particle swarm algorithm

The forming of tailor machined blanks due to the different thicknesses is more complex than the monolithic sheets. The presence of two sections with different thicknesses leads to different formability behaviors and consequently their different spring-backs. Therefore, it is important to examine the spring-back in the thin and thick sections of tailor machined blanks in order to minimize the spring-back value. In this paper, an adaptive neural-fuzzy inference system is used to model the effect of important parameters to predict the percent of spring-back corresponding to the thin and thick sections on the creep age formed tailor machined blanks. In addition, the obtained model is used to optimize the creep forming process using particle swarm optimization algorithm. In order to model the process behavior, some experiments have been performed on the creep forming of tailor machined blanks. The accuracy of the obtained model is investigated using different figures as well as by the specific criteria , , , and .

In this paper, the effect of using various intelligent algorithms to optimize the adaptive neuro-fuzzy disturbance observer has been investigated .First, a model reference adaptive control is designed for the spacecraft simulator. Then, in order to reduce the disturbance effect, an adaptive neuro-fuzzy disturbance observer is used. In this paper, the fuzzy system is optimized using Intelligent Genetic Algorithm, Particle Swarm Optimization, Imperialist Competitive Algorithm, Bee Colony, Ant Colony Optimization, and especially Policy Gradient Particle Swarm Algorithm, which speeds up and optimizes the response. The Policy Gradient Particle Swarm algorithm is a combination of gradient policy reinforcement learning and particle swarming ideas and is a hybrid optimization method to control a nonlinear complex system with many applications in the real world. In this method, influenced by reinforcement idea, the policy gradient for a non-fossilized system is calculated, and in the optimization of particle swarm relations, optimization is performed in addition to the factors included in the congestion methods in the direction of the policy gradient. It is intended to optimize the fuzzy neuro system parameters and input and output data in the cost function. Finally, the neuro-fuzzy systems optimized by these algorithms are compared and it is shown that the gradient particle swarm algorithm performs better than the particle swarm algorithm.

This paper investigates the influence parameters in optimum design of finite metallic plates with polygonal cutouts subjected to uniaxial loading. This objective is pursued by a particle swarm algorithm. The important features of this algorithm include flexibility, simplicity, short solution time, and the ability to avoid local optimums. The objective function used for in this algorithm seeks to minimize the value of stress concentration factor (optimum stress) around triangular, square, pentagonal and hexagonal cutouts. The novelty of this paper is determining of optimum stress around the polygonal cutouts in finite metallic plates under uniaxial loading. The method used for the calculation of stress concentration is based on analytical solution of Muskhelishvili complex variable and conformal mapping with plane stress assumption. The plate is assumed to be isotropic, linear elastic, and finite (the ratio of cutout side to plate’s longest side in the square and triangular cutouts and the ratio of diameter of circle circumscribing other n-gonals to the plate’s longest side are greater than 0.2). Finite element numerical solution is employed to examine the results of present analytical solution. Numerical results are in good agreement with the results of the present analytical solution. Results show that by selecting the aforementioned parameters properly, less amounts of stress could be achieved around the polygonal cutouts leading to an increase in load-bearing capacity of the structure.

In the present paper, the calculation of the optimum characteristics of thin fins attached to the hot wall in closed cavities with different aspect ratios has been investigated. Free convection is predominant in the cavity. The equations of continuity, momentum, and energy are discretized by means of finite volume method and the equations will be solved by a SIMPLER algorithm. The fins are connected to the hot wall and Particle Swarm algorithm is used to optimize the location and the length of fins. In order to model fins with high heat transfer, the dimensionless diffusion coefficients of momentum and energy equations are set equal to infinity and for the modeling of insulator fins, these coefficients are considered infinite and very small, respectively. The aim is to find the optimum characteristics of the array of fins attached to the hot wall in the rectangular cavities in such a way that the heat transfer from the cold wall is minimized. The results of particle swarm optimization algorithm are compared with the reference amounts. Results show that the particle swarm optimization algorithm is capable to find the optimum characteristics of an array of fins that is not calculated by the other methods, until now. The obtained results showed that with the increase of aspect ratio, the increase of the number of fins (increase in the number of variables), the particle swarm optimization algorithm might not have the needed ability to find the general optimum. This issue was studied by some numerical tests. Therefore, it was concluded that by decreasing the number of variables (Fixed location) and finding only the length of each fin, and also by increasing the number of particles in the sample space, the accuracy of the algorithm can be increased.

Reliability and optimization are two key elements for structural design. The reliability-based topology optimization (RBTO) is a powerful and promising methodology for finding the optimum topologies with the uncertainties. In this paper, the particle swarm algorithm (PSO) using performance measure approach (PMA) is proposed in the RBTO procedure. Conventionally, the approximate limit state function along with the most probable point (MPP) search algorithms is used for calculation the reliability index. On the other hand, the choice of penalty function for having a convergent search plays a critical role. In addition one does not need to use approximate limit state function and calculating the derivatives of limit state function with respect to random variables. Furthermore, the convergence problem of the MPP search algorithms for complicated limit state functions does not exist. This paper presents RBTO using bi-directional evolutionary structural optimization (BESO) with an improved filter scheme. The topologies obtained by RBTO are compared with that of obtained by deterministic topology optimization (DTO). Results of the RBTO using PSO show that PSO can be effectively applied to RBTO and its use is quite simple.

This paper presents a fast and efficient aerodynamic optimization method for megawatt class wind turbines. For this purpose WP_Baseline 1.5 MW wind turbine is used as a test case. Modified particle swarm optimization (PSO) algorithm is used in this study. PSO parameteric studies are conducted, to increase both efficiency and speed of optimization cycle. Since in aerodynamic optimization, it is very desirable to limit the number of the variables, in this study geometric 'class function/shape function' transformation technique (CST) is used for blade geometry parameterization and the appropriate order of shape function polynomial is proposed for S818, S825 and S826 airfoils. Improved Blade Element Momentum (IBEM) theory is implemented for wind turbine power output estimation, validated with experimental and Computational Fluid Dynamic (CFD) data of AOC wind turbine. The aerodynamic data needed for IBEM is provided by XFoil software. XFoil output data for pressure coefficient and wall shear stress which are validated against experimental and CFD data, are applied as the aerodynamic input data for IBEM method.
The twist, the chord and 3 types of airfoil for all sections of the turbine blade are optimized using IBEM method. Optimization is performed with realistic constraints to produce feasible geometry. The performance of the final optimized geometry is simulated via 3D steady incompressible Navier–Stokes equations coupled with Transition SST Model CFD simulation to predict the performance improvement. The results show about 4 percent power enhancement for WP_Baseline wind turbine.

In this paper the optimal values of effective parameters on the stress distribution around polygonal cutouts in isotropic plates are calculated. To achieve this goal, the complex variable method and PSO algorithm have been used. The expansion of the Muskhelishvili’s method are used to analyze the stress distribution in infinite isotropic plates containing various cutouts. By using conformal mapping, the area outside the non-circular cutout is mapped to the area outside of unit circle. The effective parameters on stress distribution around the cutout as design variables include: cutout shape, cutout orientatin and bluntness. The proper selection of these parameters leads to achieve minimum stress around the cutout and result in the load-bearing capacity of structures increases. The goal function in this problem is the maximum stress created around the cutout calculated by the analytical solution method. The results presented in this study shows that by choosing the appropriate shape of cutout and the optimal effective parameters, stress concentration factor can be significantly reduced and lowest stress concentration factor rather than the value of stress concentration corresponding to circular hole can be achieved.

This study explores application of a bees algorithm (BA) for economic optimal design of plate-fin heat exchangers. Therefore in this study, the optimization is targeting two single-objective functions separately. The first is the minimum heat transfer area which is mainly associated with the capital cost of the heat exchanger and the other is minimum total pressure drop that represents the operating cost for specific heat duty requirement under given space restrictions. Based on applications, heat exchanger length, fin frequency, numbers of fin layers, lance length of fin, fin height and fin thickness of the heat exchanger are considered for optimization. The constraints are handled by penalty function method. Also, the effectiveness and accuracy ofthe proposed algorithm is demonstrated through a case study. Comparing the results with the corresponding results using genetic algorithm (GA)and particle swarm optimization (PSO) algorithm reveals that the bees algorithm can converge to optimum solution with higher accuracy