Abstract—

The article considers three implementation versions of controllers operating with constant speed actuators that are used in automation systems of a technological process: with a relay-pulse control unit (RPCU) with a lagging feedback and topping differentiator (a real differentiating element, RD), with calculating the output according to a PID algorithm and using the actuator model for its implementation, and with calculating the control output speed and using a pulse-width modulator. Formulas for calculating the parameters of the RPCU and RD primary components from the condition of approaching the equivalent PID algorithm are given. The system version with a PID controller with the actuator control proceeding from its model is considered in two options: with individual integrators for the controller’s integral component and the actuator model, and with a common integrator for the controller’s integral component and the actuator model. In the latter option, the signal from the actuator model is subtracted from the controller’s integral component, due to which the possibility of the integrator’s output to reach the integral component’s limit values is ruled out. The versions are compared by analyzing the automatic control system performance in the pulsed actuator’s operation mode based on simulation for the superheated steam temperature control section of a steam boiler equipped with a surface steam desuperheater. The analysis was carried out taking into account both the usual dynamic accuracy indicators and the number of actuators switched into operation during the transient. It is shown that the advantage of the version with the PID algorithm and common integrator for the integral component and the actuator model is that there is no need in this case to calculate the controlled variable second derivative; in addition, this version is free from the problem connected with accumulation of the difference between the signals from the actuator and its model. This version can be considered to be more preferred in implementing a PID algorithm in the software of programmed controllers. Recommendations for self-tuning of pulse duration and the control unit relay element parameters during the control process depending on the algorithm’s individual components are given. By using the proposed recommendations, it becomes possible to achieve an essentially lower actuator switching frequency during the control process.

中文翻译：

---

具有恒速执行器的PID控制器：实施版本和脉冲​​持续时间的自整定

摘要-

本文考虑了在工艺过程的自动化系统中使用的采用恒速执行器运行的控制器的三种实现版本：具有滞后反馈和顶部微分器（实际微分器RD）的继电脉冲控制单元（RPCU），根据PID算法计算输出，并使用执行器模型实现，并计算控制输出速度，并使用脉宽调制器。给出了在接近等效PID算法的条件下计算RPCU和RD主分量参数的公式。带有PID控制器且执行器控制从其模型开始的系统版本有两种选择：针对控制器的整体组件和执行器模型使用单独的积分器，并带有用于控制器的集成部件和执行器型号的通用集成器。在后一种选择中，将执行器模型的信号从控制器的积分组件中减去，因此，排除了积分器输出达到积分组件极限值的可能性。通过对配备有表面蒸汽减温器的蒸汽锅炉的过热蒸汽温度控制部分进行仿真，并通过分析脉冲执行器运行模式下的自动控制系统性能来比较这些版本。进行分析时要考虑到通常的动态精度指标和瞬态过程中投入运行的执行器数量。结果表明，带有PID算法和通用积分器的积分组件和执行器模型版本的优点在于，在这种情况下无需计算受控变量二阶导数。另外，该版本没有与执行器信号及其模型之间的信号差的累积有关的问题。在编程的控制器的软件中实施PID算法时，可以认为该版本更为可取。给出了根据算法的各个组成部分，在控制过程中对脉冲持续时间和控制单元继电器元件参数进行自整定的建议。通过使用建议的建议，可以在控制过程中实现实质上较低的执行器开关频率。