Application Of Artificial Immune System In Immunized Identification Of Complex System

With the development of science and technology, the modern industry systems are becoming more and more complex. The traditional controllers can't satisfy the high performance of the system. In this background, the intelligent control theory is proposed. And it develops very quickly in theory these years. The intelligent control has many successful applications in complex control systems, too. Now it has become the new stage of the control theory [1].Artificial immune system (AIS) is an active branch of the intelligent control in recently years. AIS consists of all kinds of models, algorithms and their applications in engineering and science that are based on the main principles and mechanisms of natural immune system.This dissertation explores the theoretical foundation and main approaches of complex system immune identification based on intelligent system theory. The paper does the further research in this domain on the basis of the domestic and foreign research status, and the main contents are as follows.(1) Research on artificial immune system and its algorithm. This charpter expounds the main principles of AIS. The biological mechanisms, the research content and range are presented in detail. Then expound the principle, characteristic and structure of immune algorithm. Analyse the present problems of clone selection algorthim, advance a new improved clone selection algorithm in which the calculate methods of affination and antibody concentration have been improved.(2) Research on traditional identification methods. Expound the principles of the method of classical identification and modern identification. Using two different identification methods, a typical linear system is simulated, and the localization of traditional identification method is pointed out. Then describe some other identification methods at present.(3) Research on a kind of intelligent identification method based on neural networks. Compare the influence of different neural network learning algorithms for systemidentification through the simulation result of a nonlinear system model. Point out the problems in the present methods.(3) Research on a novel neural network identification method based on an improved hybrid immune algorithm. Aiming at the problems of the traditional system identification methods and neural networks method, this thesis proposes a novel neural network identification method based on a mixed optimizing algorthim. And then apply it to the identification of the electronic component placement process in pick-and-place machines. The result of simulation validates the efficiency of the modeling method. Not only can it avoid the disadvantage of local optimum value in process of neural networks learning, but also can improve the precision and convergence speed of tranditional GA and IA.