06-11-2014, 01:12 PM
i am looking for BACTERIAL FORAGING OPTIMIZATION ALGORITHM CODE FOR JOB SHOP SCHEDULING IN MATLAB
|
bacteria foraging optimization source code
|
i am looking for BACTERIAL FORAGING OPTIMIZATION ALGORITHM CODE FOR JOB SHOP SCHEDULING IN MATLAB
06-11-2014, 08:35 PM
Bacterial Foraging Adaptive optimization
Abstract
Bacterial optimizing foraging (BFO) is a newly developed an optimization algorithm inspired by nature, which is based on the foraging behavior of the e. coli bacteria. So far, the BFO had been successfully implemented in some engineering problems associated with simplicity and ease of implementation. However, the BFO has bad convergence behavior of complex optimization problems as compared to other natural methods of optimization. This article first Analyses how the length parameter block BFO control the exploration and exploitation of the search space of perspective directions. It is a variation on the original BFO, Adaptive Optimization Bacterial Foraging So-called (ABFO) application of adaptive strategies to improve the performance of the original foraging BFO. This improvement is achieved by the inclusion of bacterial foraging algorithm to adjust the block variable-length parameter dynamically in order to run the algorithm to balance the exploration/exploitation trade-off. Experiments to compare the performance of two versions of the original ABFO BFO, standard Particle Swarm Optimization (PSO) and real-coded genetic algorithm (GA) on four widely used tests functions.ThePredlagaemyj ABFO Shows significant improvement in performance compared to the original BFO and seems comparable to the PSO and GA.
1. Introduction
In the past several decades, research on optimization of attracted more attention. The most common objective of unconditional optimization can be defined as
where the number of parameters to be optimized.
There are various methods of optimization and algorithms that can be grouped into deterministic and stochastic. Deterministic methods depend on the mathematical nature of the problem. Weaknesses of this method depends on the gradient of local Optima, and inefficiency in a large search space. Stochastic methods is more convenient because they do not depend on the mathematical properties of the function and Hencse Given more suitable for finding optimal global solutions to various types of objective function. How much real-world optimization problems is becoming more complex, using the stochastic method is inevitable.
Nature always bylithe a rich source of ecosystem mechanisms to create artificial computational systems difficult to solve engineering and computer science problems. In optimization, the researchers were inspired by biological stochastic processes to develop some effective methods that mimic certain patterns of conduct or creatures. For example, genetic algorithms (GA) Initially conceived the Netherlands [3] rightly reflect the abstract model of the evolution of Darwin and biological genetics; Ant Colony Optimization (ACO), proposed by Dorigo etc. , Developed on the basis of real behavior of colonies of ants; Particle Swarm Optimization (PSO), proposed by Eberchart and Kennedy [6] and Chen et al. , Clay ideas of social behaviour of birds flocking and fish schooling. Woven algorithms have been found to work better than classical heuristics or gradient-based methods, particularly for the optimization of Undifferentiated, multimodal and complex discrete functions. The natural paradigms have become widely used in many fields.
In recent years, chemotactic (i.e., bacterial foraging behavior) as a rich source of potential engineering applications and computational model is attracting more and more attention. It took a couple of models developed to simulate bacterial foraging and Behavior have been applied Some practical challenges [8-10]. Among them, Optimization Bacterial Foraging (BFO) is a Population-based numerical optimization algorithm presented Passino [10]. BFO-simple But Powerful optimization tool that mimics the behavior of seeking out food bacteria Escherichia coli. So far, the BFO has been successfully implemented in some technical problems such as optimal control, harmonic measurement, reducing loss of transmission, and machine learning. However, experiments with complex and multimodal monitoring show that the original algorithm of BFO has bad convergence Behavior and search performance STI greatly decreases with increasing the search space and the dimension of the problem of complexity.
Abstract
Bacterial optimizing foraging (BFO) is a newly developed an optimization algorithm inspired by nature, which is based on the foraging behavior of the e. coli bacteria. So far, the BFO had been successfully implemented in some engineering problems associated with simplicity and ease of implementation. However, the BFO has bad convergence behavior of complex optimization problems as compared to other natural methods of optimization. This article first Analyses how the length parameter block BFO control the exploration and exploitation of the search space of perspective directions. It is a variation on the original BFO, Adaptive Optimization Bacterial Foraging So-called (ABFO) application of adaptive strategies to improve the performance of the original foraging BFO. This improvement is achieved by the inclusion of bacterial foraging algorithm to adjust the block variable-length parameter dynamically in order to run the algorithm to balance the exploration/exploitation trade-off. Experiments to compare the performance of two versions of the original ABFO BFO, standard Particle Swarm Optimization (PSO) and real-coded genetic algorithm (GA) on four widely used tests functions.ThePredlagaemyj ABFO Shows significant improvement in performance compared to the original BFO and seems comparable to the PSO and GA.
1. Introduction
In the past several decades, research on optimization of attracted more attention. The most common objective of unconditional optimization can be defined as
where the number of parameters to be optimized.
There are various methods of optimization and algorithms that can be grouped into deterministic and stochastic. Deterministic methods depend on the mathematical nature of the problem. Weaknesses of this method depends on the gradient of local Optima, and inefficiency in a large search space. Stochastic methods is more convenient because they do not depend on the mathematical properties of the function and Hencse Given more suitable for finding optimal global solutions to various types of objective function. How much real-world optimization problems is becoming more complex, using the stochastic method is inevitable.
Nature always bylithe a rich source of ecosystem mechanisms to create artificial computational systems difficult to solve engineering and computer science problems. In optimization, the researchers were inspired by biological stochastic processes to develop some effective methods that mimic certain patterns of conduct or creatures. For example, genetic algorithms (GA) Initially conceived the Netherlands [3] rightly reflect the abstract model of the evolution of Darwin and biological genetics; Ant Colony Optimization (ACO), proposed by Dorigo etc. , Developed on the basis of real behavior of colonies of ants; Particle Swarm Optimization (PSO), proposed by Eberchart and Kennedy [6] and Chen et al. , Clay ideas of social behaviour of birds flocking and fish schooling. Woven algorithms have been found to work better than classical heuristics or gradient-based methods, particularly for the optimization of Undifferentiated, multimodal and complex discrete functions. The natural paradigms have become widely used in many fields.
In recent years, chemotactic (i.e., bacterial foraging behavior) as a rich source of potential engineering applications and computational model is attracting more and more attention. It took a couple of models developed to simulate bacterial foraging and Behavior have been applied Some practical challenges [8-10]. Among them, Optimization Bacterial Foraging (BFO) is a Population-based numerical optimization algorithm presented Passino [10]. BFO-simple But Powerful optimization tool that mimics the behavior of seeking out food bacteria Escherichia coli. So far, the BFO has been successfully implemented in some technical problems such as optimal control, harmonic measurement, reducing loss of transmission, and machine learning. However, experiments with complex and multimodal monitoring show that the original algorithm of BFO has bad convergence Behavior and search performance STI greatly decreases with increasing the search space and the dimension of the problem of complexity.
|
« Next Oldest | Next Newest »
|
