In case of Nitinol you just need to fix it in some kind of mould and anneal it. For thin Nitinol wire annealing temperature is ~500oC with soaking time of ~10 minutes. It will "forgot" previous shape and new shape will be "learnt".
The sample is thin Nitinol wire. My treatment process is also like that. But I recognize something. By the second re-treatment for the same "learnt", the performance of SMA is decreasing.
Nitinol alloys exhibit two closely related and unique properties: shape memory effect (SME) and superelasticity (SE) (also called pseudoelasticity (PE)). Shape memory is the ability of nitinol to undergo deformation at one temperature, then recover its original, undeformed shape upon heating above its "transformation temperature". Superelasticity occurs at a narrow temperature range just above its transformation temperature; in this case, no heating is necessary to cause the undeformed shape to recover, and the material exhibits enormous elasticity, some 10-30 times that of ordinary metal.
For NiTi alloys, you can make them forget its memory by applying heat but do not get to +500ºC. Try only 200ºC-250ºC for a while. In my experience, I have a Af=+90ºC Nitinol wire and it begins to forget its memory above these values but again, don't reach above 500ºC because you will cause the alloy to remember a new shape. Also, I did this without fixing the wire, just applying heat.
Well, I'm not really sure. you can make some tests. Try first for only 1 or 2 minutes. You can always increase time.
I would try with a singular shape. You can take a piece of wire and fix it in a S shape. Then apply above 500ºC to set the S shape. After that you can do the first test to check the alloy doesn't recover.
I put the sample which has own shape (tend to bend 90 degrees) in to the furnace with the temp. of 200 C. I waited as like 5 minutes. But it has still memory.
If you like that the material forget completely previous shape memory effects, you must bring the material to the high temperature phase diagram region, let say 900ºC or even better 1000ºC, then perform a solid solution thermal treatment durin 3.6 Ks (1 hour) and after that kench the alloy to room temperature. This thermal treatment must be done in vacuum or in innert atmosphere, to avoid the oxidation of the material.
During this thermal treatment, the material will programe the shape of the high temperature phase (austenite). So if you like a particular shape, it must be imposed by clamping your wire during the thermal treatment.
After that you may do further aging treatments to control the microstructure of precipitates and so on.
This is in fact the initial standard thermal treatment when starting any training of a Ti-Ni SMA. I hope this could be useful.