Thien-My Dao

Numerical study of non-kinematical conical bending with cylindrical rolls

Abstract Although desired cones can be manufactured by kinematical conical roll bending, less manufacturing cost is still largely required. This paper presents the modelling and simulation of non-kinematical conical roll bending process with cylindrical rolls. Such process is achieved by using attachments to reduce the velocity on the area close to the top edge of the plate. In contrast with a kinematical conical roll bending machine, the driving outer rolls of a non-kinematical conical roll bending machine are allowed to slide on the plate close to the top edge. The modelling is based on finite element method under ANSYS/LS-DYNA environment. The bent cones obtained from numerical simulations compare well with the desired cones. Numerical simulation investigations show that the adaptive capacity of the model is available within a range of the top–bottom radius ratio of the desired cones between 0.44 and 0.8.

New Methodology for Wind Tunnel Calibration Using Neural Networks - EGD Approach

One of the hardest tasks involving wind tunnel characterization is to determine the air-flow condition inside the test section. The Log-Tchebycheff method and the Equal Area method allow calculation of local velocities from measured differential pressures on rectangular and circular ducts. However, these two standard methods for air flow measurement are limited by the number of accurate pressure readings by the Pitot tube. In this paper, a new approach is presented for wind tunnel calibrations. This approach is based on a limited number of dynamic pressure measurements and a predictive technique using Neural Network (NN). To optimize the NN, the extended great deluge (EGD) algorithm is used. Wind tunnel testing involves a large number of variables such as wind direction, velocity, rate flow, turbulence characteristics, temperature variation and pressure distribution on airfoils. NN has the advantage that multilayer perceptron neural networks can describe a 3D flow area with a small amount of experimental data, fewer numbers of iterations and less computation time per iteration. The Fluent results are used to train and optimize the proposed NN approach. The validation of this new approach is achieved by experimental tests using the wind tunnel Price-Paidoussis of LARCASE laboratory. This wind tunnel has two test chambers. The main chamber with a section Height, Width and Length equal to 0.62 x 0.91 x 1.83 meter respectively that provides a speed ranging from 0 to 30 m/s and a second chamber test with a section Height, Width and Length equal to 0.31 x 0.61 x 1.22 meters respectively that provides a speed from 0 to 60 m/s.

Optimization of cellular manufacturing systems design using the hybrid approach based on the ant colony and tabu search techniques

Cellular systems design problems (CSDP) constitute an important issue in the design of cellular manufacturing systems (CMS). A few years back, it emerged as the best alternative in manufacturing systems, representing a compromise between the efficiency of serial and the flexibility of batch production systems. In this paper, we propose a hybrid approach for solving the CSDP for large industrial data sets. This procedure comprises an ant colony optimization (ACO) and the Tabu search (TS) procedure, which is added in order to improve the quality of the ACO solutions obtained. The problem is formulated as a binary integer programming model that might minimize the dissimilarities existing between machines or parts, and that is characterized as an NP-complete model. With this proposed approach, the results obtained show that it is efficient in terms of the quality and computational time of the solutions. To demonstrate the potential ability of the proposed approach, a numerical example has been investigated.

Optimimization of group scheduling using simulation with the meta-heuristic Extended Great Deluge (EGD) approach

Many companies apply cellular manufacturing systems (CMS) in order to improve production. One of the most significant problems encountered in production management is the scheduling problem, which has also been proven to be NP-hard. The objectives of the group scheduling problem in manufacturing are considered in order to minimize the makespan, the total flowtime and machine idletime. In this paper, we propose an approach for optimizing the scheduling of the manufacturing tasks of all parts of a product family, including exceptional elements. To solve this problem, an Extended Great Deluge (EGD) approach algorithm is applied in order to determine the optimal sequence of parts in each cell, minimizing the makespan and the total flowtime; following that, a heuristic method is applied to introduce exceptional elements. The results of the proposed hybrid approach show a major improvement when compared with those obtained using one of the best algorithms that has so far been presented by other researchers.

An integrated computer-aided optimization system for process planning

Abstract Designing a manufacturing operation is a major activity of process planning, especially in the case of processes requiring a comprehensive selection of the operating parameters. This activity includes three hierarchical stages : procedure design, schedule design, and tolerance design. A new approach for process modelling and optimization has been devised to address the problems to solve at each stage. The methodology is based on sequential experimentation to model the feasible domain of the process and on a direct search technique to optimize for quality and process robustness. An integrated computer-aided system has been designed to manage the manufacturing data and the strategies for optimization and variation reduction. Applications to various manufacturing processes are presented.

New methodology for calculating flight parameters with neural network - Extended Great Deluge method applied on a reduced scale wind tunnel model of an ATR-42 wing

The determination of flight parameters such as pressure distributions and aerodynamic coefficients (lift, drag and moment) from the known parameters (angle of attack, Mach number ...) in real time is still not achievable easily by methods of numerical analysis in aerodynamics and aeroelasticity domains. For this reason, we propose a flight parameters control system. This approach is based on new optimization methodologies with neural networks (NN) and extended great deluge (EGD). The validation of this new method is realized by experimental tests using a model installed in the wind tunnel to determine the pressure distribution. For lift, drag and moment coefficient, the results of our approach are compared to the XFoil results for different angles of attack. The main purpose of this control system is to improve the aircraft aerodynamic performance.

On the Use of Theory of Rings on Nonlinear Elastic Foundation to Study the Effect of Bolt Spacing in Bolted Flange Joints

Bolted flange joints are extensively used to connect pressure vessels and piping equipment together. They are simple structures that offer the possibility of disassembly. However, they often experience leakage problems due to a loss of tightness as a result of a nonuniform distribution of gasket contact stresses in the radial and circumferential direction. Many factors contribute to such a failure; the flange and gasket stiffness and bolt spacing design combination being one of them. In our recent paper, the effects of bolt spacing were investigated based on the theory of circular beams resting on a linear elastic foundation (Do, T. D., Bouzid, A. H., and Dao, T.-M., 2011, “Effect of Bolt Spacing on the Circumferential Distribution of Gasket Contact Stress in Bolted Flange Joints,” ASME J. Pressure Vessel Technol., 133 (4), 041205). This paper is an extension of the work in which an analytical solution based on the real nonlinear gasket behavior is developed. This study focuses on the distribution of the gasket contact stress of two large diameter flanges, namely, a 52 in. and a 120 in. heat exchanger (HE) flanges. The nonlinear gasket behavior solution is compared to the Finite Element Analysis (FEA) and the linear gasket behavior solution for evaluation and comparison.

Optimisation of manufacturing cell formation with extended great deluge meta-heuristic approach

The concepts of cellular manufacturing system (CMS) and cell scheduling (CS) have been widely used to meet various production needs. The CMS is a particular case of group technology (GT) applied to improve the production efficiency and reduce operational costs. This work addresses the machine/part grouping and group scheduling problems. The cell formation problem has long been recognised as the most challenging problem in realising the concept of cellular manufacturing. It belongs to the class of NP-hard problems. One of the most important problems in the area of production management is the scheduling problem which has also been proven to be NPhard. To solve this scheduling problem an Extended Great Deluge (EGD) metaheuristic approach is employed. The results of the proposed approach show a major improvement when compared with the results of one of the best algorithms developed so far by other researchers.

A neural network controller new methodology for the ATR-42 morphing wing actuation

A morphing wing model is used to improve aircraft performance. To obtain the desired airfoils, electrical actuators are used, which are installed inside of the wing to morph its upper surface in order to obtain its desired shape. In order to achieve this objective, a robust position controller is needed. In this research, a design and test validation of a controller based on neural networks is presented. This controller was composed by a position controller and a current controller to manage the current consumed by the electrical actuators to obtain its desired displacement. The model was tested and validated using simulation and experimental tests. The results obtained with the proposed controller were compared to the results given by the PID controller. Wind tunnel tests were conducted in the Price-Paidoussis Wind Tunnel at the LARCASE laboratory in order to calculate the pressure coefficient distribution on an ATR-42 morphing wing model for different flow conditions. The pressure coefficients obtained experimentally were compared with their numerical values given by XFoil software.

Novel approach to optimize milk-run delivery: A case study

This article proposes a new approach to optimize a milk-run delivery in supply chain (SC) which was introduced by Zhou and Kelin (2011). These authors particularly formulated the mathematical model of SC total cost at operational level for milk-run delivery and used Ant Colony Optimization (ACO) to find out the results in one specific case study. Yet, the solution found by this Meta- Heuristics did not obtain optimal values. Hence, to improve the solution, the paper presents one new method namely Hybrid Ant Colony Optimization and Tabu Search (HAT). To qualify new method, Tabu Search (TS) is taken into account in testing stage. With the same data in the case study and random data, the new method outperforms the original and TS regarding SC total cost and milk-run distance.