A fuzzy adaptive turbulent particle swarm optimisation

Abstract

Particle Swarm Optimisation (PSO) algorithm is a stochastic search technique, which has exhibited good performance across a wide range of applications. However, very often for multimodal problems involving high dimensions, the algorithm tends to suffer from premature convergence. Analysis of the behaviour of the particle swarm model reveals that such premature convergence is mainly due to the decrease of velocity of particles in the search space that leads to a total implosion and ultimately fitness stagnation of the swarm. This paper introduces Turbulence in the Particle Swarm Optimisation (TPSO) algorithm to overcome the problem of stagnation. The algorithm uses a minimum velocity threshold to control the velocity of particles. The parameter, minimum velocity threshold of the particles is tuned adaptively by a fuzzy logic controller embedded in the TPSO algorithm, which is further called as Fuzzy Adaptive TPSO (FATPSO). We evaluated the performance of FATPSO and compared it with the Standard PSO (SPSO), Genetic Algorithm (GA) and Simulated Annealing (SA). The comparison was performed on a suite of 10 widely used benchmark problems for 30 and 100 dimensions. Empirical results illustrate that the FATPSO could prevent premature convergence very effectively and it clearly outperforms SPSO and GA. research interestes are in advanced computational intelligence with a focus on hybridising intelligent techniques involving connectionist network learning, fuzzy inference systems, swarm intelligence, evolutionary computation, distributed artificial intelligence, multiagent systems and other heuristics.Weishi Zhang is the Dean of School of Computer Science, Dalian Maritime University, Dalian, China. His research interests are in artificial intelligence, soft computing, nature-inspired computing, swarm intelligence and generic information systems development. He is an Editorial

Improved Particle Swarm Optimization with low-discrepancy sequences

“…a search area having low dimension), but as we go on increasing the dimension of search space the performance deteriorates and many times converge prematurely giving a suboptimal result [2]. This problem becomes more persistent in case of M. Pant is with the Indian Institute of Technology Roorkee, Saharanpur, 247001, India (phone: +91-9759561464; e-mail: millifpt@iitr.ernet.in).…”

A fuzzified systematic adjustment of the robotic Darwinian PSO

“…For instance, in the work of Shi and Eberhart [16], a fuzzy system is merged into the PSO to dynamically adapt the inertia weight of particles. Similarly, Liu et al [17] presents a fuzzy logic controller to adaptively tune the minimum velocity of the PSO particles. Several other authors considered incorporating selection, mutation and crossover, as well as the differential evolution, into the PSO algorithm.…”

Particle Swarm Optimization: Performance Tuning and Empirical Analysis

“…a search area having low dimension), but as the dimension of search space is increased, the performance deteriorates and many times converge prematurely giving a suboptimal result [5]. This problem becomes more persistent in case of multimodal functions having several local and global optima.…”

“…Although PSO has shown a very good performance in solving many test as well as real life optimization problems, it suffers from the problem of premature convergence like most of the stochastic search techniques, particularly in case of multimodal optimization problems. The curse of premature convergence greatly affects the performance of algorithm and many times lead to a sub optimal solution [5]. Aiming at this shortcoming of PSO algorithms, many variations have been developed to improve its performance.…”