Ahmad Smaili

Optimum Synthesis of Mechanisms Using Tabu-Gradient Search Algorithm

A tabu-gradient search is herein presented for optimum synthesis of planar mechanisms. The solution generated by a recency-based, short term memory tabu search is used to start a gradient search to drive the solution ever closer to the global minimum. A brief overview of the tabu-search method is first presented. A tabu-gradient algorithm is then used to synthesize four-bar mechanisms for path generation tasks by way of three examples, including two benchmark examples used before to test other deterministic and intelligent optimization schemes. Compared with the corresponding results generated by other schemes, the tabu-gradient search rendered the most optimal solutions of all.

Experimental and Theoretical Behavior of Self-healing Bolted Joints

This article focuses on improving the experimental behavior of a self-healing bolted joint using shape memory alloys (SMAs). Self-healing bolted joints are used to reduce the likelihood of failure of structures due to loosening of bolted joints. The clamping force (CF) in a bolted joint assembly plays an important role in maintaining the integrity of the structure. Controlling the preload in a bolted joint assembly by external heating activates the heat-to-recover (HTR) SMA washer. The SMA washer is initially in the martensitic state. When the bolt is loosened to a certain amount of preload, a heater enveloping the ring is activated allowing an axial constrained recovery of the SMA and a control of the preload in the bolt. Several experiments are conducted to ascertain the effect of heating rate, cooling rate, relaxation of assembly, and initial preload, on the preload-temperature relationship. Experimental results indicate that the use of SMA rings is promising for retightening and controlling preload in bolted joint assemblies. Two analytical models are established; one is based on the Liang-Rogers model (Liang, C. and Rogers, C.A. 1990. “One-dimensional Thermo-mechanical Constitutive Relations for Shape Memory Materials,” Journal of Intelligent Materials Systems and Structures, 1:207-234) and the other draws on the analogy between the cyclic stress-strain curves.

A NEW APPROACH TO SHAPE OPTIMIZATION FOR CLOSED PATH SYNTHESIS OF PLANAR MECHANISMS

ABSTRACT A method for the synthesis of four-bar mechanisms to generate closed paths through shape optimization is herein introduced. The objective function is not based on Fourier descriptors, rather on the cyclic angular deviation (CAD) vector associated with a set of desired points on the curve. A simple method is introduced to account for the starting point shift between the desired CAD and the generated one. Following shape optimization, a simple mathematical approach is devised to properly scale, rotate, and translate the mechanism to the desired configuration. A case study is presented to demonstrate the effectiveness and robustness of the proposed method. KEYWORDS Synthesis; Shape Optimization, CAD vector; Optimization; Ant Colony; 1. INTRODUCTION Methods of dimensional synthesis of planar mechanisms have evolved with time to encompass a wide range of possibilities. Precision point synthesis methods are adequate for a limited number of points [1]. If the number of desired points exceeds a few (5 pos for four-bar motion generator), optimum synthesis becomes necessary [2-23]. A hybrid of exact/approximate point synthesis schemes have also been suggested [24-28]. Exact/approximate synthesis results in a mechanism that satisfies the desired trajectory at two or three points precisely, while generating many other points approximately. The main focus of previous synthesis efforts have been on single-task applications - motion, function, or path, synthesis. A method for a hybrid task synthesis was recently introduced [29]. Mechanism synthesis involves a large number of design parameters and searching the entire design space (DS) would be expensive. One method that has been suggested to reduce the size of the DS is to optimally find a solution that satisfies the shape of the desired trajectory. Then, the final solution is obtained by scaling, rotating, and translating the solution for the final mechanism. This paper follows this approach, but in a new way. The design objective is usually to determine the optimal link lengths (and angles between them) such that the motion generated by the mechanism is as close as possible to the desired trajectory. For the discussion to follow, the parameters of the four-bar mechanism are first introduced (see Fig. 1). The following notations are adopted hereafter. An uppercase boldface letter indicates a vector representing a link; the lowercase counterpart denotes the corresponding length; and the Greek letter counterpart denotes its angle. Thus,

Design for Cultural Difference

Design, the cross-fertilization of science and art, is a basic function of all species that occupy a masterfully orchestrated and designed ecosystem in which man is but one. On the other hand, culture with its complex mix is the expression of what a group of people creates — arts, beliefs customs, institutions, products and thought — at a particular time within the context of the natural environment. Design and culture therefore are intimately linked and undoubtedly influence each other. This suggests that designers, with their problem solving skills and keen interest to preserve nature and advance quality of life are capable of reshaping culture in a positive way. This paper is not intended to provide specific answers on how to achieve that, but it highlights some aspects of the design-culture interface and asserts that designers, armed with good will and respect for all have under their disposal a strong force by which they can help fashion a peaceful world. The paper also addresses possible things designers can do to influence that objective.Copyright

A New Approach for Exact/Approximate Point Synthesis of Planar Mechanisms

The aim of this article is to provide a simple method to solve the mixed exact-approximate dimensional synthesis problem of planar mechanism. The method results in a mechanism that can traverse a closed path with the choice of any number of exact points while the rest are approximate points. The algorithm is based on optimum synthesis rather than on precision position methods. Ant-gradient search is applied on an objective function based on log10 of the error between the desired positions and those generated by the optimum solution. The log10 function discriminates on the side of generating miniscule errors (on the order of 10−14 ) at the exact points while allowing for higher errors at the approximate positions. The algorithm is tested by way of five examples. One of these examples was used to test exact/approximate synthesis method based on precision point synthesis approach.Copyright

An Elitist Ants-Search Based Method for Optimum Synthesis of Rigid-Linkage Mechanisms

Mechanical linkages are widely used in the industry and the synthesis of such mechanisms may require optimization depending on the number of precision positions required. Many intelligent optimization techniques (Genetic, Tabu, Simulated Annealing, etc) have been proposed in the literature, one of them being the Ant-Search which was first proposed by the authors in 2007. In this paper, a Modified Ant-Search (MAS) technique is proposed to optimize the synthesis of a four-bar mechanism with a path generation task. Two major improvements are applied over the previous algorithm: ants pheromone update and exploration/exploitation techniques are both modified. Unlike the previous work where a constant quantity of pheromones was added during each iteration, in this paper, the pheromone deposit rate is proportional to the error of the objective function. Such a modification in the pheromone update rule is expected to differentiate between the behaviors of different ants and better govern their motion in the subsequent iterations. Moreover, the second major improvement targets the exploration/exploitation techniques followed by the ants. Unlike the previous work where exploration dominates during the early iteration stages and exploitation during the late ones, this work implements a more dynamic strategy where ants enter and leave the exploration/exploitation processes as governed by parameters related to the objective function error and pheromone deposit levels. Such modifications applied to the Ant-Search (AS) technique are expected to ensure a better chance of converging to a global minimum. The MAS technique is applied for a few path generation tasks with prescribed timing along with a set of linear constraints. Results are compared with previous work in the literature where the newly proposed technique showed appreciable improvement as evaluated by the structural error objective function. Future work possibilities are also introduced.Copyright

Analysis and Optimum Synthesis of Planar Mechanisms With Clearances and Tolerances Using Interval Analysis

This paper presents an interval analysis based approach for analysis and optimization of planar mechanisms including the effect of link lengths tolerances and joint clearances. Mathematical models describing the effect of the mechanical errors on the desired output parameters of the mechanism are derived, followed by the commissioning of an ant-gradient based optimization scheme to find the maximum possible deviations in them. A new optimization approach is also devised to design a four bar mechanism with “minimal” sensitivity to clearances and tolerances. An analysis case study is presented and the results are displayed in the form of fuzzy membership functions and compared with results obtained using a similar approach. An example on optimum synthesis of a four-bar mechanism with joint clearances and link tolerances for hybrid exact/approximate points trajectory is also presented. Insights based on the assessment of the results are introduced.Copyright

Neuro-Fuzzy Control of Smart Flexible Mechanisms

This paper presents an analytical investigation in which a controller based on Adaptive Neuro-Fuzzy Inference System (ANFIS) is designed and implemented to control the vibrations of a flexible mechanism system with smart coupler link. The most dominant vibration mode of the mechanism is identified and the controller is then designed to reduce the effect of this mode on the response of the mechanism system. The proposed control algorithm is implemented on a mechanism system with a thin plate-type piezoceramic actuator bonded to the coupler link surface at the high strain location corresponding to the dominant mode. Simulation results showed the capability of the proposed ANFIS controller to reduce the coupler link mid point amplitudes by nearly 60%.Copyright

An Ants-Search Based Method for Optimum Synthesis of Compliant Mechanisms

Compliant mechanisms are widely used in the industry and have gained more popularity in the past few decades with the advancements in smart materials and micro-electro mechanical systems (MEMS). Compliant mechanisms offer huge advantages over the classical rigid linkages due to their flexible behavior. Such flexible mechanisms reduce production time and cost especially that they eliminate the need of joints that can get pretty hectic especially at micro level manufacturing and assembly. By avoiding multi-joints in the design and their consequent clearances, a compliant mechanism can offer higher precision over its rigid counterpart. However, these advantages come with a price; compliant mechanisms are more challenging in terms of design and analysis. Many compliant mechanisms are designed to undergo relatively large deflections which in turn impose geometric nonlinearities. In the past, many compliant designs were based on intuition, experience, and trial and error. Later on, many theories developed to assist in designing and analyzing compliant mechanisms before proceeding with the manufacturing phase. This paper covers topology optimization of compliant structures using beam elements. The swarm intelligence technique known as Ant Search (AS) is used to find the optimum design that satisfies the required mechanism performance. A case study that involves the topology design of a miniature compliant displacement amplifier is presented and results are compared with the finite element solver ANSYS. The optimized topology mechanism produced a much larger amplification ratio as compared to that presented in literature. Results produced show the high potential of swarm intelligence and AS in particular at solving multi-disciplinary optimization problems that should not be limited to designs that involve physical paths.Copyright

Application of the Flexible Link Model (FLM) and Quantum Computing-Based Algorithm for the Optimum Synthesis of Partially Compliant Mechanisms

The purpose of this paper is to integrate the concept of the flexible link model (FLM) with a modified real coded quantum inspired evolutionary algorithm (MRQIEA) for the optimum synthesis of partially compliant mechanisms. The purpose is to synthesize the compliant and rigid members of a partially compliant mechanism in a single optimization run. The compliant member parameters are defined by the FLM which facilitates the integration of the design variables associated with compliant members and rigid links in one solution vector. Four examples are presented to demonstrate the approach.Copyright