The Research On The Improvement Of Biogeography-based Optimization Algorithm And Its Application

As the science and technology update speedily, more and more attention has been paid to the complex optimization problems in science research and engineering, to which biogeography-based optimization algorithm provides an effective method. Compared with other evolutionary algorithms, BBO algorithm has the characteristics, such as novel mechanism, simple structure, easy to realize. However, in the evolutionary process of the algorithm, the operators may lead to the assimilation of the population easily, which makes the population diversity reduced and the exploration ability of the algorithm weakened. Therefore, the weaknesses about the algorithm should be studied and improved to raise the performance o f the method, so that it can meet the needs of practical application. Plenty of attention has been paid to the methods to calculate the form and position errors of the mechanical parts quickly and accurately with the rapid development of modern manufacturing technology. The essence of the evaluation of the form and position errors is an optimization problem of a complex nonlinear function, which are the intelligent optimization algorithms good at. In this paper, the algorithm is further improved based on the existing research on the problems of the algorithm to make the performance better and some research is done about its application to the evaluation of the form and position errors. The main work in this paper is summarized as follows:(1) Firstly, the research background about this paper is introduced in detail. Then the mechanism of the BBO is presented clearly and an experiment is held on several selected benchmark functions to test the BBO and some other evaluation algorithms. The weaknesses about the BBO are pointed out by the comparing about the results and performance of the algorithms. The further work about the research on the improvement of the BBO algorithm is decided by the results of the test and analysis.(2) As key operators, the immigration operator and emigration operator play important roles in the algorithm. The improvement of the performance of the algorithm can be realized by making the two operators better. In this paper, double migration operator and double mutation operator are proposed as the improved ones. Compared to the two original operators, the information can be shared more effectively by the improved migration operator and more opportunity to mutate and high-probability of a habitat to change into a better one can be ensure without the change of the mutation probability and the number of the evolutionary generation. So, the diversity of the population and explore ability of the algorithm are enhanced during the whole process of the optimization. The improvement of the algorithm is proved by the result of the experimental simulation based on benchmark functions.(3) The evaluation of the form and position error plays an important part in the machining and manufacturing of mechanical parts. With the requirement about the quality of the mechanical parts becomes higher and higher, accurate measurement of the errors of the productions has become a widespread concern in the industry. In this paper, the improved algorithm is applied to the field of form and position error evaluation. A new method is proved to be available based on the successful optimization about the cylindricity error evaluation and spatial straightness error evaluation and added to the toolbox to evaluate the errors.