Magnetic shape memory (MSM) alloys are a new class of actuator materials with high actuation frequency, energy density and strain. MSM effect occurs in alloys, which exhibit a martensite transformation and are ferromagnetic. It involves, under effect of magnetic field, a high strain achieved via reorientation of twinned martensite plates.
The presence of composition variations in the boules prepared by using the Bridgman technique is a major issue because of its effect on the transformation temperature and MSM effect is only possible in the martensite region. The major problem is that even a slight change in the alloy's composition causes drastic changes in the martensite transformation temperature. For boules with composition variations, both transformed and untransformed regions will exist over some temperature range, degrading the performance of any actuator made from them.
Therefore it is crucial to be able to predict the martensite transformation temperature of any NiMnGa alloy. Artificial Neural Networks (ANN) with their learning and generalization ability may act as a suitable tool to predict the martensite transformation temperatures of NiMnGa alloys.
ANN learn the relationships between cause and effect by using a model-free approach. They are generally used when the problem cannot be explicitly described by an algorithm, a set of equations, or a set of rules and ANN approach has recently been applied in prediction due to its adequate performance in pattern recognition.
In this paper, for martensite transformation temperature prediction, the performance of a multi-layer perceptron has been studied. Next, a radial basis function has been employed for prediction. Training and validation stages of the approach are performed by using data sets from four separate analysis results and our chemical analysis results were used for testing.
Computational results show a possible relationship between alloy’s composition and martensite transformation temperature.
[ABSTRACT TRUNCATED TO 2000 LETTERS]