Review of Neural Network Modeling of Shape Memory Alloys

Abstract

Shape memory materials are smart materials that stand out because of several remarkable properties, including their shape memory effect. Shape memory alloys (SMAs) are largely used members of this family and have been innovatively employed in various fields, such as sensors, actuators, robotics, aerospace, civil engineering, and medicine. Many conventional, unconventional, experimental, and numerical methods have been used to study the properties of SMAs, their models, and their different applications. These materials exhibit nonlinear behavior. This fact complicates the use of traditional methods, such as the finite element method, and increases the computing time necessary to adequately model their different possible shapes and usages. Therefore, a promising solution is to develop new methodological approaches based on artificial intelligence (AI) that aims at efficient computation time and accurate results. AI has recently demonstrated some success in efficiently modeling SMA features with machine- and deep-learning methods. Notably, artificial neural networks (ANNs), a subsection of deep learning, have been applied to characterize SMAs. The present review highlights the importance of AI in SMA modeling and introduces the deep connection between ANNs and SMAs in the medical, robotic, engineering, and automation fields. After summarizing the general characteristics of ANNs and SMAs, we analyze various ANN types used for modeling the properties of SMAs according to their shapes, e.g., a wire as an actuator, a wire with a spring bias, wire systems, magnetic and porous materials, bars and rings, and reinforced concrete beams. The description focuses on the techniques used for NN architectures and learning.

Mechanical Properties of NiTi Based Shape Memory Alloy used in Manufacturing of Mars Rover Wheels: Short Review

“…Shape memory alloys (SMAs) belong to this family because of their two remarkable characteristics, namely, shape memory effect (SME) and superelastic effect (SE), and have been innovatively employed in various fields, such as sensors, actuators, robotics, aerospace, civil engineering, and medicine. [22] The shape memory and superelastic effects are originated from the thermoelastic martensitic transformation (MT), which is nothing but the ability of shape memory alloys (SMAs) to undergo reversible martensitic phase transformations on application of heating or cooling within specific temperature ranges. [23].…”

Interface stress transfer model and modulus parameter equivalence method for composite materials embedded with tensile pre-strain shape memory alloy fibers

“…The intrinsic structure model of SMAC considering continuous time variation was studied to reveal the time-varying properties of SMAC for the problems of time variation of SMA composite properties and fatigue [ 35 ]. A neural network-based intrinsic structure relationship for SMAC was investigated in the presence of SMA response stresses [ 36 ]. In addition, a three-dimensional thermodynamic intrinsic structure model applicable to arbitrarily shaped objects was investigated for the variation of composite structure shapes and can be downscaled to achieve an accurate description of structures such as wires, rods, beams, and shells [ 37 ].…”

Modeling Xanthan Gum Foam’s Material Properties Using Machine Learning Methods

“…Analyzing materials and composites uses various methods such as multiple linear regression, support vector machines, artificial neural networks, and least squares methods for different purposes. These techniques have been utilized to predict screw pull-out strengths and flexural modulus values of particleboards [ 17 , 18 ], to classify alloys based on their shapes [ 19 ], to predict sound absorption coefficients of sandwich-structured materials [ 20 ], to detect protozoa in wastewater [ 21 ], to model the strength of lightweight foam concrete [ 22 ], and to determine the processing parameters for drilling glass-laminated aluminum-reinforced epoxy composites [ 23 ]. However, no studies on machine learning related to xanthan-gum-based foam materials exist.…”

“…Within the materials and composites domain, diverse methodologies, including multiple linear regression, support vector machines, artificial neural networks, and least squares methods, have been carried out for various purposes. These techniques have provided successful outcomes in predicting screw pull-out strengths and flexural modulus values of particleboards [ 17 , 18 ], classifying alloys based on their shapes [ 19 ], forecasting sound absorption coefficients of sandwich-structured materials [ 20 ], detecting protozoa in wastewater [ 21 ], modeling the strength of lightweight foam concrete [ 22 ], determining processing parameters for drilling glass-laminated aluminum-reinforced epoxy composites [ 23 ], estimating performance in different applications [ 24 ] and organic photovoltaic design [ 25 ]. Additionally, in the context of predicting material properties and minimizing both material waste and cost losses, machine learning methods provide considerable advantages.…”