Spare optimization models for systems with components connected in series and parallel are presented. The component availability at any time is obtained through Poisson process theory and the availability of the serie...
详细信息
Spare optimization models for systems with components connected in series and parallel are presented. The component availability at any time is obtained through Poisson process theory and the availability of the series-parallel system is written by system reliability analysis method. The objective of the optimization problem is to maximize the availability of the system satisfying the constraint on cost. Genetic algorithms are used to find the optimal amount of spares for each component. Finally, two illustrative examples are given.
Often in real world applications only a small number of labeled data is available while unlabeled data is abundant. Therefore, it is important to make use of unlabeled data. Co-training is a popular semi-supervised le...
详细信息
ISBN:
(纸本)9783642121265
Often in real world applications only a small number of labeled data is available while unlabeled data is abundant. Therefore, it is important to make use of unlabeled data. Co-training is a popular semi-supervised learning technique that uses a small set of labeled data and enough unlabeled data to create more accurate classification models. A key feature for successful co-training is to split the features among more than one view. In this paper we propose new splitting criteria based on the confidence of the views, the diversity of the views, and compare them to random and natural splits. We also examine a previously proposed artificial split that maximizes the independence between the views, and propose a mixed criterion for splitting features based on both the confidence and the independence of the views. Genetic algorithms are used to choose the splits which optimize the independence of the views given the class, the confidence of the views in their predictions, and the diversity of the views. We demonstrate that our proposed splitting criteria improve the performance of co-training.
In order to overcome premature phenomenon of simple genetic algorithms and inability to optimize algorithms with complex constraints, an improved genetic algorithms based on some improved methods is presented in this ...
详细信息
In order to overcome premature phenomenon of simple genetic algorithms and inability to optimize algorithms with complex constraints, an improved genetic algorithms based on some improved methods is presented in this paper and is applied in optimization design of frame structure by adopting adaptive crossover rate and mutation rate, adjusting population size, fitness and penalty function and elitist strategy in the search of GA. The experimental results indicate that the improved genetic algorithm has good performance on the global convergence and that the proposed method can be applied to optimal design of structures with discrete variables.
Aiming at the attribute reduction problems of the information system, a new algorithm based on DNA-sticker model is proposed in this paper. Firstly, the dependency matrix of the information system is established, whic...
详细信息
Aiming at the attribute reduction problems of the information system, a new algorithm based on DNA-sticker model is proposed in this paper. Firstly, the dependency matrix of the information system is established, which could convert the problem of attribute reduction to set cover problem. Then, the all true solution can be achieved by using DNA-sticker model. Finally, through the example the conclusion is drawn that this method is feasible, simple, and effective.
Solving linear variation inequality by traditional numerical iterative algorithm can not satisfy parallel. In this paper, particle swarm optimization is used to solve linear variation inequality, which sufficiently ex...
详细信息
ISBN:
(纸本)9781424458721;9781424458745
Solving linear variation inequality by traditional numerical iterative algorithm can not satisfy parallel. In this paper, particle swarm optimization is used to solve linear variation inequality, which sufficiently exerts the advantage of particle swarm optimization such as group search and global convergence and it satisfies the question of parallel solving linear variation inequality in engineering. Several numerical simulation results show that the algorithm offers an effective way to solve linear variation inequality, high convergence rate, high accuracy and robustness.
The ant colony algorithm is widely applied to optimize the complex problems in many fields with its features of being robust, parallel, flexible, demanding no artificial interference, and accurate. This paper discusse...
详细信息
ISBN:
(纸本)9781424458721;9781424458745
The ant colony algorithm is widely applied to optimize the complex problems in many fields with its features of being robust, parallel, flexible, demanding no artificial interference, and accurate. This paper discusses the application of the colony algorithm in the path search of the earthquake emergency rescue. We first construct a mathematical model for emergency rescue based on the earthquake disasters. Then we propose an improved ant colony algorithm for the rescue path searching optimization according to the specific characteristics of the mode, and the global iterative update strategy is used with limiting the rescue entries. Experiments show that the proposed algorithm can overcome the shortcomings of conventional algorithms such as slow convergence, easily trapped in local optimum, and demonstrating high and flexibility performance.
The matching design problem of the ship-engine-propeller is a non-linear constrained multi-objective optimization problem which is performed based on multiple objectives,such as system efficiency and the life cycle co...
详细信息
The matching design problem of the ship-engine-propeller is a non-linear constrained multi-objective optimization problem which is performed based on multiple objectives,such as system efficiency and the life cycle cost. A multi-objective particle swarm optimization (MOPSO) approach for matching design problem of the ship-engine-propeller problem is presented in this paper.A real matching design problem of the ship-engine-propeller is employed to demonstrate applicability of the proposed *** results indicate that multi-objective particle swarm optimization approach in comparison with most of multi-objective optimizationalgorithms such as the multiple objective genetic algorithm (MOGA) is simple ,efficient and operable.
For improving the intersection traffic capacity and reducing the vehicle emission, the solution that aim at the multi-object optimization was presented by using genetic algorithm (GA), and urban traffic microscopic si...
详细信息
For improving the intersection traffic capacity and reducing the vehicle emission, the solution that aim at the multi-object optimization was presented by using genetic algorithm (GA), and urban traffic microscopic simulation model (UTMSM) combined with GA was developed. The simulation was performed and the result indicated that the optimization method in this paper was effective to get better traffic signal control effects and to improve the environment.
In many problems of classification, the performances of a classifier are often evaluated by a factor (rate of error).the factor is not well adapted for the complex real problems, in particular the problems multiclass....
详细信息
In many problems of classification, the performances of a classifier are often evaluated by a factor (rate of error).the factor is not well adapted for the complex real problems, in particular the problems multiclass. Our contribution consists in adapting an evolutionary method for optimization of this factor. Among the methods of optimization used we chose the method PSO (Particle Swarm optimization) which makes it possible to optimize the performance of classifier SVM (Separating with Vast Margin). The experiments are carried out on corpus TIMIT. The results obtained show that approach PSO-SVM gives a better classification in terms of accuracy even though the execution time is increased.
Current methods for sensor network programming lead developers to cope with not only high-level concerns such as application logic, quality of service, adaptability, reliability, but also low-level mechanism like ad-h...
详细信息
暂无评论