Modeling of Internal Damage Evolution During Actuation Fatigue in Shape Memory Alloys

Shape memory alloys (SMAs) are unique materials capable of undergoing a thermo-mechanically induced, reversible, crystallographic phase transformation. As SMAs are utilized across a variety of applications, it is necessary to understand the internal changes that occur throughout the lifetime of SMA components. One of the key limitations to the lifetime of a SMA component is the response of SMAs to fatigue. SMAs are subject to two kinds of fatigue, namely structural fatigue due to cyclic mechanical loading which is similar to high cycle fatigue, and functional fatigue due to cyclic phase transformation which typical is limited to the low cycle fatigue regime. In cases where functional fatigue is due to thermally induced phase transformation in contrast to being mechanically induced, this form of fatigue can be further defined as actuation fatigue. Utilizing X-ray computed microtomography, it is shown that during actuation fatigue, internal damage such as cracks or voids, evolves in a non-linear manner. A function is generated to capture this non-linear internal damage evolution and introduced into a SMA constitutive model. Finally, it is shown how the modified SMA constitutive model responds and the ability of the model to predict actuation fatigue lifetime is demonstrated.

Manufacture of Lenses and Diffraction Gratings Using DLP As an Additive Manufacturing Technology

This document condenses the results obtained when 3D printing lenses and their potential use as diffraction gratings using Digital Light Processing (DLP), as an additive manufacturing technique. This project investigated the feasibility of using DLP additive manufacturing for producing custom designed lenses and gratings. DLP was identified as the preferred manufacturing technology for gratings fabrication. Diffraction gratings take advantage of the anisotropy, inherent in additive manufacturing processes, to produce a collated pattern of multiple fringes on a substrate with completely smooth surfaces. The gratings are transmissive and were manufactured with slit separations of 10, 25 and 50 μm. More than 50 samples were printed at various build angles and mechanically treated for maximum optical transparency. The variables of the irradiance equation were obtained from photographs taken with an optical microscope. These values were used to estimate theoretical irradiance patterns of a diffraction grating and compared against the experimental 3-D printed grating. The resulting patterns were found to be remarkably similar in amplitude and distance between peaks when compared to theoretical values.

Self-Powered Multifunctional Instrumented Knee Implant

Computational modeling, instrumented linkages, optical technologies, MRI, and radiographic techniques have been widely used to study knee motion after total knee replacement (TKR) surgery. Information provided by these methods has helped designers to develop implants with better clinical performance and surgeons to obtain an improved understanding of the stability and mobility of the joint. Correspondingly, overall patient satisfaction with respect to the reduction in pain and recovery of normal functioning of the joint has been improving. However, about 20% of patients are still not fully satisfied with their surgical outcomes. The main obstacle in the current state-of-the-art is that a comprehensive post-operative understanding of knee balance is still unavailable, mostly due to a lack of in vivo data collected from the joint after surgery. This work presents an attempt to develop a self-powered instrumented knee implant for in vivo data acquisition. The knee sensory system in this study utilizes several embedded piezoelectric transducers in the tibial bearing of the knee replacement in order to provide sensing and energy harvesting capabilities. Through a series of analytical modeling, finite element simulation, and experimental testing, the performance of the suggested system is evaluated and a dimensionally optimized design of an instrumented TKR is achieved. More specifically, a comprehensive platform is established in order to combine the knowledge of embedded piezoelectric sensors and energy harvesters, musculoskeletal modeling of the knee joint, multiphysics finite element modeling, additive manufacturing techniques, image processing, and experimental knee loading simulation in order to achieve the experimentally validated and optimized instrumented knee implant design. The cumulative work presented in this article encompasses three main studies performed on the sensing performance of the proposed design: first, preliminary parametric studies of the effect of local dimensional and material parameters on the electromechanical behavior of the embedded sensory system; second, investigation of the ability to sense total force and center of pressure location; and third, evaluation of an enhanced system with the ability to sense compartmental forces and contact locations. Additionally, the energy harvesting capacity of the system is investigated to ensure the achievement of a fully self-powered sensory system. Results obtained from the experimental analysis of the system demonstrate the successful sensing and energy harvesting performance of the designs achieved in this study.

Experimental and Theoretical Study of a Dual-Beam Piezoelectric Energy Harvester With Misaligned Magnets

In this study, a dual-beam piezoelectric energy harvester is proposed. This harvester consists of a main beam and an auxiliary beam with a pair of magnets attached to couple their motions. The potential energy of the system is modeled to understand the influence of the potential wells on the dynamics of the harvester. It is noted that the alignment of the magnets significantly influences the potential wells. A theoretical model of the harvester is developed based on the Euler-Bernoulli beam theory. Frequency sweeps are conducted experimentally and numerically to study the dynamics of the harvester. It is shown that the dual-beam harvester can exhibit hardening effect with different configurations of magnet alignments in frequency sweeps. The performance of the harvester can be improved with proper placement of the magnets.

Dangerous Scenes Recognition During Hoisting Based on Faster Region-Based Convolutional Neural Network

In the last couple of years, advancements in the deep learning, especially in convolutional neural networks, proved to be a boon for the image classification and recognition tasks. One of the important practical applications of object detection and image classification can be for security enhancement. If dangerous objects or scenes can be identified automatically, then a lot of accidents can be prevented. For this purpose, in this paper we made use of state-of-the-art implementation of Faster Region-based Convolutional Neural Network (Faster R-CNN) based on the monitoring video of hoisting sites to train a model to detect the dangerous object and the worker. By extracting the locations of them, object-human interactions during hoisting, mainly for changes in their spatial location relationship, can be understood whereby estimating whether the scene is safe or dangerous. Experimental results showed that the pre-trained model achieved good performance with a high mean average precision of 97.66% on object detection and the proposed method fulfilled the goal of dangerous scenes recognition perfectly.

Image-Based Comprehensive Maintenance and Inspection Method for Bridges Using Deep Learning

Bridge management and maintenance work is an important part for the assessment the health state of bridge. The conventional management and maintenance work mainly relied on experienced engineering staffs by visual inspection and filling in survey forms. However, the human-based visual inspection is a difficult and time-consuming task and its detection results significantly rely on subjective judgement of human inspectors. To address the drawbacks of human-based visual inspection method, this paper proposes an image-based comprehensive maintenance and inspection method for bridges using deep learning. To classify the types of bridges, a convolutional neural network (CNN) classifier established by fine-turning the AlexNet is trained, validated and tested using 3832 images with three types of bridges (arch, suspension and cable-stayed bridge). For the recognition of bridge components (tower and deck of bridges), a Faster Region-based Convolutional Neural Network (Faster R-CNN) based on modified ZF-net is trained, validated and tested by utilizing 600 bridge images. To implement the strategy of a sliding window technique for the crack detection, another CNN from fine-turning the GoogLeNet is trained, validated and tested by employing a databank with cropping 1455 raw concrete images into 60000 intact and cracked images. The performance of the trained CNNs and Faster R-CNN is tested on some new images which are not used for training and validation processes. The test results substantiate the proposed method can indeed recognize the types and components and detect cracks for a bridges.

Dynamics of Vibration Energy Harvester Governed by Gravity and Magnetic Force in a Rotating Wind Turbine Blade

This paper presents mathematical modeling of an energy harvester (EH) for a wireless structure health monitoring (SHM) system in wind turbine blades. The harvester consists of a piezoelectric energy harvester (PEH) beam, a gravity-induced disk, and magnets attached to both the beam and the disk. An electromechanical model of the proposed EH is developed using the energy method with repelling magnetic force considered. The three coupled equations — the motion of the disk, the vibration of the beam, and the voltage output — are derived and solved using ODE45 in MATLAB software. The result showed the blade rotation speed affects the output angular velocity of disk and the output PEH voltage. That is, as the blade speed increases, the disk angular velocity becomes nonlinear and chaotic which is more beneficial to generate larger power.

Validation of Impedance-Based Structural Health Monitoring in a Simulated Biomedical Implant System

Total Knee Replacement (TKR) is an important and in-demand procedure for the aging population of the United States. In recent decades, the number of TKR procedures performed has shown an increase. This pattern is expected to continue in the coming decades. Despite medical advances in orthopedic surgery, a high number of patients, approximately 20%, are dissatisfied with their procedure outcomes. Common causes that are suggested for this dissatisfaction include loosening of the implant components as well as infection. To eliminate loosening as a cause, it is necessary to determine the state of the implant both intra- and post-operatively. Previous research has focused on passively sensing the compartmental loads between the femoral and tibial components. Common methods include using strain gauges or even piezoelectric transducers to measure force. An alternative to this is to perform real-time structural health monitoring (SHM) of the implant to determine changes in the state of the system. A commonly investigated method of SHM, referred to as the electromechanical impedance (EMI) method, involves using the coupled electromechanical properties of piezoelectric transducers to measure the host structure’s condition. The EMI method has already shown promise in aerospace and infrastructure applications, but has seen limited testing for use in the biomechanical field. This work is intended to validate the EMI method for use in detecting damage in cemented bone-implant interfaces, with TKR being used as a case study to specify certain experimental parameters. An experimental setup which represents the various material layers found in a bone-implant interface is created with various damage conditions to determine the ability for a piezoelectric sensor to detect and quantify the change in material state. The objective of this work is to provide validation as well as a foundation on which additional work in SHM of orthopedic implants and structures can be performed.

Broadband Rotational Energy Harvesting Using Micro Energy Harvester

This paper proposes a broadband rotational energy harvesting setup by using micro piezoelectric energy harvester (PEH). When driven in different rotating speed, the PEH can output relatively high power which exhibits the phenomenon of frequency up-conversion transforming the low frequency of rotation into the high frequency of resonant vibration. It aims to power self-powered devices used in the applications, like smart tires, smart bearings, and health monitoring sensors on rotational machines. Through the excitation of the rotary magnetic repulsion, the cantilever beam presents periodically damped oscillation. Under the rotational excitation, the maximum output voltage and power of PEH with optimal impedance is 28.2 Vpp and 663 μW, respectively. The output performance of the same energy harvester driven in ordinary vibrational based excitation is compared with rotational oscillation under open circuit condition. The maximum output voltage under 2.5g acceleration level of vibration is 27.54 Vpp while the peak output voltage of 36.5 Vpp in rotational excitation (in 265 rpm).

Modular Origami Robot Inspired by a Scorpion Tail

Arthropod animals like scorpions with modular body parts can be an inspiration for a robot’s structure. The design presented here relays on inter-connected origami towers, but could also be easily disassembled. Each origami tower is fully autonomous and at the same time is part of the robot as a whole. The towers are positioned between two platforms that enable modularity. The scorpion’s tale shape is achieved by the varying platform diameter resulting in cone-like form. Each tower is actuated independently to enable multiple degrees of freedom. Maneuvering with separated units, assists in easier reparation as well as replacement. Detaching the towers into separate parts makes this structure develop more precise movements, since every unit will move autonomously. Therefore, having a higher number of separated movements combined leads to a smooth bionic movement. So, the overall hierarchy will be modular contributing to a greater curvature bending of the whole structure. Actuating and maneuvering the robot in the main concept is done by separated electro motors, built in the platform. The basic structure will be built from thick paper with plastic coatings. The thick paper itself is lightweight, but at the same time flexible. To protect the paper towers, double plastic foil is placed as an outer coating which acts as an origami cover. This transparent layer is elastic hence it can follow and support the individual units’ movements. This work is focused on understanding origami towers kinematics and different combinations of inter-connected towers to achieve multiple degrees of freedom. A conceptual model is developed, supported by CAD and mathematical models. At the end a prototype is presented.