论文部分内容阅读
研究了前向多层神经网络及其BP算法，并利用此网络设计了辨识器和控制器，进行离线辨识、在线辨识和控制等。研究结果表明，这种控制结构能对复杂的非线性系统进行有效的控制，并且具有收敛速度快、抗干扰性好、鲁棒性强等特点，效果明显优于一般的PID控制方法。
The forward multi-layer neural network and its BP algorithm are studied, and the recognizer and controller are designed by using this network for off-line identification, on-line identification and control. The results show that this control structure can effectively control complex nonlinear systems and has the characteristics of fast convergence, good anti-jamming and robustness, which is obviously better than the general PID control method.