<p>Shape memory alloys (SMAs) have the ability to exhibit moderate recoverable deformations through temperature modulation. This characteristic enables SMAs to function as lightweight and compact thermomechanical actuators. The synthesis of SMA actuators for systems with specific requirements on their actuation path (displacement stroke vs. temperature) currently relies on complex and expensive material processing and characterization. This paper first presents a geometric-based method for the synthesis of novel SMA axial actuators, termed as hybrid SMA actuators, whose size properties and material distribution are modulated to approximate a target actuation path. The hybrid SMA actuators are formed by multiple wire sections connected in series, where the length and cross-sectional area of each section are design variables. The material response of each wire section is also a design variable, where it is selected from a set of commercially available SMAs. Through the combination of multiple SMA wire sections, the hybrid SMA actuators can exhibit actuation paths not achievable by using single monolithic SMA wires. The paper then presents a machine learning-assisted framework for the modeling and design of hybrid SMA actuators. This approach allows for the prediction of their actuation path without the use of structural simulations leveraging numerical implementations of constitutive models, allowing for simplified and computationally efficient modeling and circumventing numerical issues. A regression model consisting of an ensemble of binary decision trees is trained using structural simulations. Training sets of 10000 design samples per material combination for the sections forming the hybrid SMA actuators are created using Latin hypercube sampling. During the training process of the machine learning model, actuation displacement outputs at 0.5 K temperature steps during heating and cooling are mapped to the actuator design inputs. The trained model subsequently enables the prediction of the actuation response of new hybrid SMA actuators without the need of conventional structural simulations. A validation test using 2000 design samples for hybrid SMA actuators with two sections demonstrates R-squared values of 0.99791 and 0.99762 for heating and cooling, respectively. The evaluation time for the validation samples using the trained model is less than 8 minutes, while the evaluation time using structural simulations is 59 minutes. The trained model is used as the design evaluator for the efficient design optimization of hybrid SMA actuators. The optimization approach is demonstrated through the synthesis of various hybrid SMA actuators capable of exhibiting actuation paths applicable to a diverse set of smart systems.</p>

Machine Learning-Assisted Modeling and Design of Hybrid Shape Memory Alloy Axial Actuators

Paper Proceedings: https://asmedigitalcollection.asme.org/SMASIS/SMASIS2021/volume/85499

ASME 2021 Smart Materials, Adaptive Structures and Intelligent Systems (SMASIS2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.