论文部分内容阅读
针对线性、时不变和具有不确定参数的对象进行辨别和控制研究,其辨别器和控制器的确保辨识系统全局稳定的自适应函数调参规律和结构选优组成等,都是构建在线性系统理论之上。由于非线性系统的辨别和自适应控制一直难于找到相应的数学方法,因此提出一种非线性系统数学模型构成系统模型辨识的新方法,将组合优化问题由非线性系统结构辨识问题转化而成。仿真实验表明,遗传算法求解非线性系统辨识比其他方法具有更好的近似解,证明了该算法的有效性和实用性。
For linear, time-invariant and uncertain parameters of objects to identify and control research, its discriminator and controller to ensure that the global stability of the identification system adaptive tuning rules and structure of the optimal composition of the composition, etc., are built on the linear System theory above. Since it is hard to find the corresponding mathematical method for identification and adaptive control of nonlinear systems, a new method of system model identification based on mathematical model of nonlinear system is proposed. The combinatorial optimization problem is transformed from the problem of nonlinear system structure identification. Simulation results show that genetic algorithm has better approximate solution to nonlinear system identification than other methods, and the validity and practicability of the algorithm are proved.