Search

This study deals with an identification of stiffness reduction and response predictions occurred by the delamination damage in laminated composite plates under impact loads. Combined bivariate Gamma function and microgenetic algorithms are developed to determine the crack region due to the delamination and predict future responses. The validity of the proposed method was verified using impact-induced data obtained from a two-dimensional delamination finite element model. Examples indicated that the proposed approach is a feasible and advantageous method through which future dynamic responses can be predicted and the distribution of the degraded stiffness of laminated composite structures can be inspected for different measuring locations and fibre angles. [ABSTRACT FROM AUTHOR]

Abstract: This study aims to use optimization to increase the design load of composite sandwich cylinders under external hydrostatic pressure. Unlike other studies that only consider buckling, this study took into account both buckling and material failure. MSC.NASTRAN was used for the finite element analysis, while a micro-genetic algorithm was used for the optimization. The finite element model was validated by comparison with the experiment results, and the result of the optimization using the finite element method was validated by comparison with the result of the feasible region analysis. Based on the optimization, as the thickness of the sandwich increases, the buckling load becomes larger than the material failure. Consequently, the optimum point is determined by material failure. The current results suggest that both the buckling and the static material failure should be considered in the design of the composite sandwich cylinder. [Copyright &y& Elsevier]

This paper presents parameter and topology optimization of inductor shapes using evolutionary algorithms. The goal of the optimization is to reduce the size of inductors satisfying the specifications on inductance values under weak and strong bias-current conditions. The inductance values are computed from the finite-element (FE) method taking magnetic saturation into account. The result of the parameter optimization, which leads to significant reduction in the volume, is realized for test, and the dependence of inductance on bias currents is experimentally measured, which is shown to agree well with the computed values. Moreover, novel methods are introduced for topology optimization to obtain inductor shapes with homogeneous ferrite cores suitable for mass production. [ABSTRACT FROM AUTHOR]

Learning Classifier Systems are Evolutionary Learning mechanisms which combine Genetic Algorithm and the Reinforcement Learning paradigm. Learning Classifier Systems try to evolve state-action-reward mappings to propose the best action for each environmental state to maximize the achieved reward. In the first versions of learning classifier systems, state-action pairs can only be mapped to a constant real-valued reward. So to model a fairly complex environment, LCSs had to develop redundant state-action pairs which had to be mapped to different reward values. But an extension to a well-known LCS, called Accuracy Based Learning Classifier System or XCS, was recently developed which was able to map state-action pairs to a linear reward function. This new extension, called XCSF, can develop a more compact population than the original XCS. But some further researches have shown that this new extension is not able to develop proper mappings when the input parameters are from certain intervals. As a solution to this issue, in our previous works, we proposed a novel solution inspired by the idea of using evolutionary approach to approximate the reward landscape. The first results seem promising, but our approach, called XCSFG, converged to the goal very slowly. In this paper, we propose a new extension to XCSFG which employs micro-GA which its needed population is extremely smaller than simple GA. So we expect micro-GA to help XCSFG to converge faster. Reported results show that this new extension can be assumed as an alternative approach in XCSF family with respect to its convergence speed, approximation accuracy and population compactness. [ABSTRACT FROM AUTHOR]

This research aims at characterizing and predicting the Young's Modulus of thin film materials that are utilized in Microelectromechanical systems (MEMS). Recent studies indicate that the mechanical properties such as Young's Modulus of thin films are significantly different from the bulk values. Due to the lack of proper understanding of the physics in the micro-scale domain the state-of-art estimation techniques are unreliable and often unfit for use for predicating the mechanical behavior of slight modifications of existing designs as well as new designs. This disadvantage limits the MEMS designers to physical prototyping which is cost ineffective and time consuming. As a result there is an immediate need for alternative techniques that can learn the complex relationship between the various parameters and predict the effective Young's Modulus of the thin films materials. The proposed technique attempts to solve this problem using empirical estimation techniques that utilize soft computing techniques for the estimation as well as the prediction of the effective Young's Modulus. As a proof of concept, effective Young's Modulus of Aluminum and TetrathylOrthoSilicate (TEOS) thin films were computed by fabricating and analyzing self-deformed micromachined bilayer cantilevers. In the estimation phase, 2D search and micro Genetic algorithm were studied and in the prediction phase, back propagation based Neural networks and One Dimensional Radial Basis Function Networks (1D-RBFN) were studied. The performance of all combinations of these soft computing techniques is studied. Based on the results, we conclude that performance of the soft computing techniques is superior to the existing methods. In addition, the effective values generated using this methodology is comparable to the values reported in the literature. Given a finite number of data samples, the combination of 1D-RBFN (prediction phase) and GA (estimation phase) presented the best results. Due to these advantages, this methodology is foreseen to be an essential tool for developing accurate models that can estimate the mechanical behavior of thin films. [ABSTRACT FROM AUTHOR]

In this study, we investigate a method to detect damage in a laminated composite structure by analyzing its dynamic response to impact loads. The combined finite element method (FEM) with seven degrees of freedom (DOF) and the advanced microgenetic algorithm described in this paper may allow us not only to detect the damaged elements but also to find their locations and the extent of damage. A high order shear deformation theory (HSDT) is used to predict the structural behavior and to detect damage of laminated composite plates. The effects of noise associated with the uncertainty of measurements due to the complex nature of composites are considered for [0/90] s and [±45] s layup sequences. The results indicate that the new method is computationally efficient in characterizing damage for complex structures such as laminated composites. [ABSTRACT FROM AUTHOR]

Water level monitoring and forecasting are essential tasks in flood emergency response. This study proposes an Edge COMputing-based Sensory NETwork (ECOMSNet), an innovative decentralized early warning system (EWS), for water level monitoring and prediction. A sensor-embedded algorithm integrates the direct step method (DSM) with a microgenetic algorithm (MGA). This algorithm predicts the water surface profile and corrects it once water level observations are available. It also meets efficiency requirements to accommodate sensor computation limitations. The errors in the predicted water surface profiles in channels with gradually varied flows are 5% in a laboratory flume experiment and below 10% in a field experiment. The ECOMSNet is an achievement of edge computing-based Internet of Things. It shows potential to increase emergency response efficiency. However, the system requires further refinement and testing if it is to adequately address rapidly varied unsteady flow in a scaled-up implementation. • An innovative decentralized early warning system (EWS) applies edge computing technique to provide real-time water level forecasts and corrections at local sensors. • Through integration with IoT technology such as on-site microcomputer-based sensors, the edge computing-based IoT, or ECIoT, can collect and process data and then generate decision-supporting information simultaneously, without delay. • Real-time correction can be done seamlessly to improve the performance since observations and calculations are done at the edge of the system. [ABSTRACT FROM AUTHOR]

In this paper, an innovative adaptive directional overcurrent protection system for electric power distribution systems with respect to distributed generation is proposed. The proposed system supervises network topology based on the monitoring functionality of numerical relays. The system detects any changes in the configuration and recalculates the directional overcurrent protection settings by using a microgenetic algorithm. The proposed system was evaluated for several operating scenarios and insertion levels of distributed generation, and then compared with both conventional and adaptive protection systems by means of a traditional genetic algorithm. The results showed that the performance of the proposed system is superior to the other two methods in terms of both speed and selectivity. This shows that this is a promising proposal for the protection of modern electric power distribution systems. [ABSTRACT FROM AUTHOR]