Abstract:In this study, we present a multi-objective particle swarm optimization algorithm based on density clustering for the leader particle selection strategy, called DN-MOPSO, to solve multi-objective optimization problems. The algorithm divides particles, which are composed of evolutionary population and external archive, into several classes using density clustering technique in decision variable space, and then the particles in the evolutionary population select their respective global leader particles depend on the classification results, for the purpose of making a balance between global search capabilities and local search capabilities. Moreover, the improved circular crowed distance sorting to truncate the external archive when the nondominated solutions exceed the maximum limit, in order to make the distribution of the obtained solution set better. In addition, the Gauss mutation operator mechanism is also utilized, during the iterative process, to help the algorithm jump out of premature and promote the search capacities. The performance of the proposed algorithm for multi-objective optimization problems is tested on several WFG benchmarks functions and compared to some classic multi-objective optimization algorithms. The simulation results show that DN-MOPSO allows better approximations to optimal Pareto front and get well-spread Pareto solutions. And, the strategy added in this paper not only improves the search scope of the algorithm, but also improves the search accuracy and at the same time maintains the diversity of nondominated solution. Therefore, the convergence and diversity of the solutions obtained by DN-MOPSO are significantly improved. The results also indicate that DN-MOPSO is highly feasible and competitive to solve multi-objective optimization problems.

A K-means-based Multi-subpopulation Particle Swarm Optimization for Neural Network Ensemble

Neural Networks Learning Using Vbest Model Particle Swarm Optimisation

Exploratory Study of Sorting Particle Swarm Optimizer for Multiobjective Design Optimization

Negatively correlated neural network ensemble with multi-population particle swarm optimization

Knowledge-based cooperative particle swarm optimization

Survival Density Particle Swarm Optimization for Neural Network Training

Hybridizing Niching, Particle Swarm Optimization, and Evolution Strategy for Multimodal Optimization

A Novel Multimodal-Problem-oriented Particle Swarm Optimization Algorithm

A Dynamic Neighborhood Balancing-Based Multi-objective Particle Swarm Optimization for Multi-modal Problems

Multi-objective particle swarm optimization with R2 indicator and adaptive method

AMPSO: Artificial Multi-Swarm Particle Swarm Optimization

A Scatter Learning Particle Swarm Optimization Algorithm for Multimodal Problems

Multi-objective Particle Swarm Optimization Algorithm Based on Adaptive Network and Dynamic Crowding Distance and Its Application

Chaotic Particle Swarm Optimization Algorithm Based on Hierarchical Multi-Subpopulation

Multi-objective Particle Swarm Optimization Algorithm of Multi-Swarm Co-Evolution

A hybrid particle swarm optimization and its application in neural networks

Multi-objective Particle Swarm Optimization Algorithm Based on Density Clustering

Multi-strategy self-learning particle swarm optimization algorithm based on reinforcement learning

Multi-sample Learning Particle Swarm Optimization with Adaptive Crossover Operation

A Pacesetter-Lévy multi-objective particle swarm optimization with Arnold Chaotic Map with opposition-based learning

A Multi-Swarm Cooperative Multistage Perturbation Guiding Particle Swarm Optimizer