A Hybrid Particle Swarm Optimization Embedded Trust Region Method
As one of swarm intelligence algorithms, particle swarm optimization (PSO) has good global search ability, but the main disadvantage is that it is easy to fall into the local minima, and the convergence accuracy is restricted. Trust region method is an im
- PDF / 187,298 Bytes
- 10 Pages / 439.37 x 666.142 pts Page_size
- 106 Downloads / 191 Views
School of Computer Science and Communication Engineering, Jiangsu University, Zhenjiang, Jiangsu 212013, China {2211408046,hanfei}@ujs.edu.cn 2 School of Computers Engineering, Jinling Institute of Technology, Nanjing, Jiangsu 211169, China [email protected]
Abstract. As one of swarm intelligence algorithms, particle swarm optimization (PSO) has good global search ability, but the main disadvantage is that it is easy to fall into the local minima, and the convergence accuracy is restricted. Trust region method is an important numerical method for solving nonlinear optimization problems with reliability, stability and strong convergence. In this paper, a hybrid PSO Embedded trust region method is proposed to improve the search ability of the swarm. The algorithm effectively combines the global search of particle swarm optimization with the fast and precise local search capability of the trust region method. The experiment results show that it has much better accuracy and convergence to the global optimal solution. Keywords: Hybrid particle swarm optimization Numerical experiment
Trust region method
1 Introduction Particle swarm optimization (PSO) [1] was firstly proposed by Kennedy and Eberhart in 1995. The basic idea is inspired by modeling and simulating the results of the study on the behavior of flocks of birds in the early stage. Particle swarm optimization algorithm is easy to fall into the local optimum, and the convergence accuracy is not high. In the past 20 years, many researchers have conducted research in-depth, and put forward some improved PSOs, and achieved better results than traditional PSO. In [2] an improved particle swarm optimization algorithm with inertia weight was proposed to balance the global search ability and local search ability of the algorithm. The greater the inertia weight, the stronger the global search ability, the smaller the inertia weight, the stronger the local search ability. In literature [3, 4], Clerc introduced the concept of shrinkage factor, and the method described a selection of the inertia weight w, the acceleration constants c1 and c2 to ensure the convergence of the algorithm. In [5] the author presented a particle swarm optimizer with passive congregation to improve the performance of standard PSO. Passive congregation is an important biological force preserving swarm integrity. By introducing passive congregation to PSO, information can be transferred among individuals of the swarm. In [6] Sugantan © Springer International Publishing Switzerland 2016 D.-S. Huang et al. (Eds.): ICIC 2016, Part I, LNCS 9771, pp. 762–771, 2016. DOI: 10.1007/978-3-319-42291-6_76
A Hybrid Particle Swarm Optimization Embedded Trust Region Method
763
proposed an improved particle swarm optimization based on the neighborhood region. The basic idea is that at the beginning, the neighborhood of each individual is itself. With the growth of the evolution generation, the neighborhood is also expanding to the whole population. In order to avoid premature convergence of particle swar
Data Loading...
