This is the Matlab code used to analyse the data collected for my master thesis titled:
An attempt was made to investigate how the activation behaviour of a NiTi shape memory alloy is affected by different methods of heating it in water. For this an experimental rig was build, which allowed water to be heated or cooled and to be pumped around a shape memory alloy wire connected to a return spring.
The temperature at two points, one near the wire front and one near its end, was recorded, as well as the length of the wire. To be able to measure the length, a sensor had to be designed and build.
Throughout the experiments some interesting phenomena were observed. During electric heating the one sensor recorded a very high temperature, while the other read that the water was actually cooled. This only happened when turning the heating current on or off. Furthermore while water heating unexpected behaviour of the wire, and coupled with it, the temperature was recorded.
When considering the occurrences as true measurements the following can be deducted: if the wire is suddenly electrically heated, it can sometimes lead to a dangerous heat increase at one end of the wire, while the other end is cooled. Therefore step-like electrical activation should be avoided. Also, the alloy can undergo sudden transformation spurts, during which comparatively large amounts of energy can be absorbed from and released into the environment.
These phenomena could also be the result of random or systematic errors. A number of possible error sources have been discussed as well.