Another cool energy storage technique is using subterranean flywheels. They'd probably require less maintenance than other similar energy storage mechanisms.
Flywheel energy storage has been proposed many times, and built a few times. Flywheel-powered buses have been deployed. Not too successfully.
EMALS uses a big flywheel to power aircraft carrier catapults. The flywheel energy storage seems to cause most of the problems. The next generation of that system may use ultracapacitors.[1]
It seems that the flywheel is a monolithic system, and the proposed next gen ultracapacitor is a modular system.
You probably only have one giant flywheel, as the mass is where the energy storage gains make it worthwhile. Maybe you have many of them in parallel, but the size would make scaling an issue. Not to mention that there’s an associated drivetrain and linkages and so on. It’s a fully integrated contraption. It works, but many parts must work in tandem for it to be so.
This is in contrast to an electronic system, which would benefit from miniaturization and isolated redundancy. Racks of capacitors could run as hot spares when energy is abundant next to racks that may be down for maintenance. I doubt that a flywheel can be meaningfully inspected, let alone repaired or upgraded while operating.
That is the benefit of a modular approach to certain problems and not others. The two approaches are not always mutually exclusive either. Capacitors could be used as regenerative brakes for flywheels, for instance.
UT Austin's cannon-caliber railgun had some impressive pulsed-power output from flywheels for demos (a 150g projectile launched at nearly 2km/s), but they were fired a single-digit numbers of times.
It would seem that flywheels can store a lot of energy, but due to the form of the energy storage (kinetic) it’s difficult to discharge the current quickly, so to speak. Especially compared to something like a capacitor which is designed for rapidly discharging a literal current, as opposed to the analogous metaphorical motive current in a flywheel.
The flywheels GP refers to were homopolar generators [1]. They created an impressive current pulse, and extracted almost all the kinetic energy in less than one revolution.
My Dad, (85 years old), recently posted an idea about using flywheels to store electricity -- it would give him a real kick if he knew people from here were taking a look at it: http://aoi.com.au/devices/KPW/
EMALS uses a big flywheel to power aircraft carrier catapults. The flywheel energy storage seems to cause most of the problems. The next generation of that system may use ultracapacitors.[1]
[1] https://www.ga.com/capacitors/military-mil-spec-capacitors