Abstract：
This paper proposes a novel approach for tuning Proportional Integral Derivative (PID) controllers, utilizing experimental data obtained from an open-loop step input. We propose to use the measurements of the times taken to reach 5%, 35.3% and 85.3% of the final output, as well as the process static gain, to tune the controller. The tuning equations are applicable for a wide range of stable and over damped systems. Starting from an approximate model with three real poles and time delay (obtained from the measurements), the tuning equations approximate the controller that minimizes the Integral of Absolute Error (IAE) of the disturbance response. The designer can freely decide the Sensitivity Margin (Ms) to fix a desired robustness, as a difference with respect to other known methods where the user chooses among few predefined robustness values. The user can also select freely the derivative filter parameter, N, to find the required compromise between speed response and measurement noise amplification. For N=0 a PI controller is chosen. As N increases, a faster PID controller is selected, with higher noise amplification. We have also developed a web-based application and an Android application, downloadable for free, that implement the tuning equations as well as the whole required procedure.
摘要：
本文提出了一种新的方法来调整比例积分微分（PID）控制器，利用从开环阶跃输入获得的实验数据。我们建议使用达到5%的时间测量值，最终产量的35.3%和85.3%，以及过程静态增益，调整控制器。调谐方程适用于广泛的稳定和过阻尼系统。从具有三个实极和时间延迟的近似模型开始（从测量中获得），调谐方程近似于最小化扰动响应的绝对误差积分(IAE)的控制器。设计者可以自由地决定灵敏度裕度（Ms）来修复所需的鲁棒性，作为相对于其他已知方法的差异,其中用户在几个预定义的鲁棒性值中进行选择。用户还可以自由选择导数滤波器参数，N,在速度响应和测量噪声放大之间找到所需的折衷。对于N=0,选择PI控制器。随着N的增加,选择更快的PID控制器，具有更高的噪声放大。我们还开发了一个基于Web的应用程序和一个Android应用程序，可免费下载，实现调谐方程以及整个所需程序。
