PSI - Issue 24

Riccardo Panciroli et al. / Procedia Structural Integrity 24 (2019) 593–600 R. Panciroli and F. Nerilli / Structural Integrity Procedia 00 (2019) 000–000

594

2

Composite materials are receiving particular attention when referring to bistable structures, as these o ff er a higher strength to sti ff ness ratio compared to metals (Zhang et al., 2019). It is thus easier to assemble a compliant bistable structure utilizing composite materials rather than with metals. One of the most common approaches to realize structures presenting multiple stable configurations is to exploit the thermal deformations of non-symmetric layered structures (Pirrera et al., 2010; Lamacchia et al., 2015; Zhang et al., 2016). Each stable configuration corresponds to a di ff erent shape and these structure can thus be utilized for shape morphing purposes. Indeed, manufacturing and material properties uncertainties might largely a ff ect the behavior of the developed structure as shown by Brampton et al. (2013). Other authors proposed to develop multistable structures utilizing cylindrical shells (Kebadze et al., 2004; Guest and Pellegrino, 2006). To date, the majority of the studies proposed to utilize piezoelectrics as driving force to actuate the morphing process (Betts et al., 2013b, 2012). Some other works instead proposed to utilize SMA wires, strips, or springs for the actuation (Zhang et al., 2019; Marfia and Sacco, 2005). However, so far, SMAs have been utilized as external components rather than being integrated in a single structure. The attempts to integrate SMA wires into a composite structure focused on the opportunities to utilize superelastic alloys to attenuate the vibrations rather than utilize shape memory alloys to perform actuation (John and Hariri, 2008). Primary objective of this work is to develop and verify a numerical scheme to predict the behavior of smart structures comprised by a bistable composite structure with integrated SMA wires to perform autonomous shape morphing. We choose to utilize Nitinol, a nearly equiatomic nickel-titanium alloy as shape memory material. Nitinol can show either superelastic or shape memory characteristics depending on its chemical composition and thermal treatments during the production stage. Superelastic alloys can be utilized to increase the damping of a structure, while shape memory alloys can serve for actuation. We therefore concentrate on shape memory alloys only. The shape memory e ff ect relies on phase transformations between Martensite ( M ) and Austenite ( A ). Briefly, the SMA can accumulate a large pseudo-plastic deformation (in the order of 5%) when subjected to a stress level in the order of 200MPa. Such deformation can be fully recovered upon a thermal cycle, i.e by heating the material above the austenite finish ( A f ) temperature, thus morphing back to its original shape. The material returns martensitic after cooling. In the case of one-way shape memory e ff ect, the material does not deform during the A to M cycle. However, some alloys might show a two-way shape memory e ff ect, that is, they deform upon both M to A , and A to M transformations. We here concentrate on one-way shape memory e ff ect only, and we will follow the approach proposed by Mizzi et al. (2019), who recently presented an analytical model to predict the cycling actuating capabilities of SMA wires embedded in a compliant matrix with the objective of maximizing the actuating capabilities and the recovery potential. This work is organized as follows: we initially detail the numerical modeling of SMA wires and we present the validation of the model through comparison against original experimental stress-strain curves. We then formulate a numerical model relative to the sole antisymmetric composite panel and we highlight the e ff ects of plate dimension and curing cycles on the bistable configurations, tuning the model utilizing experimental data. Finally, we introduce a model for the smart composite structure comprised by a bistable layered shell and SMA wires laying to its surface to perform the actuation. The underlying idea of the present work is to evaluate the feasibility to actuate the bistable plate by exploiting the shape memory capabilities of SMA wires. Upon preliminary considerations on the actuation capabilities of SMA alloys basing on the material properties available from the literature, we opted to utilize a commercial nitiol wire with 0.41mm diameter and nominal A f of 85 ◦ C distributed by Luminous Piertech . The stress-strain response of the wire at ambient temperature has been evaluated through an elongation test executed on a MTS Acumen3 electrodynamic test system. The tests have been executed on wires with gauge length of 83mm at a strain rate of 10 − 3 . Several wires have been tested to assess the repeatability of the results. The stress-strain curve reported in Figure 1 shows the typical response of SMA wires. The material initially behaves linearly until the pseudo-yield stress level, which in our case is approximately 150MPa, to later show a plateau until 6% elongation. Such plateau identifies the strain recovery capability of the SMA alloy. The pseudo-plastic deformation accumulated within this range can be fully recovered upon a thermal cycle by heating the wire above A f . The wire 2. Shape memory alloy wire