Development of Airfoils Based on Their Aerodynamic Characteristics Using Artificial Neural Networks

One of the main concerns in Engineering nowadays is the development of aircrafts of low consumption and high performance. For this purpose, airfoils are studied and designed to have an elevated lift coefficient and a low drag coefficient, thus generating a highly efficient airfoil. The higher the efficiency value is, the lower the aircraft fuel consumption will be; thus improving its performance. In this sense, this work aims to develop a tool for airfoil creation from some desired characteristics, such as the lift and drag coefficients and maximum efficiency rate, using an algorithm based on an ANN （artificial neural network）. In order to do so, a database of aerodynamic characteristics with a total of 300 airfoils was initially collected from the XFoil software. Then, through a routine implemented in the MATLAB software, network architectures of one, two, three and four modules were trained, using the back propagation algorithm and momentum. The cross-validation technique was applied to analyze the results, evaluating which network possesses the lowest value in RMS （root-mean-square） error. In this case, the best result obtained was from the two-module architecture with two hidden neuron layers. The airfoils developed by this network, in the regions with the lowest RMS, were compared to the same airfoils imported to XFoil. The presented work offers as a contribution, in relation to other works involving ANN applied to fluid mechanics, the development of airfoils from their aerodynamic characteristics.     

Design,Fabrication and Experimentation of a Deep Drawing Machine

This paper presents the work implemented in designing, fabricating and operating a model of a cheap hydraulic DDM （deep drawing machine）, which is currently utilized in the manufacturing processes lab in the IED （Industrial Engineering Department） at An-Najah National University. The machine is used to conduct different experiments related to the deep drawing process. This work was implemented in three stages： the first was the design stage, in which all design calculations of the DDM elements were completed based on the specifications of the product （cup） to be drawn; the second was the construction stage, in which the DDM elements were fabricated and assembled at the engineering workshops of the university; the last was the operating and experimentation stage, in which the DDM was tested by conducting different experiments. The experience gained from designing and constructing such a mechanical lab equipment was found to be successful in terms of obtaining practical results that agree with those available in literature, cost-effective relative to the cost of a similar purchased equipment, as well as enhancing students＇ abilities in understanding the deep drawing process in particular and machine elements design concepts in general.     

Vibration Analysis of Vehicle Floor Panel Using Hybrid Method of FEM and SEA

In order to improve the mass efficiency of an automotive soundproof package, it is important to predict the middle to high frequency range of noise and vibration during vehicle operation. A hybrid method of experimental and analytical SEA （statistical energy analysis） has been applied for the prediction of air-borne noise. However, for predicting structure-borne noise, there are no definitive simulation methods that can address the soundproof specifications in an actual vehicle. Thus, in this paper, a FEM （finite element method）＇SEA hybrid method is used. The FEM＇SEA hybrid method predicts structure-borne noise in the middle to high frequency range. First, we explain the basic concept of the FEM＇SEA hybrid method; Second, we describe our experiment to verify the analytical results of the FEM＇SEA hybrid method; Third, we provide the details of the FEM model versus the FEM＇SEA hybrid model; Finally, we verify the validity and availability of the FEM＇SEA hybrid method through comparisons of the FEM analysis results, FEM-SEA analysis results and measured results.     

Development and Impact Behaviors of FRP Guarder Belt for Side Collision of Automobiles

In automobiles, the CFRP （carbon fiber reinforced plastics） has a possibility of weight reduction in automotive structures which can contribute to improve mileage and then reduce carbon dioxide. On the other hand, the safety of collision should be also made clear in the case of employing the CFRP to automotive structures. In this paper, the CFRP guarder belt equipped in the automotive door is developed and examined by an experiment and a numerical analysis for replacing the conventional steel door guarder beam. As the experimental relation of impact load to displacement for CFRP guarder belt agreed well with that of numerical result, the numerical method developed here is quite useful for estimating impact behaviors of CFRP guarder belt.     

Evaluation of Adhesive Joints for Combined Load Test-Fixtures

Abstract： In a test-fixture that the authors were using, steel tabs adhesively bonded to an aluminum panel debonded before the design load on the real test panel was fully applied. Therefore, studying behavior of adhesive joints for joining dissimilar materials was deemed to be necessary. To determine the failure load responsible for debonding of adhesive joints of two dissimilar materials, stress distributions in adhesive joints as obtained by a nonlinear finite element model of the test-fixture were studied under a gradually increasing compression-shear load. It was observed that in-plane stresses were responsible for the debonding of the steel tabs. To achieve a better understanding of adhesive joints of dissimilar materials, finite element models of adhesive lap joints and ADCB （asymmetric double cantilever beam） were studied, under loadings similar to the loading faced by the test-fixture. The analysis was performed using ABAQUS, a commercially available software, and the cohesive zone modeling was used to study the debonding growth.     

Analysis of Unbalances Forces Using Methods of Identification and Finite Elements

The study of damage in rotating machineries is of fundamental interest in the fields of machine and structure design. A rotating system, supported by bearings and under some dynamic conditions, can generate a variety of problems that are encountered in many different types of rotating machines. One of these problems is the unbalance due to non-homogeneous mass distribution along the shaft. One of the techniques which are widespread today is the identification of parameters and excitation forces that may well followed by monitoring the evolution and change of possible variations of these parameters. Although several methods for the identification of unbalance excitation force are available in the literature, none of them can be considered unrestricted to be applied for all rotating systems. In this study, two methodologies to identify unknown excitations, such as unbalance, have been proposed. This project refers to the analysis of unbalanced forces from displacement parameters and speed by using methods of identification by Fourier series and Legendre polynomials together with the finite element method, state observers in reasons of the problem of absence of signs of rotational displacement, bandpass filter were used to noise suppression of the data collected from the experimental part, Quasi-Newton method to minimize a function in which the bearing stiffness and its damping are unknowns, and also the experimental verification of the methodology, using for this system owned by a rotary mechanical vibrations of the Department of Mechanical Engineering of Faculty of Engineering, campus of llha Solteira.     

The Topic of Contact Zone Problems in the Shaft and Hole Joint, Taking into Account Component Form Errors

In assemblies constructed from components manufactured with radial deviations, cross-section deviations and deviations being combination of both, there occur variable values of local stresses and displacements. Both the types of shape deviations and their values need to be taken into account in the designing process and play an important role during machine operation. They have a crucial effect on the value and scatter of maximum reduced von Mises stresses and contact stresses. Axisymmetric joints were examined, in which shafts in selected shape variants and in variable angular positions were associated with a non-deformable hole. The aspects of contact zone problems are presented using the example of numerical simulation of contact between an elliptical saddle-shaped shaft placed in a rigid, non-deformable hole in different angular positions. Occurrence of both variable relative stresses and contact stresses as well as shaft＇s axial shift and rotary movement resistance were demonstrated.     

Finite Element Analysis of the Dynamic Behaviour of Transmission Line Conductors Using MATLAB

The dynamic behaviour of power line cables have been a source of interest to researchers ever since the phenomenon was first noticed in the 1920s. Conductor oscillation is mostly caused by the dynamic forces of nature such as wind loading. This imposes a periodic force on the conductors which is highly undesirable. It is therefore important for engineers to account for the possible effect of the wind loading when designing the power line. Investigations have shown that modeling the exact dynamic behaviour of a conductor is very difficult. Based on this fact, getting the exact analytical solution to conductor vibration is difficult, which is almost impossible, hence the numerical approximation becomes an option. This paper presents the developed finite element method used to analyse the dynamic behaviour of transmission line conductors. The developed FEM （finite element method） is implemented on MATLAB. The numerical analysis using MATLAB that is presented in this paper is used to simulate the response of the conductor when subjected to external loading in the time domain. The simulation is used to analyse the transverse vibration of the conductor. The formulation of the stiffness matrix and load vector is done and the results obtained are used to evaluate the conductor＇s internal energy dissipation. This finite element solution is compared with the results documented in literature. This numerical simulation is also used to investigate the effects of varying the axial tension on energy dissipation within the strands. Hence, this evolved in physically appropriate energy characterization process that can be used to evaluate the conductor self-damping with respect to line contact.     

Influence of the Number of Rollers on a Tapered Roller Bearing

In roller bearings, the outer ring is usually fixed and the inner ring has the rolling motion. Concerning TRB （tapered roller bearings）, this motion generates forces that are transmitted to the outer ring by the tapered rollers. Thus, contact stresses occur and the number of rollers plays a major role with respect to the load distribution. This influence is analyzed in this study by the FEM （finite element method） with commercial code ABAQUS, where two models were evaluated： a common TRB and the same one but with fewer rollers. As an application, a manual automotive transmission was considered for the input loads.     

Impact Dynamics on Granular Plate

In this paper, a theoretical and numerical study on the impact of a rubber solid on the free surface of a granular plate is presented, showing a simulation of an aircraft wheel on impact with a flexible landing surface. This physical action, when we use a theological approach, becomes a fundamental parameter to investigate wear and tear, and consequently strength to micro and macro pavements failure. The study has developed initially from a microscopic point of view and subsequently on macroscale. The effects are strictly linked with material degradation associated with damage evolution. The problem is developed by energetic approach on an elastic-plastic element using the functional energy containing two contributions, bulk and surface. The model simulates the behaviour of flexible runway pavements during the landing phase.     

基于ANN和GA的机床结构优化设计

机床主轴是机床中的重要部件,其设计对机床性能影响很大。在对主轴设计的各项要求中,刚度和振动稳定性是主要考虑的指标,将其归纳在一个含区域和性能约束的主轴结构优化设计数学模型中。借助有限元软件可获得不同主轴尺寸下的各项性能指标,利用人工神经网络(ANN)实现主轴各项性能指标的近似分析,并采用遗传算法(GA)实现了性能指标为设计变量隐式表达下的机床主轴结构优化设计。     

Determination of Natural Frequencies of Multilayered Plates by Mixed Finite Elements

Natural frequencies for multilayer plates are calculated by mixed finite element method. The main object of this paper is to use the mixed model for multilayer plates, analyzing each layer as an isolated plate, where the continuity of displacements is achieved by Lagrange multipliers （representing static variables）. This procedure allows us to work with any model for single plate （so as to ensure the proper behavior of each layer）, and the complexity of the multilayer system is avoided by ensuring the condition of displacements by the Lagrange multipliers （static variables）. The plate is discretized by finite element modeling based on a primary hybrid model, where the domain is divided by quadrilateral, both for the displacement field and static variables. This mixed element for plates was implemented and several examples of vibrations have been verified successfully by the results obtained by other methods in the literature.     

Failure Analysis in Graphite Epoxy Composite Laminates

The present analysis was performed to obtain bearing strength for pinned joints in uni-directional graphite epoxy composite laminates using characteristic curve model. The characteristic dimensions used to determine the characteristic curve were evaluated using a two-dimensional finite element model that was developed in ANSYS14.5 Software. Also, two-dimensional finite element stress analysis was developed to determine the stress distribution needed to evaluate the failure. Tsai-Wu failure criterion was used in the analysis with the characteristic curve to predict bearing strength. The results of the analysis showed good agreement with experimental data.     

基于神经网络的多点柔性夹具横梁结构优化

研究了神经网络在多点柔性夹具横梁结构优化中的应用。横梁结构与最大变形量间为复杂的非线性关系或隐函数关系,且单一的有限元分析工作复杂,计算耗时大。在有限元与数学规划法基础上利用人工神经网络进行横梁结构分析,并通过测试比较从BP、Elman和RBF神经网络中选取性能较好的神经网络,实现了横梁质量的最优化,节省了计算时间。     

基于遗传神经网络的压铸机座板结构优化设计

在对压铸机合模机构进行结构设计时,利用神经网络的非线性映射能力,通过少量样本的有限元分析结果,训练出表述结构参数间函数关系的神经网络模型,然后利用遗传算法的全局寻优性找到神经网络模型表述的目标函数的最优结构参数,从而解决结构优化设计的瓶颈和智能问题,利用这种优化设计策略,设计了压铸机合模机构座板,结果表明了该方法的高效性。     

基于人工神经网络的某空间反射镜柔性支撑结构优化设计与分析

空间反射镜的支撑结构决定了反射镜的性能。通过有限元分析来优化支撑结构参数需要很大计算量,而且无法找到最优解。为了缩短设计周期,提出用人工神经网络来模拟反射镜组件输出特性和支撑结构参数间的非线性关系,并用MATLAB编制程序实现神经网络的建立和泛化,最终找到使反射镜输出特性最优的支撑结构参数。该方法不但能找到多变量优化的最优解,而且计算表明用神经网络泛化得到结果和有限元分析计算得到的结果相差在5%以内,精确度足以满足工程应用要求。     

计算机模拟与预测方法在材料科学研究中的应用

计算机模拟与预测方法在材料科学研究中得到愈来愈广泛的应用。针对材料研究的发展趋势,介绍了常用的有限元法、有限差分法、蒙特卡罗方法、分子动力学方法及人工神经网络方法等的特点,并列举了它们在材料研究开发中的应用。     

Application of a neural network to predict thickness strains and finite element simulation of hydro-mechanical deep drawing

In this paper an artificial neural network (ANN) is used to predict the thickness along a cup wall in hydro-mechanical deep drawing. This model uses a feed-forward back-prorogation neural network. After using the experimental results to train and test the network, the model was applied to new data for the prediction of thickness strains in hydro-mechanical deep drawing. The results are promising. In the present work, we also attempt to perform a finite element simulation of the process for the two dimensional axi-symmetric case using explicit finite element code LS-DYNA 2D. Counter pressure on the blank is applied by specifying the pressure boundary conditions. A comparison was made between simulated, experimental and ANN results of hydro-mechanical deep drawing using low carbon extra deep drawing grade steel sheets of 0.96 mm thickness. It was also found that by hydro-mechanical deep drawing, a higher drawability and a more uniform thickness distribution were obtained when compared to conventional deep drawing.     

Simultaneous Multi-vehicle Detection and Tracking Framework with Pavement Constraints Based on Machine Learning and Particle Filter Algorithm

Due to the large variations of environment with ever-changing background and vehicles with different shapes, colors and appearances, to implement a real-time on-board vehicle recognition system with high adaptability, efficiency and robustness in complicated environments, remains challenging. This paper introduces a simultaneous detection and tracking framework for robust on-board vehicle recognition based on monocular vision technology. The framework utilizes a novel layered machine learning and particle filter to build a multi-vehicle detection and tracking system. In the vehicle detection stage, a layered machine learning method is presented, which combines coarse-search and fine-search to obtain the target using the AdaBoost-based training algorithm. The pavement segmentation method based on characteristic similarity is proposed to estimate the most likely pavement area. Efficiency and accuracy are enhanced by restricting vehicle detection within the downsized area of pavement. In vehicle tracking stage, a multi-objective tracking algorithm based on target state management and particle filter is proposed. The proposed system is evaluated by roadway video captured in a variety of traffics, illumination, and weather conditions. The evaluating results show that, under conditions of proper illumination and clear vehicle appearance, the proposed system achieves 91.2% detection rate and 2.6% false detection rate. Experiments compared to typical algorithms show that, the presented algorithm reduces the false detection rate nearly by half at the cost of decreasing 2.7%–8.6% detection rate. This paper proposes a multi-vehicle detection and tracking system, which is promising for implementation in an on-board vehicle recognition system with high precision, strong robustness and low computational cost.