The Improvement And Research Of Several Algorithms About The Global Optimization

A lot of problems of natural science and social science can boil down to global optimization problems which often discovered in the field of economic planning models, finance, traffic transportation, engineering design and so on. Efficient global optimization methods affect the development of these subjects. Owing to there being many local optimal solutions which are different from global ones, we can only get the local optimal solutions using the methods of traditional nonlinear programming, therefore we can't receive the global optimal solution successfully. As the global optimization is applied importantly in many fields, the global optimization comes under comprehensive attention. With the speedy development of computer and the hard work of scientists,the theoretic analysis and computational methods on optimization have been highly improved.This paper mainly research the particle swarm optimization and the filled function optimization algorithm. The innovation of this paper as follows:About the particle swarm optimization, based on the analyses with the present particle swarm optimization algorithm in detail, in order to improve the multiplicity, the solution precision and the algorithm efficiency, first of all, bringing forward judgement mechanism about precocity, adjusting inertia weight using the degree of precocity and the individual value adaptively, getting the sequence of chaos making use of the logic function, a adaptive particle swarm global optimization algorithm based on chaos is proposed. Secondly using a strategy in which the velocity is not dependent on the size of distance between the individual and the optimal particle but only dependent on its direction, an adaptive scheme is adopted to adjust the magnitude of the velocity resiliently, a resilient particle swarm global optimization algorithm based on chaos is proposed. Simultaneity, combining the classical gradient method with the resilient modification of the velocity, complementarity each other, a resilient particle swarm global optimization algorithm based on gradient method is proposed. Simulation results show these algorithms can improve the algorithm's performance effectively as well as make the algorithm practical. Both the quality of global convergence and the algorithm efficiency of the algorithms are improved. About the filled function optimization algorithm. A kind of parameter-free filled function is proposed. The filled properties are discussed and proved. Then we improve it using the chaos theory and propose a global optimization algorithm based on chaos and filled function. Numerical experiments show that the method is effective.The last part concludes the research in this paper and presents the future research on the relative global optimization algorithms.