax coating should stop both types pretty much from occurring though.
Luckily, once you clean the corrosion off well and treat the surface, aluminium will just be fine and not react further by itself unlike steel where the rust devil never sleeps and keeps on gnawing/growing from every tiny fleck of rust..
Bye, Arno.
, not going to claim I'm surprised!
