A Modified Particle Swarm Optimization with an Adaptive Acceleration Coefficients

Ziyu, Tang; Zhang, Ding-Xue

doi:10.1109/apcip.2009.217

Cited by 64 publications

(32 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, during the latter stage, it is important to enhance convergence toward the global minima. Ratnaweera et al have proposed time-varying acceleration coefficients (TVAC) with iteration time based on the above concerns [17] and several investigations have been proposed modified TVAC PSO [18][19][20][21]. The modification of the acceleration coefficients in [17] are represented as…”

Section: Adaptation Of the Acceleration Coefficientsmentioning

confidence: 98%

“…In order to overcome the premature convergence, Ratnaweera et al have proposed time-varying acceleration coefficients (TVAC) with iteration time [17]. Several investigations have been modified this method [18][19][20][21]. The major consideration of there modification is to avoid premature convergence in the early iteration and to enhance convergence the the global solution during the latter iteration.…”

Section: Introductionmentioning

confidence: 96%

See 1 more Smart Citation

Estimation of inverse model based on ANN and PSO with adaptively varying acceleration coefficients

Kinoshita

Watanabe

Isshiki

2014

2014 Proceedings of the SICE Annual Conference (SICE)

View full text Add to dashboard Cite

We consider estimation of an inverse model for the uncertain systems. Multi-layered neural network (NN) can approximate any continuous nonlinear mapping, then we use NN in order to represent the inverse model. The the back-propagation rule, which is commonly used for updating the weights, requires the sensitivity function of the system. However, we can not calculate this function because of uncertainty of the systems. Hence, we apply particle swarm optimization (PSO) to update the weights. PSO is suitable for learning NN, because it dose not require the derivative of the objective function. This paper introduces a novel parameter automation strategy in order to overcome the premature convergence. The acceleration coefficients are adaptively varies with the distance between the particle and the gbest which is the best solution ever found by the swarm. We can maintain the diversity of the swarm and search a better solution around the gbest efficiently. We compare the proposed method with the basic PSO and TVAC PSO, which is proposed by Ratnaweera, through the simulation of the inverse kinematics problem. The proposed method can obtain the accurate inverse model than the basic PSO and TVAC PSO.

show abstract

Section: Adaptation Of the Acceleration Coefficientsmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 96%

Estimation of inverse model based on ANN and PSO with adaptively varying acceleration coefficients

Kinoshita

Watanabe

Isshiki

2014

2014 Proceedings of the SICE Annual Conference (SICE)

View full text Add to dashboard Cite

show abstract

“…As a result, several researchers were attracted towards its tuning [56,73,22,70]. Further, researchers also focused on tuning other parameters [53,2,20,74]. A complete survey on different parameter selection schemes for PSO is available in [24].…”

Section: Accepted Manuscriptmentioning

confidence: 99%

Dynamic mentoring and self-regulation based particle swarm optimization algorithm for solving complex real-world optimization problems

Tanweer

Sundaram

Sundararajan

2016

Information Sciences

View full text Add to dashboard Cite

“…There are some studies which have used time-varying coefficients for both cognitive and social coefficients. In 2009, Ziyu and Dingxue [22] introduced an exponentially time-varying acceleration function for adjusting both cognitive and social coefficients in order to control the global search ability and convergence to the global best solution. In 2009, Bao and Mao suggested an asymmetric time-varying acceleration coefficient adjustment strategy [23].…”

Section: Related Workmentioning

confidence: 99%

“…Using dynamic parameter tuning is a method that increases the performance of PSO without suffering from high computational cost [19][20][21][22][23][24]. The main parameters of PSO are the weighting factor (w), cognitive coefficient (c1) and social coefficient (c 2 ).…”

Section: Introductionmentioning

confidence: 99%

Autonomous Particles Groups for Particle Swarm Optimization

2014

View full text Add to dashboard Cite

In this paper a modified Particle Swarm Optimization (PSO) algorithm called Autonomous Groups Particles Swarm Optimization (AGPSO) is proposed to further alleviate the two problems of trapping in local minima and slow convergence rate in solving high dimensional problems. The main idea of AGPSO algorithm is inspired by individuals' diversity in bird flocking or insect swarming. In natural colonies, individuals are not basically quite similar in terms of intelligence and ability, but they all do their duties as members of a colony. Each individual's ability can be useful in a particular situation. In this paper a mathematical model of diverse particles groups called autonomous groups is proposed. In other words different functions with diverse slopes, curvatures, and interception points are employed to tune the social and cognitive parameters of the PSO algorithm to give particles different behaviors as in natural colonies. The results show that PSO with autonomous groups of particles outperforms the conventional and some recent modifications of PSO in terms of escaping local minima and convergence speed. The results also indicate that dividing particles in groups and allowing them to have different individual and social thinking can improve the performance of PSO significantly.

show abstract

A Modified Particle Swarm Optimization with an Adaptive Acceleration Coefficients

Cited by 64 publications

References 3 publications

Estimation of inverse model based on ANN and PSO with adaptively varying acceleration coefficients

Estimation of inverse model based on ANN and PSO with adaptively varying acceleration coefficients

Dynamic mentoring and self-regulation based particle swarm optimization algorithm for solving complex real-world optimization problems

Autonomous Particles Groups for Particle Swarm Optimization

Contact Info

Product

Resources

About