Electrical Engineering

Electrical Engineering

 

 

Poster Number: ECE-01

Authors: Yasir Al-Nadawi, Xiaobo Tan, Hassan Khalil

Title:  A Robust Adaptive Conditional Servocompensator Design for Nanopositioner Stage Control

 

Abstract: In this work, we address the design of robust adaptive control for a piezoelectric nanopositioner to track periodic desired trajectories. The nanopositioner is modeled with a linear system proceeded by a hysteresis operator, which in turn is represented by a Modified Prandtl-Ishlinskii (MPI) operator. It is assumed that the system is subjected to periodic exogenous signals, generated by a neutrally stable exosystem. For compensating these periodic disturbances, we propose to use an adaptive conditional servocompensator, which accommodates a sufficient number of harmonics, in conjunction with an approximate MPI inverse operator. The adaptation law, which is equipped with a smooth parameter projection for robustness assurance against matched disturbances, is initiated only, when the trajectory driven by a continuous sliding mode control (SMC), enters a boundary layer around its sliding manifold. To reduce the conservativeness of the continuous SMC law, we analytically derive the hysteresis inversion error, which is found to satisfy certain growth condition on the desired control action. Considering the derived inversion upper bound, the SMC control law is designed to ensure that the trajectory converges to the boundary layer without the need to tune all the controller parameters. Our analysis shows that if the hysteresis inversion error is small enough, and under certain conditions, the closed-loop system admits an asymptotically stable, periodic solution. Experiments conducted on a commercially nanopositioner confirm our theoretical analysis, and demonstrate the efficacy of the proposed controller compared to the case of a nonadaptive continuous SMC.

 

This work was supported in part by This work was supported by National Science Foundation (CMMI 1301243). The first auther is sponsored by the Higher Committee for Education Development in Iraq (HCEDIraq) scholarships program.

 

 

 

Poster Number: ECE-02

Authors: Mehmet Akif Alper, Luan Tran, Daniel Morris

Title:  Automatic Contactless Limb Tremor Measurement System

 

Abstract: Tremor measurement is an important component of quantifying the severity of diseases that cause them, such as Parkinson’s disease, and provides a feedback measure for medications enabling dosage adjustments to appropriately treat the disease. Here we propose a system for automatic tremor detection suitable for home or clinic settings that does not require wearable sensors. Rather our system uses the Kinect 2 for contactless measurements. While the in-built person pose estimation can provide rough limb motions, we show that its accuracy can be too poor to quantify small tremors. Thus we add an additional level of limb tracking that resolves limb pose to a higher precision. Using this we are able to obtain automatic tremor detection and measurement.

 

 

Poster Number: ECE-03

Authors: Mohammed Al-Rubaiai, Xiaobo Tan

Title:  An LED-based Optical Communication System with Active Alignment Control

 

Abstract: While acoustic modems have long been the dominant wireless communication method for underwater applications, they incur high cost and large power, and can only deliver low data rates. Recently, light-emitting diode (LED)-based optical communication has emerged as a promising approach to low-power, high-rate data transfer underwater over short-to-medium distances. However, LED-based communication relies on a close-to-line-of-sight link between the transmitter and the receiver, which presents significant challenges in its underwater robotic applications, where the underlying robots undergo constant motions due to propulsion and/or disturbances. We propose a novel and compact LED-based communication system with active alignment control, which maintains the communication link despite the underlying platform movement. Details on the system design and implementation are provided. The prototype is able to communicate at 115,200 bps over at least 23 m in swimming pool tests. Experiments involving a stationary transmitter and a mobile receiver mounted on a terrestrial robot are conducted to demonstrate the performance of the alignment maintenance system.

 

This work was supported in part by National Science Foundation (IIS 1319602)

Poster Number: ECE-04

Authors: Portia Banerjee, Oleksii Karpenko, Mahmood Haq, Lalita Udpa, Yiming Deng

Title:  Impact-Damage Growth Estimation in Composites using Particle Filter Based Prognostics

 

Abstract: With increasing use of fibre reinforced polymer (FRP) composites in several industries such as aviation, automotive and construction, structural health monitoring (SHM) and prognosis of composites have become an extremely critical task in recent years. Despite outstanding qualities such as light-weight, high specific stiffness and strength, components made of composite materials are often vulnerable to damages caused due to fatigue or external impacts which compromises their performance and hence propels the need for robust health monitoring techniques. Overall accurate health prognostic is critical for condition-based-maintenance (CBM) and for reducing life-cycle costs by taking full advantage of the remaining useful life (RUL) of the equipment. In this paper, delamination growth due to repeated low-velocity impact introduced by drop-weight tests on GFRP composite samples was monitored. An integrated prognostics framework for estimation of damage propagation was proposed which utilizes both physical model based on Paris law and CBM data obtained from optical transmission scans of GFRP specimens. A Bayesian method based on particle filtering was implemented to estimate model parameters using damage-sensitive features extracted from measurements. Moreover, delaminations in GFRP samples have distinguishing characteristics unlike crack growth in metals. As a result, applying piecewise Paris-Paris model to our data reduced error in estimation of damage growth path. Additionally, material and model uncertainties were taken into account during update of model parameters and RUL of the GFRP samples. Results demonstrate feasibility and potential of the proposed approach as a robust SHM technique of GFRP composites.

 

 

 

Poster Number: ECE-05

Authors: Atri Bera, Nga Nguyen, Saleh Almasabi, Joydeep Mitra

Title:  Optimal Power Flow based on Frequency Security Constraint

 

Abstract: The objective of this research is to reduce production cost while satisfying requirement of system frequency stability in addition to other operating constraints. The system frequency is required to be maintained within a safe limit, thus indicating the balance between generation and consumption. Hence, the solution for optimization of power flow should not only give a minimum cost of generation within the operating conditions, but also ensure frequency stability. In order to obtain this solution, the requirement of frequency stability is introduced as a new constraint of the power dispatch problem and is represented by the maximum frequency deviation limit. This new constraint is constructed as a non-linear function of system inertia and the frequency regulation constant, since frequency deviation is highly sensitive to these factors. A genetic algorithm is utilized in implementing the power flow optimization problem. The IEEE 30-bus test system is used to demonstrate the proposed idea.

 

 

 

Poster Number: ECE-06

Authors: Connor Boss, Joonho Lee, Jongeun Choi

Title:  Model Uncertainty and Disturbance Estimation for Quadrotor Control

 

Abstract: This poster proposes a discrete-time, multi-time-scale estimation and control design for quadrotors in the presence of external disturbances and model uncertainties. The control scheme controls all six degrees of freedom, resulting in a strategy that can track trajectories in three-dimensional space, with desired pose. Sample-data Extended High-Gain Observers are used to estimate model uncertainties and external disturbances. Discretized dynamic inversion utilizes those estimates and deals with an uncertain principal inertia matrix. In the plant dynamics, the proposed control forces the rotational dynamics to be faster than the translational dynamics. The proposed estimation and control strategy is verified through numerical simulations and experimental flight. During experimental flight tests, all sensing and computation is performed on board the vehicle.

 

This work was supported in part by National Science Foundation, CAREER award CMMI-0846547

 

 

 

Poster Number: ECE-07

Authors: Jennifer Byford, Premjeet Chahal

Title:  3D Printing Ultra-wideband Hybrid Substrate Integrated Ribbon Waveguides

 

Abstract: A new wave guiding structure and fabrication technique is introduced for high speed, low loss, ultra-wideband interconnects. It is a hybrid between a dielectric ribbon and a substrate integrated waveguide design. In this structure, a high dielectric constant valued core is surrounded by a low dielectric constant valued cladding which in turn is surrounded by a metal layer. Both cylindrical and rectangular waveguide designs are presented. Simulation and measurement results show that ultra-wide band interconnects with low-dispersion can be designed using this hybrid approach. Fabrication of the cladding layer was carried out using 3D plastic printing. Simulated and measured results are discussed as well as fabrication techniques.

 

 

 

Poster Number: ECE-08

Authors: Le Cai, Suoming Zhang, Jinshui Miao, Zhibin Yu, Chuan Wang

Title:  Fully Printed Stretchable Thin-film Transistors and Integrated Logic Circuits

 

Abstract: We present intrinsically stretchable thin-film transistors (TFTs) and integrated logic circuits that are directly printed on elastomeric polydimethylsiloxane (PDMS) substrates. The printed devices utilize carbon nanotubes and a type of hybrid gate dielectric comprising PDMS and barium titanate (BaTiO3) nanoparticles. The BaTiO3/PDMS composite simultaneously provides high dielectric constant, superior stretchability, low leakage, as well as good printability and compatibility with the elastomeric substrate. Both TFTs and logic circuits can be stretched beyond 50% strain along either channel length or channel width directions for thousands of cycles while showing no significant degradation in electrical performance. This work may offer an entry into more sophisticated stretchable electronic systems with monolithically integrated sensors, actuators, and displays, fabricated by scalable and low-cost methods for real life applications.

 

This work was supported in part by National Science Foundation under Grant No. ECCS-1549888

 

 

 

Poster Number: ECE-09

Authors: Maria L. Castano, Anastasia Mavrommati, Todd Murphey, Xiaobo Tan

Title:  Trajectory Planning and Tracking of Robotic Fish Using Ergodic Exploration

 

Abstract: In recent years, underwater robots that propel and maneuver themselves like real fish, often called robotic fish, have emerged as promising mobile sensing platforms for freshwater and marine environments. For these active monitoring applications, efficient exploration along with economical locomotion is highly important, in order to optimize sensing coverage and guarantee long field operation time. As a result, optimization of the sensing trajectory and energy saving tracking of the planned trajectory are of interest. In this paper we adopt an ergodic exploration method to calculate an optimal sensing trajectory for a tail-actuated robotic fish, and propose a nonlinear model predictive control (NMPC) approach for tracking the generated trajectory. A high-fidelity, averaged nonlinear dynamic model is used for trajectory planning and control. In particular, the bias and amplitude of the tail-beat pattern are treated as the control inputs, and their physical bounds and the constraints on their changing rates are properly accounted for in the optimization process. Finally, simulation results are presented to illustrate the effectiveness of the proposed approach.

 

This work was supported in part by National Science Foundation (DGE 1424871, IIS 1319602, CCF 1331852, ECCS 1446793)

 

 

 

Poster Number: ECE-10

Authors: Dhrubajit Chowdhury, H. K. Khalil

Title:  Fast Consensus in Multi-agent Systems Using High Gain Observers

 

Abstract: We address the consensus problem of multi-agent systems for a static undirected communication topology. It is known that for a static undirected graph, the convergence rate of the consensus protocol depends on the second smallest eigenvalue of the graph Laplacian also known as the algebraic connectivity of the graph. The fastest convergence rate can be achieved when the given communication topology is complete which is referred to as a complete graph i.e., every agent is connected to every other agent which is costly in terms of the required number of communication links. On the other hand the star topology is quite ubiquitous in nature but the convergence rate of the consensus protocol with this topology is slower than the complete graph. In this paper we show that by adding observers to each of the nodes except the root node of a given star topology, the convergence rate of the consensus protocol with the star topology approaches the convergence rate of the consensus protocol achieved with a complete graph for sufficiently small $\epsilon$, which is a high-gain observer parameter. The observers estimate the missing connections and hence the star topology with the observers acts as if the network topology was complete. Furthermore we show that for sufficiently small $\epsilon$, the trajectories of the agents with the star topology approach the trajectories of the agents with the virtual complete graph.

 

 

 

Poster Number: ECE-11

Authors: Michael Craton, Jennifer Byford, Vincens Gjokaj, Premjeet Chahal, John Papapolymerou

Title:  3D Printed High Frequency Coaxial Transmission Line Based Circuits

 

Abstract: Coaxial transmission lines are among the most elementary high frequency transmission constructions. Their use is ubiquitous in RF and high frequency design and instrumentation, favored for their superior signal isolation. Furthermore, many microwave circuit components are easily implemented using coaxial structures. Filters can be designed with higher Q-factors using coaxial transmission lines as opposed to planar structures (e.g. microstrip). Since the advent of 3D printing, many microwave circuits have been demonstrated, but a coaxial transmission line has not yet been exhibited for high frequency transmission. Current implementations of coaxial transmission lines typically require a dielectric to support the signal conductor. This limits the performance of the waveguide--higher order propagating modes can appear in smaller diameter structures than equivalent air-dielectric geometries. Other implementations use more expensive subtractive manufacturing techniques. For many high frequency designs, it is impractical to use coaxial transmission lines, thus limiting the designer’s flexibility. Semi-rigid coax requires additional tooling and other available coaxial transmission lines can very quickly become prohibitively expensive. 3D printing allows for flexible designs and quick design cycles while also providing an inexpensive design solution. This poster demonstrates a 3D printed coaxial transmission line structure using polyjet printing technology and its implementation in the design of filters for operation up to 10GHz. The unique benefits that 3D printing technology provides make it well suited to address some of the current limitations of coaxial transmission line construction. These techniques provide a template for a coaxial transmission line implementation where it would otherwise not be possible.

 

This work was supported in part by MSU Foundation Professorship

 

 

 

Poster Number: ECE-12

Authors: Zane Crawford, Jie Li, Andrew Christlieb, Shanker Balasubramaniam

Title:  Advancements in the Mixed Finite Element Method for Electromagnetics

 

Abstract: The mixed finite element method (MFEM) for electromagnetics use curl-conforming basis functions to represent fields and divergence conforming basis functions to represent fluxes. Previous work on MFEM includes work on solving the coupled Maxwell equations, namely Faraday's and Ampere's law, for the electric field and magnetic flux density. However, the choice of a leapfrog scheme to represent the time derivatives in the coupled Maxwell equations causes CFL-like restrictions on the maximum time step size. Furthermore, a non-uniform spatial grid will cause the time step size to be smaller than needed to resolve the highest frequency content supported by the spatial grid. This can cause a bottleneck for examining wave phenomena, especially in the presence of charged particles, as more time steps must be taken than necessary to see behavior. Recent work has developed alternate time-stepping algorithms for MFEM for the coupled Maxwell equations that have different stability and accuracy properties. More importantly, in this work,we present a MFEM formulation with an unconditionally stable time-stepping algorithm for the coupled Maxwell equations. Additionally, we compare the stability and accuracy properties between leapfrog, Newmark-Beta, and predictor-corrector methods and preliminary work to extend the method to higher order spatial discretizations.

 

This work was supported in part by DOE CSGF, grant number DE-FG02-97ER25308

 

 

 

Poster Number: ECE-13

Authors: Thassyo Pinto, Le Cai, Chuan Wang, Xiaobo Tan

Title:  CNT-based Sensor Arrays for Local Strain Measurements in Soft Pneumatic Actuators

 

Abstract: Soft robotics is a recent trend in engineering that seeks to create machines that are soft, compliant, and capable of withstanding damage, wear and high stress. Soft pneumatic actuators (SPAs) are the key elements of soft robots, and their elastomeric substrate enables generation of sophisticated motion with simple controls. Although several methods for fabrication, material selection, and structure design have been investigated for the construction of SPAs, limited attention has been paid to the integration of distributed sensors for performing localized measurement. Carbon nanotubes (CNTs) are molecular-scale tubes of carbon atoms with remarkable mechanical and electronic properties, showing potential application in sensing devices. In this paper, we present the design, fabrication, and testing of a novel type of CNT-based sensor array combined with silver nanowires (AgNWs) for measuring localized strain along the bottom layer of a SPA. Simulation and experimentation have been performed in order to analyze the soft actuator deformation during bending. The results demonstrate the promise of the proposed SPA with integrated strain sensing, which lays groundwork for a myriad of applications in grasping, manipulation, and bio-inspired locomotion.

 

This work was supported in part by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) under the Science Without Borders program (BEX-13404-13-0); National Science Foundation (DBI-0939454 and ECCS-1549888)

 

 

 

Poster Number: ECE-14

Authors: Osama Ennasr, Xiaobo Tan

Title:  Distribute Tracking of a Moving Target with Time-difference-of-arrival Measurements

 

Abstract: Localization and tracking of a moving target has been established as a key problem in wireless sensor networks, with many algorithms being proposed in this area. In particular, time-difference-of-arrival (TDOA) localization is considered to be a cost-effective and accurate localization technique. However, traditional TDOA algorithms rely on a central node that produces an estimate of the target's location by gathering measurements from all other nodes in the network. In this work, we look at the problem of distributed localization and tracking of a moving target using TDOA measurements. Specifically, we examine the communication topologies among a network of agents that ensure successful localization of the target’s position. In a centralized scheme, estimation is only possible if the central node has 3 or more TDOA measurements. However, in our distributed approach, we employ a networked extended Kalman filter (NEKF) and show that it is possible for the entire network to estimate the target’s position without any agent having 3 neighbors, and therefore 3 TDOA measurements. For tracking the target’s movement, control strategies for all agents in the network are devised to enhance the distributed filter’s performance. Finally, we show that employing such a distributed architecture is more robust to communication dropouts with lower communication costs when compared to the centralized approach.

 

This work was supported in part by NSF

 

 

 

Poster Number: ECE-15

Authors: Vincens Gjokaj, John Doroshewitz, Premjeet Chahal

Title:  Design and Fabrication of Multi-frequency Antenna Using Genetic Algorithms for 5G Applications

 

Abstract: Under on design approach, fifth generation wireless network will converge together most of the available wireless network. To be able to converge all of these bands in a single platform would require multiple antennas placed in close proximity to each other in a highly dense electronic environment. This is not practical and thus design of single antennas that can work at all these frequencies is necessary. Here we propose a patch antenna design that will incorporate most of these bands in a single antenna. To design such an antenna is complex and to overcome the design challenge, in this paper genetic algorithm is utilized to design antenna for design frequency bands and also to achieve high gain, especially at higher frequencies. For the first design, one antenna that can operate in Wi-Fi, Bluetooth, and 4G LTE bands and operate with high gain at all those frequencies is investigated. For the design process, a patch antenna that can operate at the middle frequency band provides the starting point. By pixelating the antenna and then removing certain pixels help provides other bands. Designs that can provide high gain at all frequency bands are determined through multi-objective optimization technique. This paper will present the design of antennas for 5G application using Matlab to run the Genetic algorithm and to control ANYSIS High Frequency Structure Simulator that carried out detailed electromagnetic simulations of different antenna structures. Several antenna designs are fabricated and characterized over a wide-band and their performance is compared to simulated results.

 

 

 


 

Poster Number: ECE-16

Authors: Jason N. Greenberg, Xiaobo Tan

Title:  LED-based Localization of Mobile Robots

 

Abstract: Achieving Simultaneous Localization And Communication (SLAC) is a great asset to resource-limited robots since it reduces the complexities of capturing the data of two essential sources of information. In this work a method for localizing a mobile robot using the line of sight (LOS) detection of an LED communication system is presented. In particular, in a two-dimensional setting two base nodes use the lines of sight between themselves and the mobile agent to acquire the latter's bearings which are then used to compute its location. The technique used in this work uses a Kalman filter to predict the position of the robot based on past localization results. This allows the base nodes to significantly reduce the search range in establishing LOS with the mobile node and consequently improve the temporal resolution and spatial precision of the localization which would otherwise be limited by extensive scanning. Extensive experimental results are presented to illustrate and support the approach.

 

This work was supported in part by National Science Foundation (IIS 1319602, CCF 1331852, ECCS 1446793)

 

 

 

Poster Number: ECE-17

Authors: Kalyanmoy Deb, Rayan Hussein, Proteek Roy, Gregorio Toscano

Title:  Metamodeling Methodologies for Multi-objective Optimization

 

Abstract: In many practical optimization problems, evaluation of objectives and constraints often involve computationally expensive procedures. To handle such problems, a meta-model-assisted approach is usually used to complete an optimization run in a reasonable amount of time. A meta-model is an approximate mathematical model of an objective or a constraint function which is constructed with a handful of solutions evaluated exactly. However, when comes to solving multi-objective optimization problems involving numerous constraints, it may be too much to meta-model each and every objective and constraint function independently. The cumulative effect of errors from each meta-model

may turn out to be detrimental for the accuracy of the overall optimization procedure. In this paper, we propose a taxonomy of various

meta-modeling methodologies for multi-objective optimization and provide a comparative study by discussing advantages and disadvantages of each method. The first results presented in this paper are obtained using the well-known Kriging meta-modeling approach. Based on our proposed taxonomy and an extensive literature search, we also highlight new and promising methods for multi-objective meta-modeling algorithms.

 

 

 

Poster Number: ECE-18

Authors: Daniel Morris, Saif Imran

Title:  Leaf Segments from Depth Images

 

Abstract: Automatically segmenting leaves in imagery of dense foliage in a challenging task. This paper proposes a method

to segment leaves and leaf-parts in depth images of foliage.

The method relies on locally defined pixel affinities and does

not require prior leaf shape models. As a result it is general enough to work on a wide variety of leaves. We develop

a novel approach to expanding affinities beyond adjacent

pixels which we call Shortest Path Affinity. These Shortest

Path Affinities lead to improved segmentation when used by

normalized cuts. We also incorporate them in a hierarchical segmentation method that creates superpixels and then

groups superpixels recursively. This new method greatly

speeds up segmentation computation while obtaining similar performance to normalized cuts. The result is an effective leaf segmentation method for depth images.

 

 

 


 

Poster Number: ECE-19

Authors: Ameer Janabi, Kellen Hilton, Bingsen Wang

Title:  Variable Mode Model Predictive Control for Minimizing the Thermal Stress in Electric Drives

 

Abstract: A control scheme based on variable mode model predictive control is proposed. By augmenting the conventional model predictive controller with a variable cutoff-frequency/order low pass filter, an additional degree of the objective function allows for optimal trade-off between the thermal stress reduction and the harmonic performance. Both the simulation and experiment results have validated the effectiveness of the proposed method.

 

 

 

Poster Number: ECE-20

Authors: William Jensen, Shanelle Foster, Elias Strangas

Title:  Online RUL Estimation of Stator Winding Insulation using Transient Peak Detection

 

Abstract: Wide bandgap semiconductor devices in machine drive topologies are becoming more prevalent. These devices improve efficiency and can operate at higher switching frequencies. However, higher switching frequencies will increase the electrical stress applied to the insulation of the machine. Electrical stress, excessive heating, mechanical vibrations, and environmental contamination are leading factors that contribute to insulation degradation. Breakdown of the insulation will lead to a short circuit fault between two conductors. Short circuit faults can quickly propagate and lead to catastrophic failure. This project proposes a method to predict the remaining useful life (RUL) of the stator insulation by monitoring a trend in the leakage current. As the insulation degrades, the magnitude of the overshoot in the transient response of the leakage current exponentially decreases. An analog peak detection circuit is used to acquire this magnitude in lieu of expensive high frequency sampling. An Extended Kalman Filter is applied to predict the RUL of the insulation. The proposed strategy will improve machine reliability, especially when using wide bandgap devices, and will eliminate system failures and unplanned shutdowns caused by short circuits.

 

This work was supported in part by James Dyson Foundation Fellowship

 

 

 

Poster Number: ECE-21

Authors: Saranraj Karuppuswami, Harikrishnan Arangali, Premjeet Chahal

Title:  A Hybrid Electrical-mechanical Wireless Magnetoelastic Sensor for Liquid Sample Measurements

 

Abstract: This poster presents a hybrid passive wireless resonant electrical and mechanical sensor for enhanced sensitivity and specificity. Mechanical resonance measures the viscosity and electrical resonance measures the dielectric properties of liquid samples. The sensor is composed of two magnetoelastic (amorphous ferromagnetic ribbons) strips placed in parallel that are separated with a dielectric spacer forming a capacitor. An inductive coil is attached in parallel to this capacitor leading to an electrical resonant inductive-capacitive (L-C) tank. Both mechanical and electrical resonance frequencies are wirelessly measured using a single pickup coil conntected to an impedance analyzer. Several liquid samples, including food items, having different viscosity and dielectric properties are measured and key advantages of this sensor are demonstrated. This sensor can be used in food quality monitoring and can be integrated with passive RFIDs.

 

This work was supported in part by Midland Research Institute for Value Chain Creation

 

 

 

Poster Number: ECE-22

Authors: Aqeel Madhag, Guoming Zhu

Title:  Estimating Slow Varying Sensor Noise Covariance using a Modified Adaptive Filter

 

Abstract: Modern control systems heavily relay on sensors for feedback control. Degradation of sensor performance due to sensor aging affects the closed-loop system stability, reliability, and performance. This paper proposes an algorithm used to identify the time-varying sensor noise

covariance online based on system sensor measurements. The covariance-matching technique, along with the adaptive Kalman filter, utilizes the information about the quality of weighted innovation sequence to estimate the time-varying sensor noise covariance. The covariance-matching of the weighted innovation sequence improves the prediction accuracy and reduces the computational load, which makes it suitable for online applications. The sequential manner of the proposed algorithm leads to significant reduction of the computational load. The convergence proof of the proposed algorithm is demonstrated. In addition, the upper and lower bounds of the estimation window length are derived such that the convergence of the proposed algorithm is guaranteed. Simulation results show that the proposed algorithm is capable of estimating the time-varying sensor noise covariance for MIMO systems with white noise whose covariances vary linearly or exponentially. Furthermore, the proposed estimation algorithm

demonstrates a reasonable convergence rate.

 

 

 

Poster Number: ECE-23

Authors: Yaqub Mahnashi, Fang Z. Peng

Title:  Smart Solar Cell: Concept and Design

 

Abstract: Wireless sensor nodes (WSN) and implantable biomedical devices require constant battery maintenance which is costly and sometimes invasive. Solutions to prolong the battery life is highly desirable. Solar cells is considered a very powerful energy source for these applications and many others. However, the open circuit voltage ‘Voc’ of a regular solar cell is highly dependent on the light intensity. To overcome this problem, solar cells with regulated output voltage feature is presented which we call a smart solar cell. In this paper, the smart solar cell concept is introduced. A case study of miniaturized smart solar cell with 300~600mV open circuit voltage and 1.5V regulated output voltage is discussed. Simulation results are conducted and discussed to show the effectiveness of the smart solar cell compared to the regular counterpart.

 

This work was supported in part by Ministry of Education of Saudi Arabia through King Fahd University of Petroleum and Minerals (KFUPM)

 

 

 

Poster Number: ECE-24

Authors: Yaqub Mahnashi, Fang Z. Peng

Title:  Variable Fibonacci Switched-capacitor Converter Synthesis

 

Abstract: Synthesizing the switched-capacitor (SC) converter to perform certain voltage conversions is considered a challenge in SC converter design and usually is done in ad-hoc way. This paper presents a systematic approach to synthesize an optimal SC converter for multiple voltage-gain-ratio (VCR) applications using Fibonacci SC topology. The optimization methodology is based on the minimum number of components count. This approach exploits the Fibonacci SC canonical model and extends it to variable converter implementation. In addition, using the method presented in this paper, the controller circuit of the designed SC converter is greatly simplified. Case studies are presented to show how different SC converters can be synthesized. Results of different designs are compared to some proposed converters in the literature which shows the versatility and simplicity of the design using proposed method.

 

This work was supported in part by Ministry of Education of Saudi Arabia through King Fahd University of Petroleum and Minerals (KFUPM)

 

 

 

Poster Number: ECE-25

Authors: Jinshui Miao, Bo Song, Zhihao Xu, Le Cai, Suoming Zhang, Lixin Dong, Chuan Wang

Title:  Infrared Photodetector and Camera Using Layered Black Phosphorus

 

Abstract: Few-layer black phosphorus (BP) is an emerging two-dimensional semiconducting material which is of great interest for applications, mainly in high performance electronics, optoelectronics, and chemical sensors. BP nanoflake has a moderate bandgap of around 0.3 eV and high carrier mobility, which lead to transistors having high on-state current and on-off ratios. Here, we demonstrate the infrared photodetector and camera using layered black phosphorus. The BP photodetector shows respectable responsivity up to ~ 100 A/W. We integrated a BP photodetector in a single-pixel camera based on compressive sensing. We proposed a system for calibrating the optoelectrical properties of micro/nano photodetectors based on digital micromirror devices (DMD), which changes the light intensity by controlling the number of individual micromirrors. The calibration sensitivity is driven by the sum of all micromirrors of the DMD. The single-pixel imaging system with the BP photodetector was used to recover a static image to demonstrate the feasibility of the single-pixel imaging system.

 

This work was supported in part by Michigan State University

 

Poster Number: ECE-26

Authors: Nicholas Miller, John Albrecht, Matt Grupen

Title:  Large-signal RF Simulation and Characterization of Power Amplifiers using Fermi Kinetics Transport

 

Abstract: Wireless communications is an ever-growing research area in the information age. Commercial applications include cellular telephones, wireless local area network (WLAN) and Wi-Fi, and global positioning system (GPS). A critical component of all wireless systems is the radio frequency (RF) and microwave amplifier. As bandwidth and efficiency criteria evolve with more sophisticated technology, so too must the underlying transistors which comprise the RF amplifiers. RF power amplifiers (PAs) are an important sub-component of TRX modules. GaN HEMTs are a popular PA transistor technology choice due to their high electron saturation velocity, large semiconductor band-gap, high 2D electron gas (2DEG) density in the channel, and high electron mobility. Design of RF PAs is significantly more involved than the design of other types of amplifiers. Basic requirements of these PAs include high gain, high power added efficiency (PAE), and higher linearity. Once the device technology, the type of PA, and the operating point is selected, design of the PA requires measured and/or simulated load-pull (LP) data. Simulated LP data is typically generated with standard commercial software, including Keysight's Advanced Design System (ADS). This requires, however, circuit models of the underlying transistors in order to simulate the PA and generate LP data. The purpose of this abstract is to utilize a TCAD framework called Fermi Kinetics Transport (FKT), which captures hot-electron and full-wave effects, to extract X-Parameter models of state-of-the-art GaN HEMT technology. These X-Parameter models will be imported into ADS to generate simulated LP data.

 

This work was supported in part by AFOSR (#17RYCOR495); AF STTR program (FA8650-16-C-1764); SMART scholarship program

 

 

 

Poster Number: ECE-27

Authors: Mohd Ifwat Mohd Ghazali, Saranraj Karuppuswami, Amanpreet Kaur, Premjeet Chahal,

Title:  3D Printed Air Substrates for the Design and Fabrication of RF Components

 

Abstract: This paper presents the fabrication and characterization of RF and microwave passive structures on an air substrate using additive manufacturing (3-dimensional, 3D, printing). The air substrate is realized by 3D printing RF structures in two separate pieces and snapped together face to face using a LEGO-like process. Spacers printed on the periphery provides the desired air substrate thickness. Metal patterning on non-planar printed plastic structures is carried out using a damascene-like process. Various RF structures such as low dispersion transmission line, T-line resonator, high gain patch antenna, slot antenna and cavity resonator are demonstrated using this process. Good performance is achieved; for example, measured 50Ω transmission line shows low loss of 0.17 dB/cm at 4 GHz and a patch antenna (center frequency of 4.5 GHz) shows gain and bandwidth of 7.6 dBi and 0.2 GHz, respectively. Details of both measured and simulation results are presented.

 

 

 

Poster Number: ECE-28

Authors: Matthias Muehle, Jes Asmussen, Michael Becker, Thomas Schuelke

Title:  High quality SCD synthesis with microwave plasma assisted CVD at pressures between 300 and 400 Torr

 

Abstract: Fabricating single-crystalline diamond (SCD) wafers exceeding 1” dimensions, requires serious synthesis effort. For example, with typical growth rates of 30 µm/hour it takes about 2000 hours of growth time to make a 3 cm by 3 cm diamond plate. Stable CVD SCD growth processes for such long deposition times have not yet been demonstrated. Thus, it is very desirable to significantly increase the growth rate, while maintaining or improving the SCD quality. The recent development of new growth reactor technology increased the safe and efficient diamond synthesis process window toward higher pressures up to 300 Torr [1]. The main motivation of increasing the process pressure is to achieve higher SCD growth rates while reducing defects [2]. However, any further increase in process pressure resulted in unstable plasma conditions due to the use of a microwave power supply pulsed at 120 Hz. In this poster, we report on further increasing process pressures to 400 Torr, which is possible with a power supply that can be switched between continuous and pulsed excitation.  In this study, we demonstrate high quality SCD synthesis in the so-far unexplored pressure region between 300 and 400 Torr. The effects of several parameters of the multidimensional parameter space are discussed. The samples were analyzed for growth rates, film morphology, optical absorption, birefringence and nitrogen content. References [1] Lu et al., Diamond and Related Materials 37 (2013), 17-28; [2] Silva et al., Diamond and Related Materials 18 (2009), 683-697

 

 

 

Poster Number: ECE-29

Authors: Asad Nawaz, Cagri Ulusoy

Title:  A 1W SiGe Power Amplifier

 

Abstract: High-speed silicon-based transistors suffer from low breakdown voltages (~2VCEO). Generation of high output power without degrading the reliability is still a major design challenge. This paper presents a watt-level X-band power amplifier realized in silicon-germanium 0.13um HBT technology. The proposed PA cell consists of modified cascode stage with base-degenerated CB transistor. Base degeneration increases amplifier reliability by sharing the voltage stress between both CE and CB transistors. Higher current swings are achieved by sizing CB transistors three time as big as CE transistors, this helps reducing output impedance while maintaining input impedance constant. Emitter and base ballasting have been included to provide thermal stability. The PA cell is biased in class-B and achieves 30dBm output power with 50% PAE. This is the state-of-art performance ever achieved from silicon-based transistors power cell.

 

This work was supported in part by IHP Microelectronics

 

 

 

Poster Number: ECE-30

Authors: Nga Nguyen, Joydeep Mitra

Title:  Effect of Wind Power on Load Frequency Control

 

Abstract: The integration of wind power into power systems has been gaining momentum in the global energy industry due to its environmental benefits and abundance of supply. However, the natural intermittent and non-dispatchable features of wind negatively impact the system’s frequency regulation capability. Wind power not only injects additional fluctuations to the already variable nature of frequency deviation, it also

decreases frequency stability by reducing the inertia as well as the regulation capability. This reduction causes the decrease of overall system frequency response characteristic. These effects of wind integration will be examined closely in this paper. In addition, the effect of wind power on tie-line power flows and area control error will be investigated. The analytical and simulation model of load frequency control are utilized to show the impact of wind on system frequency regulation. Additionally, a range of wind penetration levels is considered to determine the maximum wind power penetration level given a frequency deviation limit.

 

This work was supported in part by US Department of Energy under award no. DE-OE0000625

 

 

 

Poster Number: ECE-31

Authors: Christopher Oakley, Jennifer A. Byford, Amanpreet Kaur, Premjeet Chahal

Title:  Aerosol Jet Printing of THz Passive Components

 

Abstract: Terahertz systems offer a convenient, non-invasive platform for biomedical imaging, remote sensing, substance detection, communications systems and material characterization. Due to the high cost of system components needed for proper operation such as mixers, filters, polarizers, and amplifiers, large-scale deployment of these systems has not been realized. Additive manufacturing has recently been demonstrated to provide a viable path towards low-cost, rapid fabrication of lenses, waveguides, probes and power splitters for operation in this frequency spectrum. These methods offer many advantages over traditional micromachining and lithographic techniques by reducing the need for a skilled operator, as well as eliminating hazardous waste traditionally generated as a by-product of these fabrication techniques.

Aerosol jet printing of materials allows for a non-contact, direct-write methodology to fabricate structures with feature sizes as small at 10 micrometers, with layer thicknesses of as little as 300 nanometers. Metals, polymers and other materials can be deposited on non-planar surfaces, allowing for the integration of filters and other structures with other components of interest.

The goal of this poster is to demonstrate viability of aerosol jet printing of passive terahertz filters on thin organic substrates. Performance of both band-pass and band-stop printed filters will be compared to similar structures fabricated using traditional lithographic techniques, over several frequency bands.

 

 

 


 

Poster Number: ECE-32

Authors: Thang Pham, William Jensen, Shanelle Foster

Title:  Stator Incipient Fault Identification in Short Secondary Linear Permanent Magnet Synchronous Machines

 

Abstract: Failures in linear permanent magnet synchronous machines (LPMSMs), especially those used in transportation and mining applications, can be catastrophic. Quick detection and location of incipient stator winding faults during normal operation provides time to deploy mitigation methods that slow the propagation of the fault and prolong the post-fault life of the machine; however, little has been done to locate the position of the fault in the winding while the machine is operating.  In this work, a technique is proposed to detect and locate incipient turn-to-turn faults in short-secondary linear permanent magnet synchronous machines using voltage variations. The dynamic behavior of the circulating fault current generated by the permanent magnets introduces a unique signature voltage that is used to locate the fault. Identifying the faulty slot during normal operation provides additional flexibility for post-fault operation of short-secondary LPMSMs.

 

 

 

Poster Number: ECE-33

Authors: Jubaid Qayyum, John Papapolymerou, Ahmet Ulusoy

Title:  A System-on-package Dielectric Sensor using Additively Manufactured Interconnects

 

Abstract: In this paper, the authors present an on-chip dielectric sensor based on six-port architecture. The designed sensor, integrated on-chip with the six-port reflectometer (SPR), is being fabricated in a 0.13-µm 300GHz-fT/450GHz-fmax SiGe process technology. The heterojunction bipolar transistors (HBTs), implemented as power detectors for the SPR, yield a responsivity of 5123 V/W at 94 GHz based on simulation results. A 450-µm long shorted coplanar-waveguide (CPW) line was designed as the sensor to detect the phase variation in the reflection coefficient when the dielectric constant above the CPW changes. A system-level simulation using ADS (Advanced Design System) has been conducted to analyze and evaluate the reflectometer performance and the chip layout is done using Cadence Virtuoso. The chip occupies an area of 1179-µm × 404-µm and the reflectometer itself consumes 6.75 mW from a 1.5 V power supply. The system will be packaged using 3D printed interconnects using aerosol jet printing (AJP) technology. As a proof of concept, a trapezoidal structure was 3D printed on Liquid Crystal Polymer (LCP) and CPWs were printed on top of them to imitate mm-wave packaging. The printing was done using silver nanoparticle ink that acquired 40% conductivity of the bulk silver after sintering at 200 ˚C for one hour. The CPW interconnects yielded insertion loss of as low as 0.49 dB/mm including the trapezoid, and with a loss of 0.38 dB/mm on LCP substrate at 110 GHz. This work also represents AJP as a solution for cost-effective system-on-package (SoP)/ millimeter-wave (mm-wave) systems.

 

 

 

Poster Number: ECE-34

Authors: Mohammad R. Rawashdeh, Lalita Udpa, S. Ratnajeevan H. Hoole, Yiming Deng

Title:  Subregion Finite Element Method based Defect Detection and Characterization using Eddy Current Techniques

 

Abstract: A novel study of using Subregion Elastic Finite Element Method is presented. The idea of minimizing processing time and saving memory especially for large scale and complicated practical problems is one of the major problems that researches and engineers are facing. A practical example of detecting hidden cracks in infield military vehicles damaged by Improvised Explosive Devices is presented where subregion method is used to isolate defect region from the entire domain. This will lead to tangibly computational cost minimization. Then Finite Element solution has been developed to each region separately, next, to the affected region alone by using the Elastic mesh generation technique that improved in order to apply mathematical inverse problems techniques on this selected affected region alone then recombine solution with the entire solution of the existing problem. A flexible and elastic mesh generating scheme is developed in order to change the defect preselected design parameters each iteration. This meshing technique adds the specialty for using subregion method in such FEM inverse problems where the presented mesh generation process will go through this elastic algorithm so element numbering can be saved inside and outside defect region. In order to characterize the hidden cracks, both of Genetic Algorithm and Simulated Annealing optimization techniques are used. An excellent agreement between our results and others using different techniques. The CPU simulation process was minimized by 90% after using Subregion FEM instead of classical FEM, while an accuracy of 98% was achieved by reconstructing the original defect.

 

This work was supported in part by NDEL

 

 

Poster Number: ECE-35

Authors: Montassar Aidi Sharif, Xiaobo Tan

Title:  Modeling and Fabrication of a Bio-inspired Canal Lateral Line System

 

Abstract: It is of interest to exploit the insight from the lateral line system of fish for flow sensing applications. Biological lateral lines consist of arrays of so-called neuromasts formed with hair cells encased by a gel-like structure named cupula. There are two types of neuromasts, superficial neuromasts and canal neuromasts. In this work we investigate the modeling, design, and fabrication of an artificial lateral line system with canal neuromasts. The canal is filled with viscous fluid to match the biological counterpart. The artificial neuromast is made by embedded an ionic polymer-metal composite (IPMC) sensor within a soft molded cupula structure. The displacement of the cupula structure under an oscillatory flow is studied using both an analytical solution in the literature and finite-element computation with COMSOL 5.1 Multiphysics and fluid structural interaction (FSI) module. The effect of material stiffness and fluid viscosity on the device sensitivity is examined. The sensing response of the IPMC-based canal neuromast is experimentally characterized. Ongoing work aims to estimate the system parameters and validate the modeling analysis with experiments, followed by model-based design optimization to maximize the sensing performance.

 

This work was supported in part by Office of Naval Research (Grant N000141512246)

 

 

 

Poster Number: ECE-36

Authors: Hongyang Shi, Xiaobo Tan

Title:  Bio-inspired Design of Sea Lamprey Robot Based on the Suction and Swimming Behavior of Sea Lamprey

 

Abstract: Sea lampreys utilize attachment by suction and swim by rhythmic lateral undulations of the body axis. The goal of this study was first to understand the seal mechanism of the oral disc, the pressure creation within the sea lamprey mouth during the suction, and the neural control of swimming. Inspired from the biological mechanisms, a robotic sea lamprey would be designed. Then, we would test the suction efficiency of the robotic sea lamprey attached to some surfaces with different textures, and the swimming test would also be conducted separately to verify the neural control method. After that, we would like to conduct swimming and suction experiment in a higher flow environment, which would drive us to design a robotic sea lamprey with better suction efficiency and swimming performance.

 

 

 

Poster Number: ECE-37

Authors: Pratap Bhanu Solanki, Xiaobo Tan

Title:  Extended Kalman Filter based Active Alignment Control for LED Optical Communication

 

Abstract: Light-emitting diode (LED)-based optical communication is emerging as a low-power, low-cost and high-data rate alternative to acoustic communication for underwater applications. However, it requires a close-to-line-of-sight link between the transmitter and the receiver to achieve an efficient communication link. Alignment for maintaining line-of-sight(LOS) is challenging due to the constant movement of the underlying mobile platform that is caused by unwanted disturbances and/or the propulsion. In this paper we propose a novel and compact LED-based communication system with active alignment control using a single receiver, which maintains the LOS link despite the underlying platform movement. The system design and implementation is discussed. An Extended Kalman Filter-based algorithm is proposed, which uses light intensity measurements from the single receiver photo-diode and a scanning technique to estimate the relative orientation between the receiver and the transmitter. This estimated orientation is used to adjust the receiver's orientation accordingly. The proposed algorithm is verified in simulation and experiments. The algorithm outperforms simpler algorithms like hill-climbing and three-point-averaging under noisy environment.

 

This work was supported in part by National Science Foundation (IIS 1319602, ECCS 1446793)

 

 

 


 

Poster Number: ECE-38

Authors: Samer Sulaeman, Joydeep Mitra

Title:  A Direct Method to Calculate Capacity Value of Variable Energy Resources

 

Abstract: This work proposes a direct method to calculate the capacity value of variable energy resources (VERs) in probabilistic adequacy planning studies. The capacity value is used to measure the contribution of VERs such as wind power, to the adequacy of generation systems. Iterative methods are traditionally used to calculate the capacity value of VERs. However, iterative methods are computationally demanding even for power system planning studies. The proposed method is based on augmenting the cumulative distribution function (CDF) of the generation reserve margin (prior to adding VERs) to include the output power of VERs. From the augmented CDF of the generation reserve margin, the capacity value is analytically determined without performing iterations. The proposed method reduces modeling complexity and the computational burden associated with calculating the capacity value of VERs. The proposed approach considers the correlation with system load, output variability and forced outages of generation units in the evaluation process. Furthermore, using the proposed method, the optimal penetration level of VERs in terms of their capacity value can be determined with greater ease than with iterative methods. The proposed method is demonstrated on the IEEE Reliability Test System (IEEE RTS) and used to calculate the capacity value of wind power.

 

 

 

Poster Number: ECE-39

Authors: Yuting Tian, Mohammed Benidris, Samer Sulaeman, Salem Elsaiah, Joydeep Mitra

Title:  Optimal Feeder Reconfiguration and Distributed Generation Placement for Reliability Improvement

 

Abstract: This work presents a methodology to determine the optimal distribution system feeder reconfiguration and distributed generation (DG) placement simultaneously, and is optimal in that the system reliability is maximized. An important consideration in optimal distribution system feeder reconfiguration is the effect of the variable output of intermittent resources. The work presented in this paper considers the stochastic behavior of variable resources, and open/close status of the sectionalizing and tie-switches as variables in determining the optimal DG locations and optimal configuration that enhance system reliability. Genetic algorithm (GA) is applied to search for the optimal solution. The proposed method is demonstrated on a 33-bus radial distribution system, which is extensively used as an example in solving the distribution system reconfiguration problem.

 

 

 

Poster Number: ECE-40

Authors: Haojun Wang, Chaojian Hou

Title:  Self-assembly Nano Robot Carrier

 

Abstract: During recent years, a substantial development on precisely controlled drug delivery technology has changed medical treatments in improving patient conformity and accessibility. However, strategies, even for those most advanced, cannot eliminate side effects, especially when considering both control veracity and compatibility. Improving biocompatibility comes at the expense of reduced controllability, which has an unacceptable side effect due to carriers may interact with untargeted body cells. In favor of controllability may sacrifice the harmony between carriers and biological cells. Side effects have restricted further progress in clinical practice. Here, our Automation & Robotics Laboratory introduce a new way striking a balance between both sides, which is named Self-Assembly Nano Robot Carrier. Self-assembly nano robot carrier is made up of a certain amount of nanoparticles. Aggregation or disassembly follow scientists’ inclinations. Assembled nano-robot can be directed to areas of the body affected by a disease, such as cancer, tumor or neurodegenerative disease. Disassembled nanoparticles can be easily metabolized due to its nanoscale sizes. This controllable nano drug carrier offers a historical solution for clinicians to overcome cellular barriers and negative effects.

 

 

 

Poster Number: ECE-41

Authors: Tongyu Wang, David Torres, Chuan Wang, Felix Fernandez, Nelson Sepulveda

Title:  Maximizing Performance of Photothermal Actuators by Combining Vanadium Dioxide and Single-wall Carbon Nanotubes

 

Abstract: The development of photothermal actuators has been hindered by failing to solve the unavoidable trade-offs between performances such as high energy density, speed, deflection, power efficiency, and sensitivity. Improving some of these parameters often implies deterioration of others. Vanadium dioxide (VO2)-based actuators have demonstrated great performance in terms of strain energy density, speed, reversible actuation, programming capabilities, and large deflection. The relative low phase transition temperature of VO2 (~68 oC) gives this technology an additional advantage over typical thermal actuators in terms of power consumption. However, VO2’s low optical absorption still limits its application in photothermal devices. Here we report a VO2-based actuator technology that incorporates single-wall carbon nanotubes (SWNTs) as an effective light absorber to reduce the photothermal energy required for actuation. It is demonstrated that the chemistry involved in the process of integrating SWNT films with VO2-based actuators does not alter the quality of the VO2 film, and that the addition of such film enhances the actuator performance in terms of speed and responsivity. More importantly, the results show that the combination of VO2 and SWNT thin films is an effective approach to increase the photothermal efficiency of VO2-based actuators. Furthermore, by utilizing SWNT films with different chirality distribution, wavelength selective VO2 based photothermal actuators have been demonstrated. These devices response to selected optical spectra based on the chirality of SWNT coatings, which enables the future development of micro-robots, mechanical logic gates and electronic devices that are triggered by optical radiation from different frequency bands.

 

 

 

Poster Number: ECE-42

Authors: Suoming Zhang, Le Cai, Jinshui Miao, Zhibin Yu, Chuan Wang

Title:  Patterning Silver Nanowires on Various Substrates using a Direct Printing Method for Stretchable Conductors and Sensors

 

Abstract: we have developed a direct printing process for additively patterning AgNWs with length up to ~ 40 µm on various substrates. Well-defined and uniform AgNW features could be obtained by optimizing the printing conditions like nozzle size, ink formulation, surface energy, substrate temperature, and printing speed. Systematic characterizations were performed to investigate the electrical and electromechanical properties of the printed features with different nanowire lengths. By printing the AgNWs on a bi-axially pre-stretched PDMS substrate, we have realized a stretchable conductor that could maintain stable conductance under an areal strain of up to 156% (256% of the original area). Additionally, using the printed parallel AgNW lines as electrodes, we have fabricated an ultrasensitive capacitive pressure sensor array and a high resolution PLEC display on flexible substrates, implying the great potential of this unique additive patterning method. Furthermore, the same strategy can be applied to other material platforms like semiconducting nanowires, which may offer a new entry to various nanowire-based mechanically compliant sensory and optoelectronic systems.