Hook up a generator in the drivetrain. (What else do you possibly think could be done?)
Since the operating speed and power will not be consistent or repeatable, doing anything useful with that power will be a challenge. You can use an inverter to convert DC to line voltage and frequency AC. I can assure you without doing any calculations whatsoever that it will not be cost-effective.
You've asked similar questions before, and I have a question in return. WHY are you asking these questions? A forum on the internet is not capable of making something out of nothing.
Probably the most cost-effective approach would be to connect the stationary bikes to a fan (mechanically or electrically) blowing on the rider. That would possibly let you turn the AC back a notch or two.
Big fluid (water?) tank up on the roof (or water tower like in the movies).
Another big fluid tank as low down as you can get it (i.e. the basement if possible).
Each exercise device is connected to a pump that pulls fluid from the lower tank and pumps it up to the higher tank. (For steppers a linear pump, for bikes a rotary pump...)
The Upper tank in turn drains down to a generator that outlets into the lower tank.
You now have a built in accumulator which will smooth variations in power and also if large enough allow you to store energy to be used when needed.
Or, you just use the water pumped to the roof tank to ensure a nice high pressure shower after all that exercise - heck maybe leave some devices generating electricity and use that for an immersion heater on the tank.
Exercises in engineering and musings of Rube Goldburg-esque inventions aside, unless this is for purely novelty or demonstration purposes, or truly a subsistence means of acquiring some juice to charge a cell phone or other device, bicycle generation, while not completely useless, is pretty close to it. (imho)
It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.
I was a judge at a STEM event for school children. One of the kids proposed a mat that would generate power as someone walked across it. I could design a cinderblock that someone could lift up and down while standing on the mat. You could arrange a bunch of these mats into a circle, and ride a bicycle around the circle, again, generating power.
"Big fluid (water?) tank up on the roof (or water tower like in the movies). Another big fluid tank as low down as you can get it (i.e. the basement if possible)."
At such a small scale, a Lithium battery + inverters would be much more cost effective. Or forget the battery and push the energy straight into the grid as already suggested.
WADMW, hopefully you can tell the difference between both.
It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
Somehow, this discussion came up in the recent past and again take a look at a bicycle dynamo. If you have ever used one, you will feel the difference.
Pumped water is exactly what is used at full size power plants for load leveling/peak demand handling. No reason it is inappropriate here. In the larger scale it is how hydroelectric power is gotten using dams and turbines. It only seems odd to some because, really, people are such poor sources of power.
People are such poor sources of power that it does make the question bizarre. Might also ask how hard is it to stir water with a teaspoon in a teacup to boil the water. Followup - since reducing pressure makes liquids boil more readily, will it be easier to heat the water at higher altitude.
I would point out that stationary bikes and elliptical trainers are expensive, so putting them on Craig's list and taking the money to buy batteries or an extension cord would be more reliable.
I don't think we can really address the appropriateness of design suggestions because the OP doesn't really spell out the conditions.
Are we talking about a home gym in a western civilization, a corporate chain like Anytime Fitness with 100s of locations, or is this a very remote location with little to no infrastructure, but can cobble together some simple machines... that last case would throw out a lot of complaints about being impractical.
I'm curious if the OP is trying to make a 'green gym' in San Francisco, or trying to keep the lights on in the Outback.
