But surely if no feeder station is included that means no power west of Cardiff to Swansea should an isolation be required say at Bridgend - potentially impacting on the maintenance time for the OLE going forward.
Does new the Hitachi depot in Swansea (which ironically is home to what were probably the first OHLE masts errected in Wales) have a feeder station? If not, was one planned? Otherwise, as you suggest, an isolation would presumably have cut off everything west of the isolation point.
Can the electrification methods being used by Keilos Amey on the Valley lines (e.g. discontinuous electrification, with trains running on batteries under bridges and other unelectrified sections) be applied to Cardiff - Swansea to bring the costs down? Is this what the article is advocating?
The article mentions that, but also the bridge just west of Cardiff Central where the main line passes under the line to Queen Street. At the latter, the OHLE (I believe) is (or will be) live but the underside of the bridge is insulated to avoid rebuilding the bridge. Personally I think that approach should be used, rather than requiring all electric trains on the Cardiff-Swansea route to have a battery mode.
Lower line speeds could allow lighter, cheaper construction too. I don't think there's anything above 90mph, certainly no 125mph.
I think there is a short section with a 100mph HST differential linespeed (Pyle station is on that section I believe). If I recall correctly, that is the only section of track in south Wales above 90mph. However, the linespeed through Cardiff Central is much less than that and Network Rail's planning application for OHLE through the station showed that they planned to use "extra large" masts I believe.
Most of the route between Cardiff and Bridgend is only 75mph; they could do with improving that and putting in some four-track sections to allow fasts to overtake stoppers before putting wires up between Cardiff and Bridgend.