论文部分内容阅读
提出了一种利用基因表达式编程进行非线性系统辨识的新方法,引入了可变染色体和可变终端符集,提出了新的个体的生成机制及相应进化操作符,克服了利用遗传编程进行非线性系统辨识的不足,降低了算法对参数的依赖性,能够在相同的参数设置下进行各种非线性NARMAX模型辨识。对于适应度的定义,综合考虑模型的精确性和复杂性因素,使辨识模型能够在精确性和复杂性之间取得平衡。仿真结果表明,这种方式可以快速准确的获取非线性模型。
A new method of gene expression programming for nonlinear system identification is proposed. Variable chromosomes and variable terminal character sets are introduced. A new individual generation mechanism and corresponding evolution operators are proposed, which overcomes the problem of using genetic programming The lack of nonlinear system identification reduces the dependence of the algorithm on the parameters and enables the identification of various nonlinear NARMAX models under the same parameter settings. For the definition of fitness, taking into account the accuracy and complexity of the model, the identification model can strike a balance between accuracy and complexity. Simulation results show that this method can quickly and accurately obtain the nonlinear model.