论文部分内容阅读
为非线性控制系统设计了一种带智能补偿器的神经网络控制器.神经网络用来提供主要的控制输入信号,网络的训练可离线进行,网络的使用采用联想记忆方法.经神经网络控制器控制后,系统的实际输出值与期望值之间偏差不大,因此可将误差模型方程线性化.针对这一线性化模型,可以设计一个智能补偿器来补偿由于训练精度、干扰及时延等因素引起的系统误差.利用以上设计方法,本文对战斗机的“眼镜蛇”机动飞行控制进行了仿真研究,仿真结果表明所设计的控制系统具有良好的性能.
A neural network controller with intelligent compensator is designed for the nonlinear control system. The neural network is used to provide the main control input signal, the training of the network can be carried out offline, the associative memory method is used in the network, and the neural network controller After the control, the deviation between the actual output value of the system and the expected value is small, so the error model equation can be linearized.For this linearized model, an intelligent compensator can be designed to compensate for the factors such as training accuracy, interference and delay The system error of the maneuver flight control of the “Cobra” of the fighter plane is simulated by the above design method, the simulation results show that the designed control system has good performance.