The main problems for geneticalgorithm (GA) to deal with the complex layout design of satellite module lie in easily trapping into local optimality and large amount of consuming time. To solve these problems, the bee...
详细信息
ISBN:
(纸本)9783038350491
The main problems for geneticalgorithm (GA) to deal with the complex layout design of satellite module lie in easily trapping into local optimality and large amount of consuming time. To solve these problems, the bee evolutionary genetic algorithm (BEGA) and the adaptive geneticalgorithm (AGA) are introduced. The crossover operation of BEGA algorithm effectively reinforces the information exploitation of the geneticalgorithm, and introducing random individuals in BEGA enhance the exploration capability and avoid the premature convergence of BEGA. These two features enable to accelerate the evolution of the algorithm and maintain excellent solutions. At the same time, AGA is adopted to improve the crossover and mutation probability, which enhances the escaping capability from local optimal solution. Finally, satellite module layout design based on Adaptive bee evolutionary genetic algorithm (ABEGA) is proposed. Numerical experiments of the satellite module layout optimization show that: ABEGA outperforms SGA and AGA in terms of the overall layout scheme, enveloping circle radius, the moment of inertia and success rate.
An improved bee evolutionary genetic algorithm is proposed for enhancing the optimization capability of geneticalgorithms. Firstly, the self-adaptive selection operator is used to limit the size of the random populat...
详细信息
An improved bee evolutionary genetic algorithm is proposed for enhancing the optimization capability of geneticalgorithms. Firstly, the self-adaptive selection operator is used to limit the size of the random population in each generation, the strategy could find new space of solution timely, so as to find the optimal solution faster. And then, controlling operation that extends biodiversity of the swarm is introduced to avoid premature convergence. Finally, simulation examples show that the convergence speed and solution precision of the proposed algorithm was increased effectively.
Based on the cobweb theory and bee evolutionary genetic algorithm (BEGA), a multidimensional style intelligent design method is proposed to inherit the brand characteristics and obtain a multidimensional product style...
详细信息
Based on the cobweb theory and bee evolutionary genetic algorithm (BEGA), a multidimensional style intelligent design method is proposed to inherit the brand characteristics and obtain a multidimensional product style during the design process. Cobweb theory is used to characterise the multidimensional development of product genealogy, analyse the product morphological evolution mechanism of past generations, and distinguish morphological genes. The efficient inheritance of the BEGA guarantees the selection, crossover and mutation of genetic features. Based on the cobweb model of a certain brand of tractors, different morphological gene types were distinguished. With a human-computer evaluation, a constraint model that reflects the brand culture and consumer personality requirements was constructed. The BEGA was used to establish a multidimensional style intelligent design method for a certain brand. A tractor is used as an example to explain the process, and a set of experiments are used to verify the feasibility and effectiveness of the method.
The bee immune evolutionaryalgorithm was proposed in order to improve effectively the optimal ability of bee evolutionary genetic algorithm. In the evolutionary process of bee, the algorithm made on immune evolutiona...
详细信息
ISBN:
(纸本)9783037851555
The bee immune evolutionaryalgorithm was proposed in order to improve effectively the optimal ability of bee evolutionary genetic algorithm. In the evolutionary process of bee, the algorithm made on immune evolutionary iteration calculation, generate next-generation population, in the proportions of fitness values for the best individual and second-best individuals in each generation. Because the algorithm takes in the neighborhood of space search as well out the neighborhood of space search for the some optimal individuals, meanwhile, with iterative numbers increase, capability of local search can be strengthened gradually;the bee immune evolutionaryalgorithm can approach the global optimal solution with higher accuracy. The calculated results for typical best functions show that the bee immune evolutionaryalgorithm has better optimal capability and stability.
暂无评论