The Application Study Of Improved Gp In Circuit Modeling

When the circuit system is complicated and its function is hard to understand, the establishment of circuit equivalent model is the key to simulation analysis.Compared with standard algorithms, the GP is more flexible and it can get a variety of model structure without prior knowledge. In the paper, the method based on GP is used in circuit modeling.Main works in this paper are illustrated as follows.1 The improved GP is proposed and it substitute standards mutate for parallel mutate . The improved GP combine kinds of mutation operation to generate offspring .This algorithms comprehensive bring all the advantages of operators and change the structure obviously. It also increases different models and maintain individual's diversity and make the algorithms achieve the purpose of improving the convergence rate. By analyzing the convergence of the improved GP , it provides a theoretical basis for circuit modeling.2 The improved GP is used in circuit modeling .Two instances are given to test the improved GP. Compared with the experimental statistical methods , the results show that the improved GP obtain many models without prior knowledge and the precision is better. Compared with the standard GP, the improved GP is faster in convergence rate.