I am currently working on a huge industrial process for a student project and we need to transfer around 25MW of energy between dozens and dozens of streams, some larger and some smaller ones. Many even with gas/gas exchange, and those tend to get quite large fast. I was wondering if there was a general rule of thumb to determine if the investment for a heat exchanger is worth it in order to save on operating costs. Something like, "don't build liquid/liquid exchangers for less than 250kW and gas/gas for less than 500kW" something along those lines. Any ideas?
If this is a student design problem to see whether you understand the principles, make it as easy as you can, use steam and cooling water. Calculating all the pressure drops on the weird pipe systems and valves you need when using one stream to cool/heat the others will drive you insane. Not to mention what it can do to your choice of materials of construction an therefore building cost.
On the other hand, consider the place and resources available - e.g. if the area has low water resources you might find your design doesn't make it if your water use is too high. Or you may not have a large enough steady supply of coal/oil/gas/electricity to heat/cool all that water.
If you are designing something that will actually get built, take into acount the maintenance needs. Some designs look wonderfully efficient until someone has to hang upside-down or pull out half a mile of pipe to find a leak or change a valve. Don't get yourself lynched by the maintenance crew for making everyone's life difficult.