针对模型不确定性的连续时间时滞系统,提出了一种新的神经网络自适应控制。系统的辨识模型是由神经网络和系统的已知信息组合构成,在此基础上,建立时滞系统的预测模型。基于神经网络预测模型的自适应控制器能够实现期望轨线的跟踪,理论上证明了闭环系统的稳定性。连续搅拌釜式反应器仿真结果表明了该控制方案的有效性。 Aiming at the continuous time delay system with model uncertainty, a new neural network adaptive control is proposed. The system’s identification model is composed of the neural network and the known information of the system. Based on this, a prediction model of the system with time-delay is established. The adaptive controller based on the neural network prediction model can track the desired trajectory, and theoretically proves the stability of the closed-loop system. The simulation result of continuous stirred tank reactor shows the effectiveness of this control scheme.