Using Multiscale Modeling to Predict Material Response of Polymeric Materials

Presented is a multiscale modeling method applied to light activated shape memory polymers (LASMP). LASMP are a new class of shape memory polymer (SMP) being developed for applications where a thermal stimulus is undesired. Rotational Isomeric State (RIS) theory is used to build a molecular scale model of the polymer chain yielding a list of distances between the predicted cross-link locations, or r-values. The r-values are then fit with Johnson probability density functions and used with Boltzmann statistical mechanics to predict stress as a function of strain of the phantom network. Junction constraint theory is then used to calculate the stress contribution due to interactions with neighboring chains, resulting in previously unattainable numerically accurate Young’s modulus predictions based on the molecular formula of the polymer. The system is modular in nature and thus lends itself well to being adapted for specific applications. The results of the model are presented with experimental data for confirmation of correctness along with discussion of the potential of the model to be used to computationally adjust the chemical composition of LASMP to achieve specified material characteristics, greatly reducing the time and resources required for formula development.

Effect of Segmented Electrodes on Piezo-Elastic and Piezo-Aero-Elastic Responses of Generator Plates

In this paper, the use of segmented electrodes is investigated to avoid cancellation of the electrical outputs of the torsional modes in energy harvesting from piezo-elastic and piezo-aero-elastic systems. The piezo-elastic behavior of a cantilevered plate with an asymmetric tip mass under base excitation is investigated using an electromechanically coupled finite element (FE) model. Electromechanical frequency response functions (FRFs) are obtained using the coupled FE model both for the continuous and segmented electrodes configurations. When segmented electrodes are considered torsional modes also become significant in the resulting electrical FRFs, improving broadband (or varying-frequency excitation) performance of the generator plate. The FE model is also combined with an unsteady aerodynamic model to obtain the piezo-aero-elastic model. The use of segmented electrodes to improve the electrical power generation from aeroelastic vibrations of plate-like wings is investigated. Although the main goal here is to obtain the maximum electrical power output for each airflow speed (both for the continuous and segmented electrode cases), piezoelectric shunt damping effect on the aeroelastic response of the generator wing is also investigated.

Modeling Smart Materials Using Hysteretic Recurrent Neural Networks

This paper describes a novel approach to modeling hysteresis using a Hysteretic Recurrent Neural Network (HRNN). The HRNN utilizes weighted recurrent neurons, each composed of conjoined sigmoid activation functions to capture the directional dependencies typical of hysteretic smart materials (piezoelectrics, ferromagnetic, shape memory alloys, etc.) Network weights are included on the output layer to facilitate training and provide statistical model information such as phase fraction probabilities. This paper demonstrates HRNN-based modeling of two- and three-phase transformations in hysteretic materials (shape memory alloys) with experimental validation. A two-phase network is constructed to model the displacement characteristics of a shape memory alloy (SMA) wire under constant stress. To capture the more general thermo-mechanical behavior of SMAs, a three-phase HRNN model (which accounts for detwinned Martensite, twinned Martensite, and Austensite phases) is developed and experimentally validated. The HRNN modeling approach described in this paper readily lends itself to other hysteretic materials and may be used for developing real-time control algorithms.

Experimental Implementation of Negative Impedance Shunts on Periodic Structures

Shunt damping of structures has been heavily researched, both passively and actively. Negative capacitance shunts actively control vibration on a structure and have been shown to obtain significant broadband suppression. The use of smaller piezoelectric patches, implemented in a periodic array, can alter the behavior of the control. Assorted shunt arrangements as well as circuit configurations will be investigated. Experimental results will be compared to theoretical predictions of shunt performance.

Finite Element Martensite Ratio Derivation on NiTi via Measurable Criteria of Strain Rate

The shape memory alloys (SMAs) and smart composites have a large use in high and low level industry, while a lot of research is being done in this field. The existence of smart composite structures is because of the advance mechanical benefits of the above materials. This work refers to dynamic and quasi static nonlinear explanation of these materials. After mathematical model consideration on the rate of strain, a model which is about martensite ratio of NiTi has been presented. This work has been done because of the high sensitivity of these materials to strain rate and use of visual and measurable engineering criteria to access other variables. As the martensite ratio is not engineering measurable amount, it needs to have macro scale property to measure this important nano scale criteria. Relative experiments are done to show the rate dependency of NiTi.

Thermo-Electro-Mechanical Shakedown Response of Conditioned Shape Memory Alloy Wires

The shakedown response of conditioned shape memory alloy wires (Flexinol®) is examined experimentally during constant tension thermal cycles, at several load levels. Strain, temperature, and electrical resistivity are measured simultaneously using a specialized experimental setup that enables a relatively rapid temperature rate (1 °C/s) while preserving the temperature uniformity along the gauge length to less than 1.5 °C. Both elongation and electrical resistance are measured from the same local gauge length, allowing strain-corrected electrical resistivity to be inferred. The most repeatable behavior (least shakedown) occurs at the intermediate load of 191 MPa (consistent with the supplier’s maximum stress recommendation), with a small amount of shakedown (and some loss of two-way shape memory) at lower loads and progressively larger shakedown (strain ratcheting and reduction in hysteresis) at higher loads.

Optimal Control of Vertically Stroking Crew Seats Employing Magnetorheological Energy Absorbers

Nondimensional analyses of vertical stroking crew seats with adaptive nonlinear magnetorheological energy absorbers (MREA) and magnetorheological shock isolation (MRSI) were addressed in this study. Under consideration were single-degree-of-freedom vertically stroking seat systems consisting of a rigid occupant mass falling with prescribed initial impact velocity (sink rate). The governing equations of the vertical stroking crew seats were derived using nondimensional variables such as nondimensional stroke, velocity, acceleration and time constant, as well as nondimensional Bingham number (i.e., the ratio of MR yield force to viscous force). The critical Bingham number was defined as that Bingham number for which the available stroke was fully utilized and the seat reaches zero velocity at the end of stroke. This was done in order to maximize shock mitigation performance. Two cases were studied: (1) the MREA problem, or the case where no spring was employed in the suspension, so that the seat was used for a single shock event, (2) the MRSI problem, or the case where a spring was employed in the suspension, so that after the initial shock event, the suspension could be used for either vibration isolation or mitigation of subsequent shock events. Nondimensional displacement, velocity and acceleration were analyzed for MREA and MRSI vertical stroking crew seats for three different payload masses of 47, 77 and 97 kg corresponding to 5th percentile (%tile) female, 50th %tile and 95th %tile male, respectively, with initial impact velocities of 4, 5 and 6 m/s. An optimal control solution was derived for both the MREA and MRSI cases. The effects of payload mass and initial impact velocity on the optimal responses of the vertical stroking crew seats were analyzed for a feasible range of Bingham number based on a realistically constrained (in diameter and volume) MR damper design.

The Application of Reconfigurable Logic to Active Magnetic Bearing Controller

The article presents the main problems of implementing the PID control law in the reconfigurable logic, namely FPGA integrated circuit. The consecutive steps of discretizing and choosing the fixed-point representation of the continuous, floating-point PID algorithm are described. The FPGA controller is going to be used in the active hetero-polar magnetic bearings system consisting of two radial and one axial bearings. The results of the experimental investigations of the controller are presented. The dynamic performance of the controller is better when compared with the dSPACE controller, that was used so far. The designed hardware and software, the developed implementation procedure and the experience acquired during this stage of the whole project are going to be used during the implementation of more sophisticated control laws (e.g. H∞ robust) in the FPGA for AMB controllers.

Topology Optimization of Mechanical Structures Using Level Set Method

This article introduces a novel algorithm in topology optimization of mechanical structures to achieve a desired compliance using the level set method. In contrast with the literature in this area that attempts to minimize the compliance of a structure, the present study concerns with an innovative formulation to reach a desired compliance. It is shown that a more comprehensive technique is required to achieve this goal.

The Calculation of Stiffness for Semiconductor Components

In this study, the correlation between macroscopic and microscopic properties of the II-IV semiconductor compounds CdX (X = S, Se, Te) is investigated. Based on constructing orthonormal tensor basis elements using the form-invariant expressions, the elastic stiffness for cubic system materials is decomposed into two parts; isotropic (two terms) and anisotropic parts. A new scale for measuring the overall elastic stiffness of these compounds is introduced and its correlation with the calculated bulk modulus and lattice constants is analyzed. The overall elastic stiffness is calculated and found to be directly proportional to bulk modulus and inversely proportional to lattice constants. A scale quantitative comparison of the contribution of the anisotropy to the elastic stiffness and to measure the anisotropy degree in an anisotropic material is proposed using the Norm Ratio Criteria (NRC). It is found that CdS is the nearest to isotropy (or least anisotropic) while CdTe is the least near to isotropy (or nearest to anisotropic) among these compounds. The norm and norm ratios are found to be very useful for selecting suitable materials for electro-optic devices, transducers, modulators, acousto-optic devices.