جستجوی مقالات مرتبط با کلیدواژه « objective genetic algorithm » در نشریات گروه « فنی و مهندسی »

Ensemble classifier is an effective method in machine learning that attempted to provide a better approximation of an optimal classifier with combination of some classifiers results. To achieve better performance, the base classifiers should have acceptable efficiency and different classification error, also a suitable method used to combine their results. Various ensemble classification methods such as bagging, voting and strengthening methods have been presented. In this paper, we proposed the ensemble classifier based on weighted mean of the base classifiers output. The weights were estimated using a multi-objective genetic algorithm with taking classification error, sparsity, diversity and density criterion. The results of implementations on UCI datasets show that the proposed method causes more increasing classification accuracy related to other traditional ensemble classifiers.

Simultaneous application of mechanical pressure and electrical charge on powder samples in spark plasma sintering process, has resulted in a sample with a density close to the theory. In the present study, a thermal-electrical-mechanical coupled finite element model of spark plasma sintering system using multi-objective optimization algorithm is proposed to optimize the mold variable. The simulation performed for Si3N4-SiO2 (1:1 mol) specimen has good agreement with the experimental results. Multi-objective genetic algorithms was used for optimization of mold design in order to maximize the temperature of sample core and minimize the mises stress in the mold. The results show that the optimized dimensions cause 8% increase in sample temperature and about 18% decrease in temperature difference between mold surface and sample core. This leads to better uniformity in the porosity distribution of final sample.

Traditional management of water resources, which is often based on cost and benefit, is not sustainable approach; therefore reassessment of the former approaches seems necessary nowadays. Optimum reservoirs operation is one of the important water resource management problems. Present study has used the non dominated sorting genetic algorithm (NSGA-II) for multi-objective optimization of Sefidrud reservoir in North of Iran. The main goal of the study is to reach sustainable operation of the reservoir by considering these objectives simultaneously: supply of downstream demand, sediment evacuation and hydropower generation

In an anechoic chamber, the flat multi-layer sound absorber is cheaper, easier to install and less complicated in manufacturing than conventional wedge and pyramid absorbers. Therefore, design of the optimum flat multi-layer absorber which has minimum thickness is desirable. In this paper, the genetic algorithm has been employed as an effective optimization tool to determine flat triple layer porous absorber. To obtain a broader range of porous materials, combination of foam and fiber types is used. Theoretical and numerical method (finite element method specifically COMSOL Multi-physics version 4.4) have been used to investigate on the operation of sound absorption correspond to the multi-layer structure. In the first step, mathematical model is verified and finite element method, theoretical and experimental results are compared together for two different samples of structures which show appropriate matching. Furthermore to verify the operation of programmed genetic algorithm, the results obtained from the optimization of flat triple layer porous absorber are compared with others that show accuracy and efficiency of this method. The optimization results indicate that a flat triple layer porous structure can achieve results comparable with quality wedge type structure with overall thickness slightly smaller than a fifth of a wavelength at 80 Hz cut-off frequency.

In the area of critical and sensitive facilities location, on the one hand, adequate coverage can reduces injuries and damages caused by disasters and on the other hand, dispersion of critical facilities’ locations can decreases the probability of identification and destruction of the facilities. In this paper, a bi-objective model is presented for the location problem of critical facilities, in which the principals of passive defense are considered via the objective functions of the minimization of the uncovered demand and the maximization of the minimum distance between facilities. In the proposed model, it is assumed that each facility is identified and destructed based on a known probability. In addition, it is supposed that injuries (customers) can be served by those facilities located within a predefined coverage distance. Since the proposed problem is NP-hard, a bi-objective genetic algorithm is developed to solve large size real-world problems.

Conventional methods for solving the problem of site location for booster chlorination stations have assumed the use of linear superposition principle for first-order kinetics of chlorine decay in water distribution systems in order to speed up the process of evolutionary algorithms. However, examination of this assumption in this paper shows that it causes a non-trivial error, especially when the order of chlorine decay rate is more than one. This paper presents a novel meta-model for solving the multi-objective optimization problem of optimal locations for booster chlorination stations for the nonlinear order of chlorine decay rate. In the proposed model, the meta-model is the water quality simulation model quantified by the principle of linear superposition for nonlinear kinetics of chlorine decay. To do so, residual chlorine concentration is calculated from the injection unit chlorine concentration at each node in the network as offline outside the optimisation loop. Then, within the optimization loop, the residual chlorine concentration in the network obtained for the optimal solutions is calculated as the combined chlorine injected from different locations and at different concentrations based on the linear superposition meta-model of the previous part. Objective functions of the optimisation solutions are quickly evaluated by this meta-model. In order to mitigate the significant error due to the estimation of this meta-model, the fitness of the best solutions are again evaluated using the real water quality simulation model (nonlinear chlorine decay rate) and replaced with the evaluations previously approximated by the meta-model. The results show the desirable accuracy of the proposed model and the high speed in the run time of the hybrid optimisation model.

In this paper the creep behavior of a functionally graded (FG) rotating disc made of Aluminum 6061 and Silicon Carbide is investigated and the optimum volume fraction of FG disc and its profile has been obtained. For this purpose, the temperature gradiant along the disc radius is obtained by solving the govering heat transfer differential equation. All the thermal properties of the material are assumed to be the function of temperature and volume fraction. To obtain material properties, two models of Mori-Tanaka and Hashin-Schtrickman are used. To validate the results, they are compared with those given in the literature. Two solution semi-analytical and closed form are employed and the results are compared. The optimum design is carried out with one, and multi-objective methods which are based on genetic algorithm. The objectives are increasing the factor of safety, reducing the weight of the disc and reducing the range between minimum and maximum safety factors. The design variables are percentage of volume fraction, the power of material distribution formula, and the thickness of the disc. The results show that two solution methods compare well. Also, it has been shown that high fraction of Silicon Carbide in the outer side the disc provide optimum results. Also, contradiction of the objectives is reviled, hence the results are presented as Pareto front.

In this paper، multi-objective Genetic Algorithm has been employed to find the communications satellite optimal transfer trajectory from geosynchronous transfer orbit (GTO) to geosynchronous orbit (GEO) as the destination operational zone. Because of satellite high specific impulse، Orbital maneuvers are considered as impulsive maneuvers in this paper and its validation is verified by comparison of results with continues one. Number of intermediate orbits has been found in order to physical constraints. In this systematic approach، both minimum-time and fuel-saving using constant acceleration simultaneously as the problem strategy to find the optimal transfer trajectory between two orbits. Set of optimal trajectories are plotted in Pareto Front and transfer trajectory can be selected from these points. Finally، the results are compared with the results obtained using STK software and good agreement has been revealed. The satellite access to ground station (GS) is analyzed and optimal position for GS is found by defining lack of access and mass consumption as objective functions. Finally، disturbance torques induced by liquid apogee motor (LAM) is investigated and thrusters'' fuel mass for this disturbance torque control is proposed.

This paper deals with dynamic modeling and path planning of a spherical mobile robot. Spherical robot is the prototype of mobile robot used in unknown environments. In this paper, it has five degrees of freedom such that its driver mechanism is 2DOF pendulum moved with two motors. This spherical robot is a non-holonomic and non-linear system. First, the kinematic of the spherical robot is analyzed and next its equations of motion are derived,using the Kane method. For path planning, inverse dynamic of the robot on straight and on circular trajectories are solved. For this, new iterative method, using genetic algorithm for path planning in motion on circular path is used.Finally, the optimal path for achieving minimum power and time of motion are obtained, using multi-objective genetic algorithm. Good agreements were observed between the results of this method and previous computer simulations.