论文部分内容阅读
针对锅炉温度系统的非线性、滞后、时变等特性,提出了一种串联控制策略。提出基于神经网络的微分型预测控制算法,该方法的突出优点是能够加快调节时间。在此基础上结合常规PID控制器构成了预测-PID串联控制,这种串联控制的方法既有基于神经网络的预测控制在实时系统中抗干扰能力强的优点,又充分利用了PID控制方法响应速度快的特点。通过对锅炉温度系统的实时控制实验,证明了所提方法的有效性,极大地提高了系统的控制品质。
Aiming at the nonlinearity, hysteresis and time-varying characteristics of boiler temperature system, a series control strategy is proposed. The differential predictive control algorithm based on neural network is proposed. The outstanding advantage of this method is that it can speed up the adjustment time. On this basis, combined with the conventional PID controller constitutes a predictive-PID series control, this series control method based on neural network-based predictive control in real-time system has the advantages of anti-interference ability, but also take full advantage of the PID control method to respond Fast features. Through the real-time control experiment of the boiler temperature system, the effectiveness of the proposed method is proved and the control quality of the system is greatly improved.