Particle Swarm Optimization

Particle Swarm Optimization is a popular heuristic search algorithm which is inspired by the social learning of birds or fishes. It is a swarm intelligence technique for optimization developed by Eberhart and Kennedy [1] in 1995. Inertia weight is an important parameter in PSO, which significantly affects the convergence and exploration-exploitation trade-off in PSO process. Since inception of Inertia Weight in PSO, a large number of variations of Inertia Weight strategy have been proposed. In order to propose one or more ...

Abstract: This paper proves that non-convex quadratically constrained quadratic programs have an exact semidefinite relaxation when their underlying graph is acyclic, provided the constraint set satisfies a certain technical condition. When the condition is not satisfied, we propose a heuristic to obtain a feasible point starting from a solution of the relaxed problem. These methods are then demonstrated to provide exact solutions to a richer class of optimal power flow problems than previously solved.

In Computer Vision tracking of moving object in a real life scene is considered as a very challenging problem. Many factors such as illumination of light, noise, occlusion, sudden start and stop of moving object, shading etc. are associated with the task of tracking an object in a real scene under dynamic background(varying background) as well as static background(fixed background). In this paper we present a new object tracking algorithm using Quantum particle swarm optimization (QPSO).QPSO essentially deals with the dominant points of an object to be tracked. The novelty of our approach is that QPSO with a set of dominant points as stated above can be successfully applied for object tracking with both variable background as well as static background. Thus a unified attempt has been made for object tracking in a real life scene. In our approach we first detect the dominants points of objects to be tracked, then a group of particles form a swarm are initialized randomly over the image search space. It start searching the curvature connected between two consecutive dominant points until they satisfy a fitness criterion. As the curvature moves and the curvature movement is tracked by the swarm throughout the video and eventually when the swarm reaches optimal solution, a bounding box drawn based on particles final position. Experimental results demonstrate that this proposed QPSO based method works efficiently and effectively for object tracking in both dynamic and static environments. A comparative study shows that QPSO based tracking algorithm, on an average, runs 90% faster than PSO based tracking algorithm.

Burden distribution is one of the most important operations, and also an important upper regulation in blast furnace (BF) iron-making process. Burden distribution output behaviors (BDOB) at the throat of BF is a 3-dimensional spatial distribution produced by burden distribution matrix (BDM), including burden surface output shape (BSOS) and material layer initial thickness distribution (MLITD). Due to the lack of effective model to describe the complex input-output relations, BDM optimization and adjustment is carried out by experienced foremen. Focusing on this practical challenge, this work studies complex burden distribution input-output relations, and gives a description of expected MLITD under specific integral constraint on the basis of engineering practice. Furthermore, according to the decision variables in different number fields, this work studies optimization of BDM with expected MLITD, and proposes a multi-mode based particle swarm optimization (PSO) procedure for optimization of decision variables. Finally, experiments using industrial data show that the proposed model is effective, and optimized BDM calculated by this multi-model based PSO method can be used for expected distribution tracking.

The application of soft computing in decision support system in disease prediction is one of the emerging interdisciplinary research areas in the field of computer science. Machine learning algorithms plays an important role in risk prediction of diseases. Attribute selection among the dataset is the key factor that influences prediction accuracy. Mathew's correlation coefficient performance metric is also taken into account. Particle swarm optimization algorithm is altered and applied for performing attribute selection. Improved fuzzy artificial neural network classifier performs the prediction task.

—Particle Swarm Optimization (PSO) is a member of the Swarm Intelligence family of population-based optimizers. The Particle Swarm Optimization (PSO) algorithm, as one of the latest algorithms inspired from the nature, was introduced in the mid-1990s. PSO gets better results in a faster, cheaper way compared with other methods. In this paper introduction of PSO algorithm is given, which is followed by the survey of major PSO based algorithms. The PSO have several advantages and applications which are given in this paper. Accordingly, various methods, advantages and application are discussed of the PSO algorithm.

This paper aims to model a consumer goods cross-docking problem, which is solved using metaheuristics to minimize makespan and determine the capacity in terms of inbound and outbound docks. The consumer-goods cross-docking problem is represented through inbound and outbound docks, customer orders (products to be delivered to customers), and metaheuristics as a solution method. Simulated annealing (SA) and particle swarm optimization (PSO) are implemented to solve the cross-docking problem. Based on the results of statistical analysis, it was identified that the two-way interaction effect between inbound and outbound docks, outbound docks and items, and items and metaheuristics are the most statistically significant on the response variable. The best solution provides the minimum makespan of 973.42 minutes considering nine inbound docks and twelve outbound docks. However, this study detected that the combination of six inbound docks and nine outbound docks represents the most efficie...

Structural optimization of Lennard-Jones (LJ) clusters is a classical NP problem. There are many local minima locating near the global minimum, and the local optima number is increased exponentially. Social cognitive optimization algorithm (SCOA) is a new swarm intelligent technique by simulating the human society. However, its local search capability is still weak. Therefore, in this paper, a novel local search strategy, Limited memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) is employed to enhance the exploitation capability of SCOA. Simulation results show the proposed hybrid algorithm has successfully found the lowest-energy structures of the LJ2-LJ5, LJ7-LJ9 and LJ12 comparing with PSO and the standard SCOA.

Two modern optimization methods including Particle Swarm Optimization and Differential Evolution are compared on twelve constrained nonlinear test functions. Generally, the results show that Differential Evolution is better than Particle Swarm Optimization in terms of high-quality solutions, running time and robustness.

Particle swarm optimization (PSO) is a population based optimization technique developed by Dr. Eberhart and Dr. Kennedy in 1995, inspired by social behavior of bird flocking or fish schooling. It is a computational method that optimizes a problem by iteratively trying to improve a candidate solution with regard to a given measure of quality. It is an optimization of continuous nonlinear functions that are used to find the best solution in problem space.. In this paper, this nature inspired behaviour has been considered to optimise the neural network i.e Radial Basis Function by PSO. The RBF is a Neural Network. It is based on moving or directed along the radius. Radial Basis Function emerged as a variant of artificial neural networks in late 80's. RBF's are embedded as two layer neural network, where each hidden unit implements a radial activated function. The output units implement a weighted sum of hidden unit outputs. In order to use a Radial Basis Function Network we ne...

In recent years, the reconfiguration of the distribution network has been proclaimed as a method for realizing power savings, with virtually zero cost. The current trend is to design distribution networks with a mesh network structure, but to operate them radially. This is achieved by the establishment of an appropriate number of switchable branches which allow the realization of a radial configuration capable of supplying all of the normal defects in the box of permanent defect. The purpose of this article is to find an optimal reconfiguration using a Meta heuristic method, namely the particle swarm optimization method (PSO), to reduce active losses and voltage deviations by taking into account certain technical constraints. The validity of this method is tested on a 33-IEEE test network and the results obtained are compared with the results of basic load flow.

This thesis explores the theme of improving, in terms of speed and regularity, the road-operation of public transport systems, through control strategies in real time.
Specifically in this work are highlighted intervention strategies that falls in the two categories of "Vehicle Priority" and "Vehicle Holding", potentially conflicting the one with the other.
The analysis of scientific literature and best practice has shown that the two intervention strategies, leading the first at maximizing the commercial speed and the second to the maximization of regularity of service, are currently considered one independently from the other, both in the field of practical applications than in the one of theoretical speculation.
In this work is presented a new unified approach that reconciles both the strategies of intervention; moreover a method is introduced, implemented into a prototype software, for the calculation of the optimal strategy of the vehicle operating through regulation of fixed control points (eg. Signals road traffic lights) to allow a joint pursuit of speed maximization objectives and regularity of service.
The optimization method here formulated and tested is based on the combined use of (meta) heuristics (Genetic Algorythm, PSO and Hill Climbing) and traffic simulation methods able to reproduce and predict the temporal trajectories of individual public transport vehicles. The short-term prediction capabilities inherent in the use of simulation tools allows to generate solutions to the regulation problem which are extended "forward" both in the road space and time, thus allowing an approach to regulation that is both anticipatory and non-local. These characteristics allow in general to achieve road capacity relocation between neighboring intersections thus enabling, where possible, to reduce the competition between different modes of transport in the use of road capacity.
In this work are finally presented the results of a series of tests carried out both in the case of realizing the control through a normal signal pattern with ripetitive cycle and phases for each signal, both in the hypothesis less restrictive of red and green times generally variables during time for each signal.
The calculation effort of the prototype also suggest the possibility, through further developments, to use this method for real-time applications.
This PHD thesis is written in Italian Language.

ABSTRACT: This Paper presents a comparative study of Genetic Algorithm method (GA) and Particle swarm optimization (PSO) method to determine the optimal proportional-integral-derivative (PID) controller parameters, for load frequency control in a single area power system. Comparing with conventional Proportional–Integral (PI) method and the proposed PSO the performance of the controller is improved for the step input in Load frequency control. This paper presents a comparative study of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) method to determine the optimal parameter of proportional integral-derivative (PID) controller parameters, for load frequency control in single area power system. For this application, MATLAB –Simulink software is used.

– The Android Operating System covers almost 80% of the smartphones market. However, being so successful and popular amongst its users, Android Platform may misbehave in spite of all the rigorous security mechanisms it can afford. In our preceding work, we tried to state the analysis of malwares and Android malwares, how device application may be affected due to the malwares injected along with the Android Application Package. Static and Dynamic analysis incorporates the features extracted to define the proper functioning of the application. In this paper a packet capture cum classifier framework has been projected which will be a security mechanism for Android and is on the verge of implementation. In near future, our mechanism will sniff and capture the network transportation and determine the suspiciousness of the packets exchanged between Android devices or on the network and catalogue them as benign or suspicious/malicious applications. Our later work would be a framework for Android applications for recognition of Android packet misbehavior grounded on the features concluded from the procedure of analysis and thereby categorize them as benign or suspicious/malicious.