This paper intends to describe the process of derivation of loading surfaces with respect to phase transformation, when a structure is subjected to cyclic loading. This structure is realized as a Schwartz Primitive unit cell, and appropriate boundary conditions are applied to simulate the presence of adjacent unit cells. One hypothesis is that the curves in a stress space will shrink as more cycles are performed, due to a higher martensite volume fraction. This is a consequence of functional fatigue.
We analyzed the low Reynolds number (Re=100) flow generated behind a slitted 2D cross-sectional cylinder placed in a stream flow and oriented at various azimuth angles, to explore the possibility of passive control of the incompressible laminar boundary layer. Experiments were conducted using a 2D soap film tunnel that we have setup in our laboratory at Khalifa University. The visualization in soap film tunnel exploits the optical properties of soap film and relies on the wake formation patterns and the vortices shedding frequencies. These flow visualizations of the vortex shedding behind the cylinder with/without slit were recorded and analyzed and can be used to validate the 2D numerical simulations of similar flow. The developed soap-film technique will be attempted over more complex 2D shapes subjected to other controls for CFD validation. We will provide the details of the soap-film technique and guidelines for successful experimentation demonstrated on slitted cylinder.
The topic of Supernumerary Robotic Limbs (SRLs) has become of great interest in recent years. Researchers from all around the world have been involved in the analysis and development of the extra robotic limbs. This is of no surprise as SRLs can be used in many applications, especially in the industrial field where they can be used to reduce workers' fatigue and offer them help in different demanding tasks. However, there are many challenges in all the stages of this development process. Many factors should be considered and optimized such as weight, wearability, and Safety; which is the most critical factor. In this paper, the mechanical design and workspace analysis of an extra robotic arm will be presented and discussed. Also, future work will be included at the end.
To reduce equipment size and enhance mass transfer, Process Intensification (PI) technologies are extensively used. Mass transfer plays an important role in heterogeneous reactor systems where chemical species must interact with the catalytic surface. As part of this study, a heterogeneous reactor system, the Rotating Bed Reactor (RBR), is modeled using computational fluid dynamic (CFD) tool ANSYS. Two different geometries are used to investigate the flow characteristics, i.e., a detailed geometrical model involving a quarter (90o segment) of the reactor and a simplified geometrical model employing a 22.5o segment of the whole reactor. The CFD models are validated using a previous study. It is found that the pressure and velocity profiles computed using a simplified geometrical model are in close agreement with the model based on detailed geometry. Further, it is shown that the simplified model required less computational power and can be used for further studies.
Firstly, we present a design and protocol to add binary numbers using discrete solitons in waveguide arrays. We show that the nonlinear interaction between discrete solitons in waveguide arrays can be exploited to design half and full adders. Secondly, we present a design and protocol to achieve an essential feature of an optical transistor, namely the amplification of input signal with the use of discrete solitons in waveguide arrays. We consider the scattering of a discrete soliton by a reflectionless potential in the presence of a control soliton. We show that at the sharp transition region between full reflectance and full transmittance, the intensity of the reflected or transmitted soliton is highly sensitive to the intensity of the control soliton. This suggests a setup of signal amplifier. The suggested devices will be very important components in the all-optical data processing.
It is attractive to learn physics via machine learning because physics describes our complicated real-world both elegantly and economically, with simple laws of physics to govern the evolution of complex states. In the case of classical mechanics, nature favors the object to move along the path according to the time integral of the Lagrangian, called the action S. We consider setting the reward/penalty as a function of S, so the agent could learn the physical trajectory of particles in various kinds of environments with reinforcement learning (RL). In this work, we verified the idea by using a Q-Learning based algorithm on learning how light propagates in materials with different refraction indices, and show that the agent could recover the minimal-time path equivalent to the solution obtained by Snell's law or Fermat's Principle. The success sheds light on the possibility of further applications for combining RL and physics.
Thin films of TiO2 were deposited on FTO by spray pyrolysis technique. The chosen precursor is titanium disopropoxide stabilized with acetyl acetone in ethanol solvent. The X-ray investigation revealed polycrystalline nature of film with single anatase phase. Scanning Electron Microscopy assured the uniform deposition of films on FTO substrate. The estimated optical band gap was found to be 3.4eV that is in accordance with the literature for thin film anatase TiO2 . Since, the spray pyrolysis is a versatile technique with interdependence of precursor chemistry for effective modulation of electron transport features of TiO2 films in enhancing the performance of perovskite solar cells. Therefore, single-phase, uniform, and transparent TiO2 films have the potential to be utilized as an efficient electron transport layer for enhanced efficiency of Perovskite Solar Cells. Moreover, the sprayed titania films will also be tested for photocatalytic degradation activity of organic dyes.
Ultracold molecular physics is a promising research field with great potential to revolutionize various important applications such as quantum information. One of the methods that can be used to produce ultracold molecules is Laser cooling. This work investigates diatomic molecules that can be cooled through Doppler laser cooling, using quantum computational calculations: ab initio complete active space self-consistent field (CASSCF), (MRCI+Q) calculations. The molecular candidates of interest in this work are PH and LuF. We confirm that PH is feasible for laser cooling for A3? - X3?- transition (the main cooling cycle) with radiative lifetime of 192.8 ns. We also find that LuF molecule can be feasible for laser cooling for the main cooling cycle between (1)1? - X1?+ states having a radiative lifetime of 9.22 ns
We propose a high accuracy power series method for solving partial differential equations with emphasis on the fundamental nonlinear Schr?dinger equation. The method is based on employing an iterative power series for time stepping and a multi-point formula for the spacial discretization of the second derivative. The accuracy and calculation speed can be arbitrarily increased to orders of magnitude larger than those of other methods and easily reaching machine precision. Two parameters are characterizing the presented method accuracy, the maximum power of the time power series and the number of points in the multi-point formula. The present method successfully captures the exact moving bright soliton of the nonlinear Schr?dinger equation.
With the prevalence of big data, feature selection has become a necessary preprocessing step across many data mining applications. In this work, a general framework for the ensemble of multiple feature selection methods was implemented. Based on diversified datasets generated from the original set of observations, the importance scores generated by multiple feature selection techniques were aggregated using two methods: Within Aggregation Method (WAM) which refers to aggregating importance scores within a single feature selection, and Between Aggregation Method (BAM) which refers to aggregating importance scores between multiple feature selection methods. The experimental evaluation of 13 real datasets shows that WAM provides an effective tool for determining the best feature selection method for a given dataset. Both method exhibit comparable computational demands, though WAM shows greater stability. The results of this work suggest that by applying both WAM and BAM, practitioners can gain deeper understanding of the feature selection process.
