Research On Control Application Of Immune Algorithm With Inverter

With the development of artificial intelligence, all sorts of smart optimization algorithms gradually arise. These algorithms have made a series of achievements in theories and applications, and are now showing tremendous potential of engineering applications. Immune algorithms are extracted from immune principles in biology and they make up for some inherent deficiencies of genetic algorithms, become another research hotspot of intelligent optimization algorithms, and now it is in the stage of rapid development. How to combine immune algorithms with engineering projects have become an important issue for many researchers.This thesis starts from the concepts of immune-related properties and reviews the processes of refining, formation and derivation of immune algorithms, discusses their applications about control of inverter. On the basis of genetic and immunological principles, an improved immune optimization algorithm is designed for optimize the PWM control sequences of inverters. We regard single-phase full-bridge inverter as research object, and take into account both static and dynamic loading to verify its effectiveness. In the end, the results of simulation of the algorithm designed is analyzed and compared with ones of genetic algorithm, the superiority of immune algorithm is shown.Firstly, origination and development processes of immune algorithms are overviewed, then we introduce concepts and work principles of biological immune system and artificial immune system, and how to design immune algorithms from immune mechanisms. A brief report of relevant terminology of immune algorithms and applications about engineering is given too. Meanwhile, the current control strategies of inverter are listed simply.Secondly, the thesis introduces three main algorithms(clone immune algorithm, immune network algorithm, negative selection algorithm), all of them base on immunology principle, some vivid descriptions are given about their basic principles and execution steps, operators designing and characters respectively. Moreover, with the inverter control sequences for study, we establish optimization model base on PWM control sequences from the perspective of mathematical, design related operators of immune algorithm, put forward an improved immune optimization algorithm, and apply it about how to obtain PWM control optimization sequence of inverter.Finally, the proposed algorithm in the thesis is programmed and code implemented with MATLAB,a simulation platform is built based on Simulink too. we design numerical experiments of inverter load about both static and dynamic circumstances to verify the algorithms for application of optimizing PWM control sequences. Lateral correlation of Simulation results about the algorithm designed and traditional genetic algorithm, the results show that the algorithm designed have certain advantages in PWM inverter control sequence optimization under the same conditions,...