Search

In this paper, an asymmetric breakup of non-Newtonian droplet (with power law behavior) in a new geometry (network junction) has been investigated. The geometry can break an initial droplet into six unequal size droplets. The research method is numerical simulation with Volume of Fluid (VOF) algorithm. The numerical results are compared with the results of a benchmark problem and a very good agreement is seen. The results showed that in areas close to the wall, mixing of materials of inside droplet is performed better, which is important in industrial applications of droplet based flows, especially in pharmaceutical and chemical industries. The results showed that the maximum vorticity magnitude in the K1 branch (the lowest output branch in the system) is 26, 44 and 28 % more than the maximum vorticity magnitude of the branches of K2, K3 and K4 (K4 is the highest output branch is in system). Also, maximum effective viscosity in the K1 branch is 27, 29 and 24 % less than the maximum effective viscosity in the K2, K3 and K4 branches, respectively. Therefore, K1 branch has the best performance in mixing of the material of inside droplet among the output branches. It was also revealed that the pressure of inside of droplet (both before and after breakup) is constant along the channel width [ABSTRACT FROM AUTHOR]

An alarm threshold plays an important role in an industrial fault detection system and directly contributes the False Alarm Rate (FAR) and Missed Alarm Rate (MAR). A crucial consideration for designing a threshold is estimating the Probability Density Function (PDF) of both normal and abnormal state based on recorded samples. The existence of measurement error in samples will be the contributors to an inaccurate estimation, following that, the alarm threshold will also be inaccurate. Therefore, grasping and recognizing measurement errors is highly important; in this paper, this problem will be investigated. For this purpose, firstly, a mathematical closedform of statistical parameters will be estimated, and, then, based on error propagation rule, the computation error of estimated parameters will be explored. It is assumed the high limit and low limit values of the measurement error are known or computable. Secondly, an approach is introduced to design a varying alarm threshold adapting to the current value of measurement error computing based on change detection algorithm. The proposed method is confirmed via a Monte Carlo simulation and it is finally applied to an industrial benchmark, Gas Turbine V94.2, experiencing fouling fault. [ABSTRACT FROM AUTHOR]

In this paper, the result of a numerical study on the natural convection in an inclined T shap cavity filled with Water-Cu nanofluid with the presence of a constant magnetic field was investigated. A heat source embedded on the bottom wall of enclosure, the upper wall is cold and the other walls are adiabatic. Discretization of the governing equations are achieved through a finite volume method and solved with SIMPLE algorithm. The Hartmann number has been varied from 0 to 80 and the cavity has been twisted under the angles between 0 to 90 degrees. The findings of study show that the effect magnetic field on the average Nusselt number is higher in high Reyleigh number. In Ra=105, the increase in nanofluid, to the Hartman number 20, contributes to decrease of the average number and in the Hartman number 40 and more, causes the average Nusselt number to increase. In Ra=106, the increase in nanofluid, to the Hartman number 20, contributes to increase of the average number and in the Hartman number 40 and more, causes the average Nusselt number to decrease. The results also indicate that, the maximum heat transfer, in Ra=105 and Ra=106 accurse at 67.5° angle. The minimum heat transfer, in Ra=105 and Ra=106 accurse at 0° and 22.5° angle respectively. [ABSTRACT FROM AUTHOR]

In this paper, the problem of the navigation error effect for the optimal and constraint Trajectory de of the UAVs that are required to fly at low altitude over terrains has been discussed. Due to the increasing deviation problem of inertial navigation systems in terms of time, having a safe flight and collision avoidance with terrain at low altitude is the main point in the trajectory design of this type of the vehicles. On the other hand, some of these vehicles use Terrain Contour Matching (TERCOM) as a navigation aiding system. This system is more efficient in rough terrains, and providing the requirements of this system beside other constraints is a complex task. In this paper is tried to meet these constraints in the trajectory design process. For this purpose, an algorithm based on the layered network flow on the digital terrain maps used in a manner that has a high potential in adoption of various constraints and optimal trajectory is generated. Then, using equations of motion on a terrain digital data in 3D space with the dynamical constraints and different optimality criteria, a complete model of navigation error and also parameters affecting TERCOM has been developed to generate feasible path reducing terrain collision probability to zero. The resulting algorithm applying initial conditions and in the least time possible, produces the trajectory. Numerical results show validity of this issue. [ABSTRACT FROM AUTHOR]

In this paper spacecraft formation flying control in leader -- follower structure is investigated. A non-linear PID controller is designed based on predictive control. The formation relative equation is obtained from nonlinear Hill's equation. The control input is achieved using predictive control algorithm. In controller design, disturbances have been replaced by disturbance observer. Equations are rewritten in the form of PID gains. Stability of the closed-loop system is proven by closed-loop error dynamics. Nonlinear PID controller performance in the pursuit of desired arrangement has been tested in simulations. Also, effects of various factors on the quality of response are studied. It is proven that predictive horizon time and disturbance observer gains have the most effect on system response. It is shown that if predictive time increases the settling time increases and control effort decreases if disturbance observer gain increases from a limit, it has no effect on settling time but control effort increases. The simulation shows the ability of this nonlinear control method in tracking. [ABSTRACT FROM AUTHOR]

Optimal trajectory planning is an important task which is required in most of guidance missions. This paper introduces a new method that utilizes the most important characteristics of global optimization methods along with a new gradient-based method in a two layered scheme for the trajectory planning. In the first layer to construct a convenient shooting method based algorithm, some of the most important global methods of optimization are used in an information transform structure. Exchanging the information between selected algorithms helps for increasing the efficiency of problem solving. To do this, a comprehensive model for parameterization of the control history is introduced which allows the method to search for the best profile in a variety of different profiles. Results of this layer are transformed to the second layer that uses one of direct methods of solving the optimal control problems. This gradient based method named Radau pseudospectral method using of the results of global methods, completes the optimization process. Finally, developed algorithm is used to find the optimal trajectory of a reentry capsule and effects of the path constraint values on the total heat absorbed is investigated. [ABSTRACT FROM AUTHOR]

In this paper, trajectory tracking control of a wheeled mobile robot is analyzed. Wheeled mobile robot is a nonlinear system. This system including three generalized coordinates (x, y, ∅), and a nonholonomic constraint. First, system kinematic and dynamic equations are obtained. A non-model-based control algorithm using PD-action filtered errors has been used in order to control the wheeled mobile robot. Non-model-based controllers are always more appropriate than model-based algorithms due to independency from dynamic models, lower computational costs and also robustness to uncertainties. Asymptotic stability of the closed loop system for trajectory tracking control of wheeled mobile robot has been investigated using appropriate Lyapunov function and also Barbalat's lemma method. Finally, in order to show the effectiveness of the proposed approach simulation and experimental results have been presented. Obtained results show that without requiring a priori knowledge of plant dynamics, and with reduced computational burden, the tracking performance of the presented algorithm is quite satisfactory. Therefore, the proposed control algorithm is well suited to most industrial applications where simple efficient algorithms are more appropriate than complicated theoretical ones with massive computational burden. [ABSTRACT FROM AUTHOR]

The need for complex surfaces in CAD motivates researchers for methods which can produce smooth and visually pleasing surfaces. In this research, a new method is presented for creating compatible cross-sectional curves for surface fitting to certain sections or lofting. In this method, the distribution of sections' data points along with basis knot vectors are improved in order to reach a desired smooth surface. In compatibility process, the section curves' degrees and their knot vectors must be set equal before implementing lofting process. Based on proposed algorithm, in this research, the constructed smooth and faired surfaces can be used in many engineering applications such as reverse engineering, biomedical engineering, quality control, etc. The main focus of the method is improvement of data points' distributions and their assigned parameters in a way that by a few iterations, data points' distribution are improved in order to reach a common knot vector for all cross-sectional curves. The method is implemented on some benchmarking examples and its efficiency are confirmed. In addition, the amount of final data points' deviation from the initial section curve is analyzed using the vigorous Hausdorff method. It is worth mentioning that the quality of obtained final surface is visually pleasing. In order to quantitatively confirm that the proposed method will result in smooth and fair surfaces, MVS is used. Finally the application of the method in modeling the rootjoint zone of a wind turbine blade is presented. [ABSTRACT FROM AUTHOR]