scholarly journals An adaptive particle swarm optimization method for multi-objective system reliability optimization

2019 ◽  
Vol 233 (6) ◽  
pp. 990-1001 ◽  
Author(s):  
Mohamed Arezki Mellal ◽  
Enrico Zio
Keyword(s):  
Particle Swarm Optimization ◽  
System Reliability ◽  
Particle Swarm ◽  
Superior Performance ◽  
Reliability Optimization ◽  
Swarm Optimization ◽  
Adaptive Particle Swarm Optimization ◽  
Multi Objective ◽  
Objective System ◽  
System Reliability Optimization

Multi-objective system reliability optimization has attracted the attention of several researchers, due to its importance in industry. In practice, the optimization regards multiple objectives, for example, maximize the reliability, minimize the cost, weight, and volume. In this article, an adaptive particle swarm optimization is presented for multi-objective system reliability optimization. The approach uses a Lévy flight for some particles of the swarm, for avoiding local optima and insuring diversity in the exploration of the search space. The multi-objective problem is converted to a single-objective problem by resorting to the weighted-sum method and a penalty function is implemented to handle the constraints. Nine numerical case studies are presented as benchmark problems for comparison; the results show that the proposed approach has superior performance than a standard particle swarm optimization.

scholarly journals Multi-Objective Power Distribution Network Reconfiguration using Chaotic Fractional Particle Swarm Optimization

2021 ◽  
Vol 19 (1) ◽  
pp. 43-50
Author(s):  
Yashar Mousavi ◽  
Mohammad Hosein Atazadegan ◽  
Arash Mousavi
Keyword(s):  
Particle Swarm Optimization ◽  
Power Distribution ◽  
Distribution Network ◽  
Particle Swarm ◽  
Superior Performance ◽  
Network Reconfiguration ◽  
Power Distribution Network ◽  
Swarm Optimization ◽  
Distribution Network Reconfiguration ◽  
Multi Objective

Optimization of power distribution system reconfiguration is addressed as a multi-objective problem, which considers the system losses along with other objectives, and provides a viable solution for improvement of technical and economic aspects of distribution systems. A multi-objective chaotic fractional particle swarm optimization customized for power distribution network reconfiguration has been applied to reduce active power loss, improve the voltage profile, and increase the load balance in the system through deterministic and stochastic structures. In order to consider the prediction error of active and reactive loads in the network, it is assumed that the load behaviour follows the normal distribution function. An attempt is made to consider the load forecasting error on the network to reach the optimal point for the network in accordance with the reality. The efficiency and feasibility of the proposed method is studied through standard IEEE 33-bus and 69-bus systems. In comparison with other methods, the proposed method demonstrated superior performance by reducing the voltage deviation and power losses. It also achieved better load balancing.

Interactive enhanced particle swarm optimization: A multi-objective reliability application

2007 ◽  
Vol 221 (3) ◽  
pp. 177-191 ◽  
Author(s):  
M. K. Pandey ◽  
M. K. Tiwari ◽  
M. J. Zuo
Keyword(s):  
Particle Swarm Optimization ◽  
Decision Maker ◽  
Optimization Problems ◽  
Particle Swarm ◽  
Reliability Optimization ◽  
Swarm Optimization ◽  
Multi Objective ◽  
New Variant ◽  
Conflicting Objectives ◽  
Enhanced Particle Swarm Optimization

In reliability optimization problems, it is desirable to address different conflicting objectives. This generally includes maximization of system reliability and minimization of cost, weight, and volume. The proposed algorithm of a metaheuristic nature is designed to address multi-objective problems. In the presented algorithm, interaction with a decision maker guides the search towards the preferred solution. A comparison between an existing solution and the newly generated solution substantiates the desirability or fitness of the latter. Further, the utility function expresses the preference information of the decision maker while searching for the best solution. During the development of the algorithm, a new variant of particle swarm optimization (PSO) is proposed and named as ‘enhanced particle swarm optimization’ (EPSO). EPSO considers the difference between the particle's best position and the global best position for efficient search and convergence. The developed algorithm is applied to the reliability optimization problem of a multistage mixed system with four different value functions that are used to simulate the designer's opinion in the solution evaluation process. Results indicate that the algorithm effectively captures the decision maker's preferences for different structures. Superior results in multi-objective reliability problem-solving prove the algorithm's superiority over other approaches.

Sign in / Sign up

Export Citation Format

Share Document