Session 4 Abstracts

Kite Wind Energy Plant Ground Station Design and Implementation

Jing Guo
Electrical Engineering
Co-author: Florian, Electrical Engineering Ph.D

Absolutely, the clean and alternative energy is important and it is more commercially available than ever, but still less than 5% of global energy production with the huge investment and low energy density. Over time, wind turbine structures have generated more power by becoming taller, heavier and more expensive. In order for clean energy sources like wind to gain mass adoption, they must become as cheap and available as traditional fossil fuels. It?€?s time to rethink wind power.

An energy kite operates on the same principles as a conventional wind turbine, but tethered to the ground like a kite. Kite wind energy plants consist of a parachute winch to Winch Generator System (WGS) on ground. Energy is produced through a tethered airfoil that flies in large circles at up to 1,000 feet altitude, where the wind is much stronger and pretty more consistent than the winds reached by conventional systems. As it flies in a yoyo process, air moving across the energy kite rotors forces them to spin, producing electricity, which travels down the tether and into the grid. It is amazing that power kite has the ability to eliminate 90% of the material used in conventional wind systems, while generate the totally same amount energy.

However, there are some born deficiencies of the Power Kite against all of the advantages above. The energy generated by power kite shakes a lot and it is easily disturbed even by a little influence by errors inside the system or by the unpredicted atmosphere outside. At the same time, when the generation mode of power kite has been done, the Kite Wind Energy Plant Ground Station promises to pull the kite back safely. During the pull-back mode, zero error happening must be guaranteed. Kite Wind Energy Plant Ground Station should keep the stability of power kite, generate high quality electric and fight against all kinds of disturbance both inside and outside.

Fortunately, the OCC Converter is very suitable to solve this challenge. In recent years, the UCI Power Electronics Laboratory has proposed several new topologies for multi-level converters. These converters have many advantages over conventional converters, such as reduced input disturbance, constant power flow, reduced energy storage requirements, etc.
With OCC-Converter the capacity factor of Kite Wind Energy is higher, then the capital investments are much lower, finally, the price of energy will be lower.

Timestamp-Free Synchronization for Wireless Body-Area Networks

Rohan Ramlall
Electrical Engineering

Telemonitoring of biosignals is a growing area of research due to the aging world population. Telemonitoring utilizes a wireless body-area network (WBAN) consisting of wearable biosignal sensors (i.e., wearable technology) equipped with ultra low power radios. The measured data from each sensor on the patient is sent to a central communication node (e.g., smartphone or personal computer), which then sends the data to a healthcare provider via the internet. Thus, the patient?€?s health is monitored continuously and remotely in real-time without the need for the patient to visit their doctor.

One of the major constraints in WBANs is power consumption, since these sensors are meant to be used for weeks, months, and even years. The power consumed by wirelessly transmitting the data to the central communication node is orders of magnitude higher than the power consumed by any other operation, and thus, must be minimized.

To enable real-time monitoring of the biosignals, it is critical to have accurate timestamped data from the sensors in the WBAN. For example, if a sensor uses a typical low cost 32,768 Hz crystal oscillator with a frequency stability of 100 ppm, the time offset can be as high as 259 seconds after 1 month of use without any synchronization algorithm. Most of the synchronization algorithms presented in the literature require the exchange of dedicated timing messages containing digital timestamps on the network. However, this is not feasible for WBANs due to the high power cost associated with transmitting messages.

The contribution of this research is a novel timestamp-free synchronization algorithm applicable to power constrained networks such as WBANs. In the proposed algorithm, the synchronization algorithm is embedded in the existing network messages, so there is no additional overhead and power cost of exchanging dedicated timing messages. ”

Building energy: how to make a fuel cell and a battery work together to generate electricity effectively?

Gia Nguyen
Mechanical and Aerospace Engineering
Co-author: Jack Brouwer (UCI)

What is the best strategy to operate a fuel cell-battery system to generate electricity for a building? Traditionally, large central power plants produce the electricity for use in buildings with good efficiency and reliability. However, with the diminishing fossil fuel reserve, and the environmental impact of energy production, it is important to produce electricity with higher efficiency and reliability, and cutting harmful emissions. A different approach is to use small and efficient engines near buildings. This method can increase the reliability of the electricity and reduce the emissions of pollutants. To produce electricity for a building, an energy system needs to vary the power output quickly to accommodate the dynamic electrical demand.

A candidate for distributed generation is fuel cells. Fuel cells can produces electricity with high efficiency and low pollutant emissions, but can only vary the power output slowly. In contrast, batteries can store and discharge electricity quickly. Batteries can act as buffers for fuel cells, storing electricity when the demand is low, and discharging electricity when the demand is high. The combination of the fuel cell and battery creates a system with high efficiency and ability to vary the electrical output. My research studies the coordination needed for a fuel cell-battery system.

First, we collected high-resolution electrical demands from buildings in Southern California. With these data, we can simulate how the fuel cell-battery system must operate to satisfy the electrical demand.

Second, we develop a controller for the fuel cell-battery system. Using weather and time of the day, the controller creates a forecast on electrical consumption for every hour over a 24-hour period. With this forecast, the controller determines the operation to run the fuel cell with slow variation in power output, and uses the electricity produced from the fuel cell for the building or store it in the battery. The controller attempts to maintain the fuel cell and battery within their operation regime. The result of the optimization is the power set points for fuel cell and battery for every hour over a 24-hour period.

At the next time step, the controller updates a new forecast of electrical demand, and reruns the optimization. By constantly updating the new forecast and running the optimization, the controller developed in this research automatically coordinates the operation of a fuel cell-battery hybrid system, thus providing electricity for building and maintaining all equipment within their operation regime.

Products two Cantor sets and application to the Labyrinth model

Yuki Takahashi

Cantor set is a bounded subset of the real line which does not contain any interval, and has no isolated point. Physicists and Chemists have long believed that such sets never appear in nature, but recently, it was discovered that quasicrystals, which is a material gave rise to a paradigm shift in material science, have Cantor set as their spectrum. Investigation of the spectral properties of quasicrystals is a subject which have fascinated Physicists and Chemists, but it is known that even numerical investigations present a challenge, due to the aperiodic structure of quasicrystal.

Recently, it was found out that many spectral properties which have been considered to be impossible to show experimentally, can be proven rigorously using Mathematics. Nowadays, Mathematics is one of the most crucial tool to understand quasicrystals.

In our research, we consider the Labyrinth model, which is a two dimensional quasicrystal model. So far, almost nothing was known rigorously for this model, and all the works have been centered on numerics. We prove some spectral properties of this model, which gives an insight to the Physics and Chemistry society for further understanding of quasicrystals.
We first show that the spectrum of this model is given by the product of two Cantor sets, and is an interval in certain parameter region. Then, we consider the so called integrated density of states, and show that it is absolutely continuous with respect to Lebesgue measure. This is called scattered states, and physically, this means that the particle eventually runs out to infinity.

Motivated by the fact that the spectrum of the Labyrinth model is given by the products of two Cantor sets, we then consider this products of two Cantor sets more in detail. It is known that for any Cantor set, a real number called thickness can be assigned. In terms of this thickness, we get beautiful optimal estimates that guarantee that the product of two Cantor sets is an interval. Surprisingly, it turns out that if the thickness of two Cantor sets are the same, then the optimal value of the thickness is the golden mean, which is a value appears everywhere in nature, art, and architecture. For example, it is the ratio between a side and a diagonal of pentagon.

Lastly, we discuss the surprising connection between this problem and the “intersection of two Cantor sets” problem, which is a problem considered in many papers before.

Movement Anticipation and EEG: Implications for BCI-Contingent Robot Therapy

Sumner Norman
Mechanical and Aerospace Engineering

Brain-Computer interfacing is a technology that can potentially be used to improve patient effort in robot-assisted rehabilitation therapy, leading to increased motor outcomes after stroke. Past studies have used BCI as a control system where the patients?€? attempted movements are read at the cortical level, and translated into robotic orthosis movement. For example, movement intention reduces mu (8-13 Hz) and beta (13-35 Hz) wave oscillation amplitude over the sensorimotor cortex, a phenomenon referred to as event-related desynchronization (ERD). In what can be called an BCI-contingent assistance paradigm, initial studies have most commonly used ERD as the trigger signal for providing robotic assistance to limb movement. There are, however, a few previous reports of ERD occurring in response to externally imposed movements in which the subject remained passive. This introduces the possibility that the presence of ERD cannot be relied on to ensure active effort in movement training. Here we investigated how ERD changed as a function of audio-visual stimuli, overt movement from the participant, and robotic assistance. Eight unimpaired subjects played a musical computer game designed for rehabilitation therapy using the FINGER robotic exoskeleton. In the game, the participant and robot matched movement timing to audiovisual stimuli in the form of notes approaching a target on the screen set to the consistent beat of popular music. The audiovisual stimulation of the game alone did not cause ERD, before or after training. In contrast, overt movement by the subject caused ERD, whether or not the robot assisted the finger movement. Notably, ERD was also present when the subjects remained passive and the robot moved their fingers to play the game. This ERD occurred in anticipation of the passive finger movement with similar onset timing as for the overt movement conditions. These results demonstrate that ERD can be contingent on expectation of robotic assistance; that is, the brain generates an anticipatory ERD in expectation of a robot-imposed but predictable movement. Therefore, in a predictable therapy environment, the use of ERD as an orthosis control signal does not necessarily require the patient’s active engagement in the motor task, but simply in the expectation of robotic movement. As such, ERD is suboptimal in the context of patient engagement in BCI-contingent robot therapy, which may limit motor outcome after stroke. This is a caveat that should be considered in designing a BCI-robot therapy system for enhancing patient effort in robotically assisted therapy.

Optical trapping of dielectric nanoparticles by plasmonically enhanced evanescent wave

Qiancheng Zhao
Electrical Engineering
Co-authors: Rasul Torun, Shah Rahman, Tuva Cihangir Atasever, Ozdal Boyraz

We demonstrate the optical trapping of polystyrene nanoparticles by evanescent wave which is plasmonically enhanced by gold bowtie antennas. The light intensity between two antennas is 26 times larger than that of non-antenna enhancement case. The trapping force is boosted by 2 orders of magnitude with the help of plasmonic enhancement. Compared to gravity, optical force is 8 times larger. It shows a strong tendency that the nanoparticle tends to move to the highest light intensity spot, yielding a trapping phenomenon. Optical trapping starts from launching 785 nm light into a silicon nitride waveguide that is transparent to both infrared and visible region. TE mode is preferred because of its significant tangential electric field with respect to the top surface of the waveguide, which is capable of facilitating coupling mode into antennas. Due to discontinuity of electric field and magnetic field at the boundary, there is evanescent wave near the waveguide surface. However, the evanescent wave decays exponentially and thus cannot interact with nanoparticles effectively. To increase the trapping efficiency, a pair of 20 nm thick gold bowtie antennas is deposited on top of the silicon nitride waveguide. By manipulating the length of the antenna, resonance is achieved, and field enhancement occurs at the gap between two antennas. Although smaller gap leads to stronger field, the gap between two antennas is set to be 30 nm for the sake of fabrication convenience. A polystyrene nanoparticle with 20 nm diameter is placed in the gap of bowtie antennas. Due to light intensity gradient, the particle experiences different forces, which can be characterized by Maxwell Stress Tensor, at different locations. The net force upon the particle can be further calculated by integration over the particle surface. Simulation reveals that vertical optical force pulls the particle down towards the waveguide, and that transverse force pulls the particle to the center of the antenna gap. Therefore the nanoparticle is trapped into the gap between antennas. The simulation matches the theoretical expectation well, because high intensity region attracts optical denser medium but repels optical sparse medium. Compared to optical trapping by prism, waveguide structure with antenna enhancement is more compact and robust. The lab-on-chip system can be fabricated in a CMOS compatible way, rendering itself promising in low-cost production.