Showing papers on "Multi-swarm optimization published in 2020"

PID neural network decoupling control based on hybrid particle swarm optimization and differential evolution

Abstract: For complex systems with high nonlinearity and strong coupling, the decoupling control technology based on proportion integration differentiation (PID) neural network (PIDNN) is used to eliminate the coupling between loops. The connection weights of the PIDNN are easy to fall into local optimum due to the use of the gradient descent learning method. In order to solve this problem, a hybrid particle swarm optimization (PSO) and differential evolution (DE) algorithm (PSO-DE) is proposed for optimizing the connection weights of the PIDNN. The DE algorithm is employed as an acceleration operation to help the swarm to get out of local optima traps in case that the optimal result has not been improved after several iterations. Two multivariable controlled plants with strong coupling between input and output pairs are employed to demonstrate the effectiveness of the proposed method. Simulation results show that the proposed method has better decoupling capabilities and control quality than the previous approaches.

Multi-state particle swarm optimization for discrete combinatorial optimization problem

Abstract: The binary-based algorithms including the binary particle swarm optimization (BPSO) algorithm are proposed to solve discrete optimization problems. Many works have focused on the improvement of the binary-based algorithms. Yet, none of these works have been represented in states. In this paper, by implementing the representation of state in particle swarm optimization (PSO), a variant of PSO called multi-state particle swarm optimization (MSPSO) algorithm is proposed. The proposed algorithm works based on a simplified mechanism of transition between two states. The performance of MSPSO algorithm is emperically compared to BPSO and other two binary-based algorithms on six sets of selected benchmarks instances of traveling salesman problem (TSP). The experimental results showed that the newly introduced approach manage to obtain comparable results, compared to other algorithms in consideration.

A Particle Swarm Optimization -Based Heuristic for Scheduling in FMS Review

Abstract: This paper focus on the problem of simultaneous scheduling of machine and automated guided vehicle (AGV) in a flexible manufacturing system (FMS) so as to minimize the makespan The FMS scheduling problem has been tackled by various traditional optimization techniques. While these methods can give an optimal solution to small-scale problems, different scheduling mechanisms are designed to generate optimum scheduling; these include non-traditional approaches such as genetic algorithm (GA), memetic algorithm (MA) and particle swarm algorithm (PSA) by considering multiple bjectives,i.e.,minimising the idle time of the machine andminimising the total penalty cost for not meeting the deadline concurrently. Two optimization algorithms ( genetic algorithm and particle swarm algorithm) are compared and conclusions are presented

Aeroengine Performance Prediction Based on Double-Extremum Learning Particle Swarm Optimization

Abstract: Abstract To predict the performance parameter changing trend of an aeroengine, a novel double-extremum learning particle swarm optimization (DELPSO) algorithm is proposed. Inspired by human learning behavior, this algorithm simulates this behavior, including the strategies of collective learning, private tutoring, and research behavior, so that obtained final solutions would be in the global optimal area or its neighbor area as close as possible. Meanwhile, to improve the prediction performance, a nonlinear mapping function is designed to describe the feature relationship between inputs and outputs of historical data. Based on the DELPSO, the fitness function synthetically considers the changing trend and the prediction error and can adaptively select optimal parameters of the nonlinear mapping function. The experimental results demonstrate that the DELPSO has globally stable and reliable performance. To validate the prediction performance of the proposed DELPSO, it is also applied to an aeroengine. Its good prediction performance indicates that the proposed DELPSO is an important reference for maintenance decision-making of aeroengines.

Multi-Objective particle swarm optimization algorithms – A leader selection overview

Abstract: Multi Objective Optimization (MOO) problem involves simultaneous minimization or maximization of many objective functions. Various MOO algorithms have been introduced to solve the MOO problem. Traditional gradient-based techniques are one of the methods used to solve MOO problems. However, in the traditional gradient-based technique only one solution is generated. Thus, an alternative approach such as Particle Swarm Optimization (PSO), which able to produce a number of possible solutions are highly desirable. In PSO, particles search the optimum solution under the influence of a better solution known as leader. This leader facilitates cooperation between all particles. However, this strategy to select the leader has to be changed when it is used for MOO problems. This paper presents an overview of the multi-objective PSO algorithms, which emphasize on the leader selection. In addition, the description of PSO and multi-objective optimization problems are also provided.

Adaptive Multi-swarm Bat Algorithm (AMBA)

Abstract: Modified shuffled multi-population bat algorithm (MSMPBat) is a recently proposed swarm algorithm. It divides its population into multiple sub-populations (SPs), each of which uses different parameter settings and evolves independently using an enhanced search mechanism. For information exchange among these SPs, a solution from one SP is copied to the next after every generation. This process leads to duplication of solutions over time. To overcome this drawback, different techniques are introduced. Opposition-based learning is used to generate a diverse starting population. For information exchange, if a solution comes too close to the swarm best, only then it is sent (moved, not copied) to another swarm. Four techniques are proposed to select this second swarm. Initially, the selection probability of each technique is same. The algorithm adaptively updates these probabilities based on their success rate. The swarm which gave up the solution uses a modified opposition-based learning technique to generate a new solution. These changes help to maintain the overall diversity of the population. The proposed approach, namely, adaptive multi-swarm bat algorithm (AMBA), is compared to six algorithms over 20 benchmark functions. Results establish the superiority of adaptive multi-swarm bat algorithm.

Particle Swarm Optimization

Abstract: Optimization techniques based on behaviors of some animal species in natural environment are strongly developed. Algorithms simulating behaviors of bees colony, ants colony, and birds flock have appeared. The last algorithm is named in the literature as a particle swarm optimization (PSO) algorithm, and is a new technique dedicated to optimization problems having continuous domain. However, its modifications to optimize discreet problems have been developed lately. The PSO algorithm has many common features with evolutionary computation techniques. This algorithm is operating on randomly created population of potential solutions, and is searching optimal solution through the creation of successive populations of solutions. Genetic operators like cross-over and mutation, which exist in evolutionary algorithms, are not used in the PSO algorithm. In this algorithm, potential solutions are moving to the actual optimum in the solution space. There exist two versions of PSO algorithm: local, LPSO, and global, GPSO, algorithm.