You're not the first to go down this route. Laithwaite in the 60's tried to make a train based on these ideas, as did the Japs. I saw Laithwaite's forlorn scrap at Cranfield in the 70's.
They got <30% the efficiency of conventional motors, and a used lot of complicated expensive unreliable electronics, so the idea has been quietly dropped.
Suggest you do the same.
You're never going to get round the terrible efficiency of the induced magnetic loop. Whatever you do , the gaps have to be large, which is a BAD idea. A train trundling along at 100mph is unlikely to be able to maintain a <1mm gap, even if the track is perfect. And what about curves? How will you upgrade existing track and install your driving strip to those accuracies? What about distortion in extremes of heat and cold?And any disgruntled yob could ruin it with a few decent thumps with a sledgehammer.
A linear BLDC might be worth a second look - with a segmented steel track and variable frequency drive. But to what gain over much cheaper and more efficient conventional systems?
Especially when we're all meant to care about energy consumption......
Linear motors do have a place - for example in 'flying faders' for mixdown on music recording consoles. It's been done. But not for transport.
Allan