Research And Application Of Firefly Algorism Based On Chaos Optimization And VFSA

The field of intelligent optimization algorithms is the field which is newly arisen in recent years and causes the extensive attraction of scholars. The swarm intelligent optimization algorithm can be taken as the key research direction in the field and is a novel bionic algorithm generated by simulating the process of foraging and information interaction of insects and under the observation and research of swarm behaviors of biology in the nature. In addition, the swarm intelligent optimization algorithm is expressed through a mathematical version. At present, swarm intelligent optimization algorithms, such as the Particle Swarm Optimization, the Ant Colony Optimization and the Wolf Pack Algorithm, are used frequently by scholars.Xin-she Yang, the scholar of Cambridge finds that fireflies have the behavior of mutual attraction and the characteristic of glowing by observing the characteristics of fireflies and proposes the Firefly Algorithm(FA) in the year of 2008. The algorithm is a heuristic optimization algorithm proposed after the particle swarm algorithm and the ant colony optimization and has the main advantages of being relatively simple in concept, few in applied parameters and convenient to use and achieve.Based on existing research results of the firefly algorithm, furhter analysis verification is conducted in the text, and the firefly algorithm based on chaos optimization and VFSA is designed. A novel argument and nalyssi method is adopted for conducting performance comparison analysis on the novel algorithm and the traditional algorithm. Specific content is shown below:(1) In the text, comparison analysis is conducted on the traditional firefly algorithm, the algorithm mechanism is researched, and the reasons for the phenomenon of slow convergence and the occurenace of the locally optimal solution in the later iteration stage of the algorithm is proposed.(2) Comparison analysis is conducted on the existing improved firefly algorithm. The theory of optimization layer upon layer is adopted for delicate optimization on all the operation steps of the firefly algorithm. The chaos theory, andvarious swarm study mechanism ideals, the very fast simulated annealing algorithm, a self-adaptation step-length mechanism and the like are used in the initial stage, the iteration stage, the evolution optimization stage and other stages. Different optimization methods are adopted in different stages. The elolution advantages of the original algrithm are maintained as much as possible, and meamwhile the global optimization performance of the firefly algorithm is obviously improved.(3) Specific to the defects of the current firefly algorithm, a representative function is selected for conducting an experimental test on the algorithm after the traditional algorithm is improved. It is show through simulation experiment results that, the performance of the optimized firefly algorithm is greatly improved.(4) The novel algorithm is applied to the question setting paper constructing algorithm, the question setting process of the paper constructing algorithm is improved, and scientificity and reasonability of the question setting process of the paper constructing algorithm are further optimized. Simulation tests show that an ideal effect is achieved for the application of the improved algorithm.