采用智能化的PID控制器是新一代DCS发展的趋势。本文针对用于某油田的DCS,提出了一种适用于时滞对象并具有参数自整定和自适应功能的PID控制器,这种PID将系统辨识和基于模式识别的专家自整定方法相结合,对PID控制波形进行特征分类,并从实用的角度提出了控制目标由波形模式和控制指标两个要素组成。计算机仿真结果表明,当对象参数变化时仍能取得满意的控制结果,具有实际应用价值。 The use of intelligent PID controller is a new generation of DCS development trend. In this paper, we propose a PID controller which is suitable for time-delayed objects and has the function of parameter self-tuning and self-adaptation. This kind of PID combines the system identification with the expert self-tuning method based on pattern recognition. The characteristics of PID control waveform classification, and from a practical point of view the control target by the waveform mode and control index of two elements. Computer simulation results show that when the object parameters change can still achieve satisfactory control results, with practical value.