Never tried it myself, but "functional print Friday" channel on YouTube uses IPA after printing with slow steady prying to remove stuck tpu. The IPA wicks under the print and releases it.

I use a bit of glue and try to remove it when it's still warm but not hot. Zero issues so far.

In my experience, TPU benefits from a higher first layer - it prefers being just "deposited" by the nozzle instead of slightly squished like for PLA and PETG. You can try raising your Z adjustment by 0.02 - 0.05mm and see if that makes it easier to remove.

If you work with transparent TPU and don't want it to turn white, print slow and cold.

I've never had too much trouble removing TPU from a textured PEI plate. If the part doesn't want to come off, lift up a corner of the part with a knife and put a drop of IPA under it.

Turn off the heat on the hotbed. TPU doesn't need it, and it makes removal much easier if you don't heat the hotbed.

Also reduces the amount the printer heats your room by 90% or so.

I've got a textured PEI bed and when I've printed TPU, the adhesion has been perfect, i.e. good enough that the part wasn't going to go anywhere unless I wanted it to, but still easy enough to remove when the print was done and the bed had cooled. I guess it could vary from filament brand to brand, so it's possibly worth trying the same brand as I used, which was cheap Geeetech stuff. It's £8 a roll, and I've used their cheap PLA for ages. I wouldn't recommend their ABS+, though, as it seems to break down at the lowest temperature that gives reasonable layer adhesion.

Smooth prusa powder coated works great.

Drying makes a gigantic difference. It only takes around an hour in the open before TPU changes significantly from any ambient moisture and become visible in the print.

If you can control the moisture to a minor degree, you can alter the mechanical properties significantly. Once you hear popping, you'll likely start blowing holes in prints, but there is a stage before this where the bubbles of gases are present but are not coalescing into the larger audible voids of escaping steam. You will see this on long prints using dried TPU filament left out in the open. There will be a much tougher start to the print that gradually degrades into a slightly softer and more flexible texture. It will likely turn slightly foam-like spongy rubbery soft for a section and then it will start popping and dropping walls with holes in the structure.

If you write down the room temperature and humidity and note the time it takes to get to this moisture property, it becomes possible to alter the flexible properties or empirical hardness of a TPU to make it behave in off label ways. This is essentially creating your own rudimentary foaming or light weight filament. It works best for vase mode or other small single wall structures. I have used this based on intuition alone. I imagine with a bit of record keeping one could control the humidity of a box to do longer prints within this state of foaming softness. I don't know of anyone using a humidifier like the ones for acoustic guitar cases or cigars in a filament box, but that would be an interesting thing to play with too.

I just use my textured sheet and remove very carefully - but I tend to use harder TPUs which makes flexing the sheet more effective.

I have heard people say that TPU is the reason they still keep a glue stick handy and the Prusa material matrix says to use it on their smooth PEI sheet. And it suggests a polypropylene sheet is also applicable without glue stick.

I used to throw the textured pei plate with the sticking print into the freezer for 15 min, made removal much easier. Now I just add baby powder onto my build plate, and remove the print at room temp. Much easier, a little messier