Earticle

Task scheduling is one of the most important issues in heterogeneous environments when high efficiency is required. Because task scheduling is a Nondeterministic Polynomial (NP)-hard problem, many evolutionary algorithms have been adopted to solve this problem. Since the convergence speed of solutions in population-based algorithms is low, they are integrated with local search algorithms. Thus, in this paper, to optimize the task scheduling makespan, a hybrid particle swarm optimization and hill climbing algorithm is proposed. The experimental results on random and scientific Directed Acyclic Graph (DAG) showed that the proposed algorithm performs effectively in terms of the makespan compared to the current well-known heuristic and particle swarm optimization algorithms.

In this paper, the particle swarm optimization (PSO) algorithm is adopted to design a modified ring-slot type patch rectenna with a resonance frequency of 2.45GHz. In order to accomplish minimization of dimensions and circular polarization (CP) and harmonic suppression, axial direction slits and side-cuts are added to the patch of the ring. The PSO manipulated this kind of multi-dimensional problem very well, and as a result, the designed rectenna shows a desirable performance of return loss of 21.36dB and axial ratio of 2.92dB at the frequency of 2.45GHz with compact sizing.

In this paper, a study was conducted on the methodology for learning the parameters of a neural network using an evolutionary algorithm such as a particle swarm optimization algorithm. The possibility of using the particle swarm optimization algorithm for deep learning was analyzed, and various methods were considered for practical use.

최근에 진화연산방법은 건설 분야의 여러 문제에 성공적으로 적용되어 왔으며 대표적인 방법으로는 유전자 알고리즘과 Particle Swarm Optimization이 있다. 유전자 알고리즘은 생물유전학과 자연선택이론에 바탕을 둔 병렬적인 최적화 탐색방법이며 PSO는 새나 물고리 무리의 집단적인 행동에서 영감을 얻은 진화형 통계탐색방법이다. 두 방법은 많은 유사한 점을 공유하고 있지만, 어느 한 방법이 다른 방법도 효율과 효과 면에서 항상 뛰어나다는 것을 증명할 수 없다. 이에 본 연구는 유전자 알고리즘과 PSO방법 중 어떤 방법이 공동주택 사업초기단계에서의 공사비 예측문제에서 더 뛰어난 예측성능을 보여주는 지를 조사하였다. 한국건설감리협회의 감리자 입찰공고를 통해 수집한 219개의 공동주택 사업자료를 가지고 NeuroShell Predictor 소프트웨어에서 두 방법을 이용한 예측모델을 제시하였고 모델의 예측성능을 점검하였다. 소프트웨어에서 제공하는 높은 R-squared 및 상관계수값을 통해서 두 예측모델의 성능은 우수한 것으로 판명되었으나 PSO 기반 예측모델이 유전자 알고리즘 기반 모델보다 평가시험그룹에서의 예측정확도면에서 약간 더 우수한 것으로 나타났다.

In recent years, evolutionary computation methods have been successfully applied to various problems in the construction field. Two of these popular methods are Genetic Algorithm(GA) and Particle Swarm Optimization(PSO). GA is a parallel search method for an optimal solution based on the genetics and natural selection and PSO is a population-based stochastic optimization method inspired by social behavior of bird flocking or fish schooling. While two methods share many similarities, one cannot be proven to always outperform the other in terms of efficiency and effectiveness. From this point, this paper examined whether GA or PSO is superior at a problem to predict construction cost on apartment housing projects at the early stage. With 219 apartment housing project data obtained from bidding announcements for construction supervisors in the Korea Construction Consulting Engineers Association, prediction models using two methods were suggested in the NeuroShell Predictor software and prediction performances of two models were checked. From high R-squared and correlation coefficient values provided by the software, performances of two prediction models were found to be good. However, the findings also showed that the PSO-based model is slightly better than the GA-based one in the comparison of prediction accuracy on the testing dataset.

According to the relationship between feature subset and parameters of RBF neural network, in order to improve the intrusion detection accuracy, it proposed an improved particle swarm optimization neural network of network intrusion detection model. Network feature subset and parameters of RBF neural network were regarded as a particle, through collaboration and information exchange between particles to find the optimal feature subset and parameters of RBF neural network, so as to establish the optimal network intrusion detection model, and using KDD Cup 99 data sets to carry out simulation experiment. The simulation results showed that, IPSO-RBF neural network reduced the feature dimensions, and the better parameters of RBF neural network was obtained then, which is a kind of network intrusion detection model with high detection accuracy and high speed.

In the age of Internet, the data transmission and storage got rapid progress, however, data processing and information extraction is still exist many problems to solve. Under the condition of so much data, processing data, get useful information; In cloud computing, big data environment to adopt the method of distributed computing, such a large complex networks, however, requires a simulation environment, for comparison and optimization platform, it can save development costs. Hadoop can evaluate the performance of distributed cloud computing platform, so the Hadoop performance directly affects the evaluation on the performance of the big data cloud computing, which fully show the importance of performance of Hadoop. Algorithm is improved based on Hadoop platform, using the particle swarm optimization algorithm improved the calculation and implementation of the Hadoop platform, so as to improve its ability to execute and compute, the calculation results and analysis show that the proposed scheme is effective.

A major challenge facing cloud computing is virtual resource allocation with dynamic characteristics. Evaluation of a resource allocation strategy using a single aspect can no longer meet the real world demands. We resolve this issue from the perspectives of users and resource providers using a particle swarm algorithm for resource allocation. With this algorithm, we establish an allocation model using the shortest task completion time and the lowest cost as the constraints. The fast convergence rate of the particle swarm algorithm is then used to find the optimal solution for resource allocation. The velocity weight of each particle is self-adaptively adjusted based on the fitness value of each particle, resulting in an improvement in the global optimization and convergence capabilities. Finally, a simulation with the CloudSim platform shows that this algorithm can take into account the completion time and cost, which ensures the minimum cost in the shortest possible time to complete the task to improve resource utilization.

A strategy which using the particle swarm algorithm improved by the artificial fish swarm algorithm to optimize the BP (Back propagation) algorithm was proposed. It can conquer the shortcomings that the convergence rate of BP is too slow and it is easy to fall into local extreme value, and can improve the learning ability of BP neural network. Finally, the improved algorithm has been used to analysis the earthquake prediction. The results of simulation and test show that the optimized algorithm can improve the predicting accuracy of the BP network.

For solving the rotation error problem of the shafting, a three-dimensional motion model was developed, and it turned into a multi-parameter optimization problem. As the multi-parameter optimization problem solving approach, the particle swarm optimization (PSO) was introduced to calculate this problem, and carried out shaft rotary motion trajectory simulation experiments. The PSO algorithm to calculate the rotational error was prepared and developed an analysis software, the correctness of PSO algorithm was also verified. Finally, an example of rotational error measurement experiment by atomic force microscope (AFM) was calculated and obtained the rotational error results.

This research presents novel Particle swarm optimization inspired by gravitational based search method to solve active power dispatch problem in electrical power system planning. The proposed PSO utilizes the operator of social thinking coupled with search capacity of gravity inspired algorithm to formulate and develop technique for active power dispatch problem to satisfy power demand requirements. Optimal scheduling of generators and system constraints to match load demand and losses is successfully done with proposed method. Total operating cost is minimized satisfying various bounds of system with proposed method. Exploration and convergence efficiency are evaluated to checklist the computational efficiency and robustness of the proposed technique. The suggested technique is tested and evaluated on different test systems comprises three, five, six test systems. Test results are compared with other techniques presented in literature .Investigations shows promising results which further benchmark the effectiveness of proposed method to solve complex optimization non linear problems.

Prediction of network public opinion is a complicated prediction featuring poor information, small samples and uncertainty. A prediction model of network public opinion based on grey support vector machine (SVM) is specified to increase prediction accuracy. First, network data are preprocessed by text clustering, hotpot extraction and data aggregation. Then a time series model GM(1,1) is established and SVM is used to modify prediction outcomes of GM(1,1). At last, simulation experiment is conducted to test performance of the model. Simulation results indicate that grey SVM improves the prediction accuracy of network public opinion compared with traditional prediction models. The predictions have certain practical values.

Task scheduling and resource scheduling are the core issue in cloud computing. Pointing at the premature problem in the scheduling algorithm of particle swarm, we propose a scheduling algorithm of cloud task particle swarm based on “fission” mechanism in this paper. The particle in traditional particle swarm algorithm gets “fission” by the new algorithm in appropriate place, to get more kinds of the particles, contributing to the particle swarm diversity, avoiding premature convergence of the swarm. As the experimental result shows that, the algorithm in this paper has faster scheduling efficiency than the FIFO and the PSO, also solves the premature problem in PSO.

Image mining is used to mine knowledge from large image databases. Image segmentation, image compression, image clustering, image classification and image retrieval are significant image mining tasks. Face detection methods are used to identify the similar faces from the large collection of facial images. It has numerous computer vision applications and it has many research challenges such as rotation, scale, pose and illumination variation. Facial expression is defined as the position of the muscles beneath the skin of the face and it is a form of nonverbal communication. Facial expressions are the expression which shows the emotions and different feelings of human beings. Different facial expressions are sad, happy, fear, normal, surprise and angry. In this research work facial expressions are classified by using the optimization algorithms. PSO with LIBSVM algorithm is proposed for facial expression classification and the performance of this algorithm is compared with the existing BAT algorithm. The results of the existing and proposed algorithms are analyzed based on the two performance factors; they are classification accuracy and execution time. From the experimental results, we observed that the proposed PSO with LIBSVM algorithm has produced good results compared to existing BAT algorithm. This work is implemented in MATLAB 7.0.

The thesis mainly studies bi-criteria no-wait flexible flow shop problem, whose optimi-zation objective is to minimize the maximum completion time and the maximum delay time. This problem is NP hard, yet enjoying important theoretical research value, thereby this thesis proposes elite particle swarm optimization (EPSO) to solve bi-criteria no-wait flex-ible flow shop problem. EPSO algorithm applies five modified heuristic algorithms and random methods to generating initial population. Moreover, for the particle personal best, this thesis puts forward elite crossover algorithm, which retains continuous fragments of the identical workpieces among excellent individuals, avoiding the destruction of good continuity between solutions of workpieces. In addition, in order to avoid algorithm into local optimum, this thesis raises double insertion disturbance algorithm to help particles jump out the local optimal state and expand the feasible search range. For the purpose of effectively evaluating algorithm quality, there is a comparison among EPSO algorithm, PSO algorithm and ICA algorithm in simulation experiment that is respectively aimed at small-scale problem and large-scale scheduling problem, the results of which show that the proposed EPSO algorithm, due to better validity and accuracy, is superior to the PSO algorithm and ICA algorithm.

Based on particle swarm optimization (PSO) algorithm and its power system reactive power optimization method to in-depth study and research proposed a new hybrid particle swarm optimization algorithm (HPSO). Algorithm combines the differential evolution algorithm and simulated annealing algorithm and particle swarm optimization algorithm, in particle searching optimal except for tracking individual and global, and tracks produced by particle information difference of the three value. At the same time, when the particle search space of one dimension speed lower than the setting value will be re initialized the dimensional particle velocity and the particle of differential evolution mutation. For the crossover and mutation operations, new solution may be worse than the original solution to, the introduction of simulated annealing algorithm, the metropolis rule in a certain extent accept bad solutions, allows the target function in a certain degree of deterioration, practical calculation is not according to the probability to choose the poor solution, but rather the judgment target function difference is less than allows the target function deterioration range. Hybrid particle swarm optimization algorithm combines the advantages of particle swarm optimization algorithm, differential evolution algorithm and simulated annealing algorithm, to maintain the diversity of particles, has very strong practicability.

Traditional flexible flow shop scheduling cannot adapt to the work processes with existence of parallel machines, and blocks or limits the processes with no-wait constraints. Firstly, according to the problem in NWBFFSSP, which minimizes the maximum time used in the flow shop, an optimal solving model has been designed to realize the flexible flow shop scheduling with multi constraints; besides, for the distribution of machinery is improved, Finally, in the solving process, a real-time release priority strategy has been proposed to determine processing machine for each work piece. Furthermore, a methodology to detect work piece conflicts has been introduced while the conflicts are then eliminated by a kind of right moving strategy based on the maximum difference. The experimental results verify the effectiveness and feasibility of the proposed algorithm.