Applied Mechanics and Materials Vol. 598

Abstract: Offset in computer aided design and manufacturing (CAD/CAM) is a kind of important geometric operation, such as rapid prototyping technology, NC machining, the collision detection, etc. When offset distance d is greater than the minimum radius of curvature on the original surface, or part of the distance between parts of curved surface pieces, it causes self-intersection. Self-intersection makes distance between two models smaller than the offset distance r, leading to offset model local intersection or distortion. This paper proposes a method based on light projection algorithm to remove self-intersections and achieve good results. Firstly, set point coordinates on the original model as the light source locations and build the projection light according to adjacent plane vector weighted sum.Secondly, the projection light passes through the offset model. We extract the triangles on the outermost layer and load it into a list. Finally it is essential to filter the intersection points and construct triangle mesh. We use different discrete data models to do experiments and prove the validity and practicability of the algorithm.

Abstract: An improvement is made on the adaptive step size of the gradient Descent (GD). We propose that the improved GD can be used in Molecular Dynamics (MD) to determine the shortest distance between an irregular curve (described in high-order equation) and a point outside the curve, that we show that it can solve the problem with affordable computational power and high accuracy (an average of 1% error in around 30 iterations).

Abstract: Nowadays, there are several different kinds of methodology in selecting recommendation systems (CRS), and every method has its own evaluation criteria to pick up the best one. In this paper, a new MCDM method for recommendation system selection based on fuzzy VIKOR with multiple distances is introduced. It selects the best system by calculating values using three different distance calculation methods, which are Hamming distance, Euclidean distance and Hausdorff distance, and voting via Condorcet method. It minimizes the effect of distance and offers a more objective result than other methods and helps enterprises to select the most suitable recommendation system.

Abstract: In this paper, the eigenfunctions of Rayleigh wave in viscoelastic halfspace is studied. By comparing the eigenfunctions of Rayleigh wave in viscoelastic halfspace and the corresponding elastic halfspace, we found the penetration depth of Rayleigh wave in viscoelastic halfspace is always larger than in the corresponding elastic halfspace. But the difference is small. The ratio of penetration depth and wave length under different cases is calculated, the results show penetration ability of Rayleigh wave in viscoelastic halfspace is always stronger than the corresponding elastic case, and penetration ability of Rayleigh wave gets weaker when the Poisson's ratio is smaller. The penetration ability of Rayleigh wave is stronger when the Q value is smaller. The study in the paper is meaningful for the further research of Rayleigh wave in viscoelastic media.

Abstract: The paper describes the derivation of the numerical formulation of a second order time accurate and Geometrically Conservative Backward Difference Scheme (BDF) for transient flow simulation of Arbitrary Lagrangian Eulerian (ALE) problems using the control volume approach. The required modification to implement the scheme in an implicit adaptive flow solver is explained. The accuracy and robustness of the current formulation is demonstrated by simulating unsteady flow field over a sinusoidally pitching NACA0012 airfoil with larger allowable timestep in comparison to an existing Mid-point scheme.

Abstract: Tube hydroforming process is a well-established manufacturing process widely employed to form tubular parts that are lighter and stronger compared to those from stampings. Nevertheless, determination of process loading paths, i.e. axial feed distance versus hydraulic pressure, still typically relies on trial-and-error FEM approach. In this paper, a semi-forward adaptive simulation concept is proposed as an effective FEM approach, able to select a feasible THF loading path within a single FEM simulation run. The semi-forward adaptive simulation technique is based on the ability to “adapt” or adjust the loading path as to keep the forming strains within a preferred stain trajectory over the course of a simulation run. Forming strains at the current simulation time step are used as inputs to the fuzzy logic control; the output sets are then used to readjust the loading path for the current and next time steps. This semi-forward adaptive simulation scheme allows one to “correct” the loading path at the current time step as well as to better predict the forming strains in the next time step. It was found that the corrective and predictive nature of this semi-forward simulation approach coupled with the strain trajectory based fuzzy logic control scheme could handle a highly non-linear forming behavior of tube hydroforming processes effectively. In this work, a feasible loading path was determined thru only one simulation run for successful hydroforming of an eccentric bulged tubular part.

Abstract: In the present study CFD simulation with delayed detached eddy simulation (DDES) are performed to investigate an open cavity at Mach 0.85. Two cavity configurations, clean cavity and cavity with a leading-edge saw tooth spoiler, are modeled. The results obtained from clean cavity prediction are compared with experimental sound pressure level (SPL) data from QinetiQ, UK. Furthermore, comparisons are made with the predicted SPL between the two configurations. The main focuses of this investigation are to obtain a further understanding of the cavity aeroacoustics and test the noise suppression effect by a saw tooth spoiler.

Abstract: This paper presents the implementation of a fuzzy logic controller (FLC) for an FES-assisted elliptical stepping exercise in stroke rehabilitation. The purpose of the system is to restore the function of the lower extremities for the stroke subject through the application of a Functional Electrical Stimulation (FES) in specified exercise technique. The physiologically-based muscle model is developed to represent the quadriceps muscle of the paretic leg. The FES-induced movement at the paretic leg and voluntary movement at the non-paretic leg is coordinated using a control strategy to achieve a smooth exercise movement. Simulation result verifying the control strategy is presented and discussed.

Abstract: This paper describes the implementation of PID control strategy for a functional electrical stimulation (FES) in specific rehabilitation exercise known as elliptical stepping exercise. The PID controller is used to regulate the FES in controlling the knee joint torque of the paretic and non-paretic legs of the hemiplegic stroke patient during the exercise. The cadence speed error is used as the control variable and input to the PID controller. The Particle Swarm Optimization (PSO) technique is used to optimize the PID control parameters and the performance is compared to that of the manual tuning. The result shows that the system has good interaction with PID controller in completing the exercise. The result also shows that PSO is an efficient technique in optimizing the control parameters compared to manual tuning. Further investigation of the control and optimization technique will be considered in the future in order to improve the efficiency of the FES-assisted elliptical stepping exercise.

Abstract: Ensuring vehicle drive comfort and securing drive safety are the leading topics among the most interested issues for researchers in vehicle dynamics area. In this paper, a method utilizing a linear actuator is proposed for active control of the vehicle vibrations which are caused by road profile, intending to improve drive comfort and safety of road vehicles. The mathematical model belonging to the system that is evaluated as two degrees of freedom quarter car suspension system is derived by using Lagrange Equation of Motion and MATLAB/Simulink software. In addition to modeling technique, dynamic model of proposed system is created in MSC-ADAMS software and it is simulated in both Matlab and Adams programs together. Moreover two different controllers are designed, which are PID and Artificial Neural Network Based Fuzzy Logic (ANNFL) control in order to use in active vibration control simulations. Performances of the designed controllers are examined and the suitability of the designed controllers is studied by comparing their performances in case of using two different road profile functions.