I'm personally in agreement with Bjorn Schrammel, in as much as a "contruction yard" would make most sense first. I'd suggest a little lower in orbit, however. Only by about 100,000 meters. This is also done without math etc, and purely by similar projects in KSP.
If you're not wasting resources lifting from surface, and instead mining, refining and producing materials in space, then construction of enhanced density radiation sheilding isn't so much of an issue, and modular station components can be launched further up to about the 350,000 mark and assembled on site. Constant adjustment of velocity will be a minor thing with something like a Q-thruster system deployed.
To assume a permenantly inhabited, long term residental station, I disagree with the L1 placement. Mostly because that's a single point of failure. Sensibly, there'd be a collection of stations, evenly distributed aorund the orbital path. Each station vastly able to exceed it's citizen's draw, it will be able to act as a nearby "lifeboat" destinaiton in case of some critical failure that causes abandonment, splitting a stations occupants to those either side...
And velocity is the key determinant of kinetic energy, but however, it's relative velocity. Where you are in regards to orbital distance doesn't matter half as much as the direction the projectile is comming from, and the speed it's doing in relation to you.