Particle Swarm Optimization Strategy Research Articles

A generation-based optimal restart strategy for surrogate-assisted social learning particle swarm optimization

Evolutionary algorithm provides a powerful tool to the solution of modern complex engineering optimization problems. In general, a great deal of evaluation effort often requires to be made in evolutionary optimization to locate a reasonable optimum. This poses a serious challenge to extend its application to computationally expensive problems. To alleviate this difficulty, surrogate-assisted evolutionary algorithms (SAEAs) have drawn great attention over the past decades. However, in order to ensure the performance of SAEAs, the use of appropriate model management is indispensable. This paper proposes a generation-based optimal restart strategy for a surrogate-assisted social learning particle swarm optimization (SL-PSO). In the proposed method, the SL-PSO restarts every few generations in the global radial-basis-function model landscape, and the best sample points archived in the database are employed to reinitialize the swarm at each restart. Promising individual with the best estimated fitness value is chosen for exact evaluation before each restart of the SL-PSO. The proposed method skillfully integrates the restart strategy, generation-based and individual-based model managements into a whole, whilst those three ingredients coordinate with each other, thus offering a powerful optimizer for the computationally expensive problems. To assess the performance of the proposed method, comprehensive experiments are conducted on a benchmark test suit of dimensions ranging from 10 to 100. Experimental results demonstrate that the proposed method shows superior performance in comparison with four state-of-the-art algorithms in a majority of benchmarks when only a limited computational budget is available.

Chaos-enhanced multiple-choice strategy for particle swarm optimisation

ABSTRACT In this paper, we use pseudo-random number generators based on six chaotic systems to enhance the performance of multiple-choice strategy for particle swarm optimisation (PSO). The multiple-choice strategy is a heterogeneous swarm-based method. We present the results of benchmark testing using the latest CEC’17 benchmark suite. The performance of the proposed method is compared with the canonical PSO and evaluated for statistical significance. Utilising of pseudo-random number generators based on six chaotic systems to enhance the performance of multiple-choice strategy for particle swarm optimization.

Memetic particle swarm optimization scheme for direction-of-arrival estimation in multipath environment

Memetic particle swarm optimization scheme for direction-of-arrival estimation in multipath environment

The impact of PHEVs charging and network topology optimization on bulk power system reliability

Exhausting fossil fuels and increasing environmental pollutions call for the wider deployment of plug-in hybrid electric vehicles (PHEVs), which enable the interactions of transportation and electric sectors. The extra loads for charging massive PHEVs could compromise the power system reliability and impose considerable stress on the power system. Transmission line congestions and generation capacity inadequacy could be caused under this circumstance. However, there is little literature which comprehensively studies the impact of large-scale PHEVs on the bulk power system reliability when smart grid technologies are incorporated including smart charging and flexible transmission topology control technologies. This work studies the impacts of different charging strategies and transmission network operation strategies on the bulk power system reliability with high PHEVs integration. An evolution strategy particle swarm optimization (ESPSO) algorithm is applied to optimize the charging load of PHEVs and shave the peak load. A network topology optimization (NTO) operation strategy is employed to lessen the transmission congestions and reduce the load curtailment. Based on the sequential Monte Carlo simulation method, these technologies are integrated into the reliability evaluation process. Numerical studies and sensitivity analysis are conducted on modified IEEE RTS-79 systems. The results verify the effectiveness of the technologies to realize the potential of existing power system infrastructures so as to relieve the stress caused by the PHEVs load, increase the available penetration of PHEVs and promote the bulk power system reliability which is meaningful for power system planning and risk management.

Optimal strategies of energy management integrated with transmission control for a hybrid electric vehicle using dynamic particle swarm optimization

In this study, dynamic particle swarm optimization was used to develop optimal strategies for two-variable energy management and gear-shifting in hybrid electric vehicles. To perform simulations, six mathematical subsystems, including a spark ignition engine, lithium battery module, traction motor, automatic manual transmission, longitudinal vehicle dynamics, and driver model were constructed. The control strategies for energy management and gear-shifting were modeled using a dynamic particle swarm optimization strategy with three inputs (i.e., battery state-of-charge, engine speed, and demanded torque) and two outputs (i.e., power-split ratio and gear number). The optimization process was divided into six sequential steps. Four different cases were selected for comparison. The results demonstrate that compared with the baseline case, the improvements in the equivalent fuel and energy consumed in the two-variable optimization case were 30.75% and 59.55%, respectively, during the standard European city driving cycle, and 20.54% and 46.95%, respectively, during the federal test procedure cycle. Consequently, the proposed optimization strategy brought superior performance when applied to hybrid energy management and transmission control. The results of a hardware-in-the-loop verification also confirm the effective performance of proposed the online control. In future work, tests using a real vehicle will be conducted.

Power Allocation in PON-OCDMA with Improved Chaos Particle Swarm Optimization

In this work, it is investigated an improved chaos particle swarm optimization (IC-PSO) scheme to refine the quality of the algorithm solutions regarding to solve the optimal power allocation in next generation passive optical networks (NG-PON)s. The proposed IC-PSO scheme utilizes the Beta distribution instead of uniform distribution of the traditional PSO. A factor of damping based in the chaotic logistic map related to the updating of the best global value is successful introduced. The numerical results corroborate the best relation between the performance-complexity tradeoff and the quality of the algorithm solutions for the proposed IC-PSO when compared with the classical PSO power allocation scheme.

Predicting host CPU utilization in the cloud using evolutionary neural networks

The Infrastructure as a Service (IaaS) platform in cloud computing provides resources as a service from a pool of compute, network, and storage resources. One of the major challenges facing cloud computing is to predict the usage of these resources in real time. By knowing future demands, cloud data centres can dynamically scale resources to decrease energy consumption while maintaining a high quality of service. However cloud resource consumption is ever changing, making it difficult for accurate predictions to be produced. This motivates the research presented in this paper which aims to predict in advance the level of CPU consumption of a host. This research implements evolutionary Neural Networks (NN), a powerful machine learning method, to make these predictions. A number of state of the art swarm and evolutionary optimization algorithms are implemented to train the neural networks to predict host utilization: Particle Swarm Optimization (PSO), Differential Evolution (DE) and Covariance Matrix Adaptation Evolutionary Strategy (CMA-ES). The results of this research demonstrate that CMA-ES converges faster to a better solution on the training data. However when evaluated on the test data, DE performs statistically equal to CMA-ES. The results also demonstrate that the trained networks are still accurate when applied to CPU utilization data from different hosts with no further training needed. When evaluated to predict multiple steps into the future, the accuracy of the network understandably decreases but still performs well on average.

Coordinated trajectory planning of dual-arm space robot using constrained particle swarm optimization

Application of the multi-arm space robot will be more effective than single arm especially when the target is tumbling. This paper investigates the application of particle swarm optimization (PSO) strategy to coordinated trajectory planning of the dual-arm space robot in free-floating mode. In order to overcome the dynamics singularities issue, the direct kinematics equations in conjunction with constrained PSO are employed for coordinated trajectory planning of dual-arm space robot. The joint trajectories are parametrized with Bézier curve to simplify the calculation. Constrained PSO scheme with adaptive inertia weight is implemented to find the optimal solution of joint trajectories while specific objectives and imposed constraints are satisfied. The proposed method is not sensitive to the singularity issue due to the application of forward kinematic equations. Simulation results are presented for coordinated trajectory planning of two kinematically redundant manipulators mounted on a free-floating spacecraft and demonstrate the effectiveness of the proposed method.

Frequency-domain Tuning of Robust Fixed-structure Controllers via Quantum-behaved Particle Swarm Optimizer with Cyclic Neighborhood Topology

This paper presents a constrained particle swarm optimization (PSO) algorithm with a cyclic neighborhood topology inspired by the quantum behavior of particles, and describes its application to the frequency-domain tuning of robust fixed-structure controllers. Two main methodologies for improving the exploration and exploitation performance of the PSO framework are described. First, a PSO scheme with a neighborhood structure based on a cyclic network topology is presented. This scheme enhances the exploration ability of the swarm and effectively reduces the probability of premature convergence to local optima. Second, the above PSO scheme is hybridized using a distributed quantum-principle-based offspring creation mechanism. Such a hybridized PSO framework enables neighboring particles to concentrate the search around the region covered by those particles to refine the candidate solution. A frequency-domain tuning method for fixed-structure controllers is then demonstrated. This method guarantees certain preassigned performance specifications based on the developed PSO technique. A typical numerical example is considered, and the results clearly demonstrate that the proposed PSO scheme provides a novel and powerful impetus with remarkable reliability for robust fixed-structure controller syntheses. Further, an experiment was conducted on a magnetic levitation system to compare the proposed strategy with a well-known frequency-domain tuning method implemented in the MATLAB tool for Structured H∞ Synthesis. The comparative experimental results validate the effectiveness of the proposed tuning strategy in practical applications.

Fuel loading pattern optimization of a pressurized water reactor by varying internal weights-based particle swarm optimization

Fuel reload pattern optimization is essential for attaining maximum fuel burnup for minimization of generation cost while minimizing power peaking factor (PPF). The aim of this work is to carry out detailed assessment of particle swarm optimization (PSO) in the context of fuel reload pattern search. With astronomically large number of possible loading patterns, the main constraints are limiting local power peaking factor, fixed number of assemblies, fixed fuel enrichment, and burnable poison rods. In this work, initial loading pattern of fixed batches of fuel assemblies is optimized by using particle swarm optimization technique employing novel feature of varying inertial weights with the objective function to obtain both flat power profile and cycle keff > 1. For neutronics calculation, PSU-LEOPARD-generated assembly depletion-dependent group-constant-based ADD files are used. The assembly data description file generated by PSU-LEOPARD is used as input cross-section library to MCRAC code, which computes normalized power profile of all fuel assemblies of PWR nuclear reactor core. The standard PSO with varying inertial weights is then employed to avoid trapping in local minima. A series of experiments have been conducted to obtain near-optimal converged fuel-loading pattern of 300 MWe PWR Chashma reactor. The optimized loading pattern is found in good agreement with results found in literature. Hybrid scheme of PSO with simulated annealing has also been implemented and resulted in faster convergence.

A new structural damage detection strategy of hybrid PSO with Monte Carlo simulations and experimental verifications

Structural damage detection (SDD) is originally an optimization problem by minimizing discrepancy between measured and calculated data of structures. In most cases, natural frequencies and mode shapes are selected to define objective functions for SDD. In this study, a new SDD strategy of hybrid particle swarm optimization (HPSO) is proposed and its availability solution to SDD is studied via Monte Carlo simulations. First of all, PSO algorithms with different parameters are tested via Monte Carlo simulations to decide which combination of parameters is more beneficial for SDD. After that, a powerful local searching Nelder-Mead algorithm is embedded into the PSO with a new strategy, which confirms helpful to enhance the PSO global searching ability numerically and experimentally. Simply-supported beams with 10 and 20 finite elements are simulated respectively which prove our proposed method to be effective in SDD. Further, a series of experiments of a box-section steel beam are designed and fabricated in laboratory. Structural frequencies and mode shapes are measured under different crack damage patterns. The experimental verifications confirm the applicability of the proposed new SDD strategy. The SDD results of the experimental beam also show the outperformance of the proposed new SDD strategy. Some relative discussions are also described in detail.

Opposition-Based Hybrid Strategy for Particle Swarm Optimization in Noisy Environments

Particle swarm optimization (PSO) is a population-based algorithm designed to tackle various optimization problems. However, its performance deteriorates significantly when optimization problems are subjected to noise. PSO is strongly influenced by its previous best particles and global best one, which may lead to premature convergence and fall into local optima. This also holds true for various PSO variants dealing with optimization problems in noisy environments. Opposition-based learning (OBL) is well-known for its ability to increase population diversity. In this paper, we propose hybrid PSO algorithms that introduce OBL into PSO variants for improving the latter’s performance. The proposed hybrid algorithms employ probabilistic OBL for a swarm. In contrast to other integrations of PSO and OBL, we select the top fittest particles from the current swarm and its opposite swarm to improve the entire swarm’s fitness. Experiments on 20 benchmark functions subject to different levels of noise show that the proposed hybrid PSO algorithms outperform their counterpart PSO variants as well as composite differential evolution in most cases.

Web-Spam Features Selection Using CFS-PSO

This paper proposes Swarm based hybrid technique CFS-PSO, which combines the characteristics of Correlation Based Feature Selection Technique (CFS) and Particle Swarm Optimization (PSO) strategy. PSO is an optimization approach motivated by swarm conduct which uses the real-number randomness & the global communication among the swarm particles. Feature selection (pre-processing technique) is very crucial part of Data Mining & Machine Learning.The aims of feature selection includes building of simpler & more logical models and improving the performance in terms of reducing the time to build the learning model and increasing the accuracy. We assess the performance of CFS-PSO on WEBSPAM-UK2006 with five classifiers. Experimental results show reduction in original features & increasing the F-measure upto 88% & 45.83% respectively.

Simultaneous Identification of Model Structure and the Associated Parameters for Linear Systems Based on Particle Swarm Optimization

In this study, a novel easy‐to‐use meta‐heuristic method for simultaneous identification of model structure and the associated parameters for linear systems is developed. This is achieved via a constrained multidimensional particle swarm optimization (PSO) mechanism developed by hybridizing two main methodologies: one for negating the limit for fixing the particle’s dimensions within the PSO process and another for enhancing the exploration ability of the particles by adopting a cyclic neighborhood topology of the swarm. This optimizer consecutively searches the dimensional optimum of particles and then the positional optimum in the search space, whose dimension is specified by the explored optimal dimension. The dimensional optimum provides the optimal model structure, while the positional optimum provides the optimal model parameters. Typical numerical examples are considered for evaluation purposes, which clearly demonstrate that the proposed PSO scheme provides novel and powerful impetus with remarkable reliability toward simultaneous identification of model structure and unknown model parameters. Furthermore, identification experiments are conducted on a magnetic levitation system and a robotic manipulator with joint flexibility to demonstrate the effectiveness of the proposed strategy in practical applications.

Magnetic Flux Leakage Signal Inversion Based on Improved Efficient Population Utilization Strategy for Particle Swarm Optimization

In this paper, an improved efficient population utilization strategy for particle swarm optimization (IEPUS-PSO) for high dimension problem is proposed to estimate defect profile from magnetic flux leakage (MFL) signals. In the IEPUS-PSO, a mutation probability is proposed to distinguish local version and global version in particle change model and a self-adapted mutation operator, which is used to update the particles’ positions randomly, is introduced into EPUS-PSO. The IEPUS-PSO- based inversing technique is used to estimate the defect profiles. The estimated defect profiles of simulation signals demonstrate that the inversing technique based on the IEPUS-PSO outperforms the one based on EPUS-PSO. The results estimated from real MFL signals by the IEPUS-PSO-based inversing technique indicate that the algorithm is capable of decreasing the computation time. The results show that the IEPUS-PSO-based inversing technique could improve the reconstruction precision by two orders of magnitude for the MFL simulation signals, and for the real MFL signals, the computation time is reduced by about 30% nearly under the same reconstruction precision.

A novel intelligent diagnosis method using optimal LS-SVM with improved PSO algorithm

Aiming at the problem that the most existing fault diagnosis methods could not effectively recognize the early faults in the rotating machinery, the empirical mode decomposition, fuzzy information entropy, improved particle swarm optimization algorithm and least squares support vector machines are introduced into the fault diagnosis to propose a novel intelligent diagnosis method, which is applied to diagnose the faults of the motor bearing in this paper. In the proposed method, the vibration signal is decomposed into a set of intrinsic mode functions (IMFs) by using empirical mode decomposition method. The fuzzy information entropy values of IMFs are calculated to reveal the intrinsic characteristics of the vibration signal and considered as feature vectors. Then the diversity mutation strategy, neighborhood mutation strategy, learning factor strategy and inertia weight strategy for basic particle swarm optimization (PSO) algorithm are used to propose an improved PSO algorithm. The improved PSO algorithm is used to optimize the parameters of least squares support vector machines (LS-SVM) in order to construct an optimal LS-SVM classifier, which is used to classify the fault. Finally, the proposed fault diagnosis method is fully evaluated by experiments and comparative studies for motor bearing. The experiment results indicate that the fuzzy information entropy can accurately and more completely extract the characteristics of the vibration signal. The improved PSO algorithm can effectively improve the classification accuracy of LS-SVM, and the proposed fault diagnosis method outperforms the other mentioned methods in this paper and published in the literature. It provides a new method for fault diagnosis of rotating machinery.

Ab initio calculations on structural and electronic transport properties of six-atom GaN clusters

The structural and electronic transport properties of GaxNy ([Formula: see text]) clusters are investigated in the framework of density functional theory (DFT). To get their most stable structures, a strategy of particle swarm optimization (PSO) algorithm is adopted. It is found that the most stable cluster’s binding energy and HOMO–LUMO gap energy decrease with Ga atom’s number in cluster increasing. The electronic transport properties of the clusters connected with two Al(100) electrodes are obtained by a method of combining nonequilibrium Green’s function (NEGF) with DFT. Equilibrium conductance of all six-atom GaN cluster is low (less than 0.65 G0), and Ga2N4 has the highest one (0.635 G0). Significant negative differential resistance (NDR) phenomenon is observed in configurations with cluster Ga2N4, Ga3N3 and Ga5N1, and these three clusters have almost the same current value in voltage region from 0.8 V to 1.3 V.

Attribute weight computation in a decision making problem by particle swarm optimization

The objective of this paper is to introduce a method for computing weights of attributes in a decision making problem under intuitionistic fuzzy environment. Many weight generation methods exist in the literature under intuitionistic fuzzy setting, but they have some limitations which can be pointed out as: the entropy measures used in entropy weight methods are invalid in many situations and also there are lots of entropy formulae for intuitionistic fuzzy sets, which will be better to use, and thus a confusion may arise; the other weight generation methods may lose some information since it needs to transform the intuitionistic fuzzy decision matrix into an interval-valued decision matrix. This conversion distorts experts original opinions. In this point of view, to overcome these demerits, we develop a weight generation method without changing the original decision information. The proposed method maximizes the average degree of satisfiability and minimizes the average degree of non-satisfiability of each alternative over a set of attributes, simultaneously. This leads to formulate a multi-objective programming problem (MOPP) to compute the final comprehensive value for each alternative. The scenario of an MOPP itself is subjective and can be modeled by fuzzy decision making problem due to the conflicting objectives and the way of human choice on conflict resolution. This problem is solved by using particle swarm optimization scheme, and the evaluation procedure is illustrated by means of a numerical example. This work has also justified the proposed approach by analyzing a comparative study.

A novel optimized hybrid fuzzy logic intelligent PID controller for an interconnected multi-area power system with physical constraints and boiler dynamics

In the fast developing world nowadays, load frequency control (LFC) is considered to be a most significant role for providing the power supply with good quality in the power system. To deliver a reliable power, LFC system requires highly competent and intelligent control technique. Hence, in this article, a novel hybrid fuzzy logic intelligent proportional-integral-derivative (FLiPID) controller has been proposed for LFC of interconnected multi-area power systems. A four-area interconnected thermal power system incorporated with physical constraints and boiler dynamics is considered and the adjustable parameters of the FLiPID controller are optimized using particle swarm optimization (PSO) scheme employing an integral square error (ISE) criterion. The proposed method has been established to enhance the power system performances as well as to reduce the oscillations of uncertainties due to variations in the system parameters and load perturbations. The supremacy of the suggested method is demonstrated by comparing the simulation results with some recently reported heuristic methods such as fuzzy logic proportional-integral (FLPI) and intelligent proportional-integral-derivative (PID) controllers for the same electrical power system. the investigations showed that the FLiPID controller provides a better dynamic performance and outperform compared to the other approaches in terms of the settling time, and minimum undershoots of the frequency as well as tie-line power flow deviations following a perturbation, in addition to perform appropriate settlement of integral absolute error (IAE). Finally, the sensitivity analysis of the plant is inspected by varying the system parameters and operating load conditions from their nominal values. It is observed that the suggested controller based optimization algorithm is robust and perform satisfactorily with the variations in operating load condition, system parameters and load pattern.

New mathematical models and a hybrid Grouping Evolution Strategy algorithm for optimal helicopter routing and crew pickup and delivery

Most jacket installations located in offshore regions require daily transportation of crew with helicopter, which imposes huge operational costs on oil and gas companies. In order to save on crew exchange costs, the daily flight schedule must be planned in a way that all constraints related to origin-destination itinerary, sortieing, flight length, capacity and crew member’s arrival time to the onshore base be met and the finish time of the final sortie be minimized. For this helicopter routing and crew exchange problem, two different mathematical models are developed which differ in the way the nodes and arcs are defined, and accordingly, the way the decision variables are defined. These models were solved to optimality for test problems up to 11 jackets and 28 crew members, and for the real case of South Pars Gas Field, the world's largest gas field, up to 11 jackets and 46 passengers. Considering the future extensions of the South Pars Gas Field and for problems of larger sizes, given the nature of the problem which is a grouping problem, a new algorithm, namely the hybrid Grouping Evolution Strategy (HGES), is proposed which works based on the composition of the flight sorties rather than working with each nodes in isolation. Several heuristics and a local search algorithm are incorporated in the body of GES to enhance its performance. Extensive computational experiments demonstrate that HGES can produce good solutions in an acceptable amount of time which is fit for operational use, as compared with particle swarm optimization (PSO) and pure Grouping Evolution Strategy algorithm. A GUI has also been designed to eliminate the complexity and inefficiency of the current method which is on “by-eye” basis.