You Should Know

Let's assume we need 100MWth of cooling capacity. At a CoP of 5 (generous case) we need 20MWe of power to move 100MWth (120MWth total with our electrical input converted into heat eventually as well)

Comparing that to absorber chillers: A single effect absorber at a CoP of 0.4 needs 250 MWth of heat energy at or above 90⁰C. If you go dual effect you can get 0.9-1.2, but need higher grade heat. I'll do the worst case for now.

For reference, 250MWth is roughly what goes up the stack in a 125 MWe plant (hilariously tiny plant. Might run a remote community and be a portable turbine generator or two on a trailer? I've worked on gas compressors with higher nameplate capacities than that by miles)

A 1GWe coal plant at 33% efficiency burns 3GWth of heat energy and discharges 2GWth up their stack. That's 2 GWth you have available as waste heat. By definition it has to be vented to atmosphere or to water at the condenser. That's a lot of already-paid-for heat being literally thrown out.

If you count cooling efficiency using just the thermal watts you have to burn at the powerplant for the electric cooler to move 100MW of heat:

20MWe / .33 = 60 MWth needed to generate it.

If your electric chiller is a more reasonable CoP of 4:

100MWth heat moved, takes 25MWe.

25MWe at 33% efficiency, is 75.75 MWth of heat energy burned at the power plant to make 25MW of electricity.

If you used a dual effect absorber at 1.2 CoP:

100MWth cooling with 83MWth of heat energy at 170⁰C.

What these two numbers show is we can move 100 MW of thermal energy with an electric chiller by burning 60MW (or 75.8MW) of heat we then turn into electricity from a power plant, or 83MW of heat that already was burned, if we steal heat from the flue gasses and use an absorber chiller. The raw energy input to do 100MW of cooling differs by only 23 MW, so an absorber is in the same order ballpark still.

What's important to note here is the energy needed to run an electric cooler is $money$ while the energy to run an absorber is waste heat going up the stack... These are used on earth already BTW just typically as a combined heat and power plant for universities or government/company campus. There's no reason you can't also apply that to an entire city.

For example, Aarhus University Hospital, Denmark uses a 3 MW absorption chiller powered by 105 °C district heating water. It consumes about 4.3 MW of heat to produce 3 MW of cooling, while simultaneously returning useful heat to the district heating network. It's operated alongside electric chillers so the most economical option can be chosen depending on electricity prices

Crucially you don't need to oversize your power grid for a summer heatwave peak, when your power lines are hottest (highest resistance, expand, droop lower and maybe short out on trees) so you can instead use literal waste to cool a city instead. Cooling tends to be ⅓ the amount needed for heating roughly, so you a home that needs 12kW of heating roughly needs 4kW of cooling.

2GWth of cooling capacity at 4kW per home is 500 000 homes cooled with waste heat. Of course this is assuming perfect efficiency, but it illustrates my point. District cooling typically uses 1-3% of the cooling nameplate in electrical demand for pumps afaik, so a 2GWth cooling system needs 60 MWe of pumping. To achieve the same cooling with electric chillers you need 571 MWe (assumed 3.5 CoP) just for chilling, consuming more than half of the power output of our 1GWe plant.