Scholarlink

<p>In the scientific community, there are many algorithms for computing the parameters of First-Order (FOPTD), Second-Order (SOPTD) and Integrating (IPTD) plus time delay processes in open-loop. There are, however, few methods available for identifying this type of process model in closed-loop, controlled by a Proportional-Integral-Derivative (PID) or a Model Predictive Control (MPC) controller. In this paper, it is proposed a novel method to perform identification of these three model structures in closed-loop. The proposed method uses the collected data of the controlled and manipulated variable from the step response in closed-loop. The proposed algorithm is formulated in two steps: (1) the direct identification of open-loop model using an ARX model on Laguerre orthonormal bases, looking for minimizing the error of real and predictive values and (2) the identification of the process based on graphical response. Several simulation results and an industrial real case, in the presence of different noise levels, demonstrated the effectiveness of the proposed method.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Estimation of FOPTD, SOPTD and IPTD Processes Parameters in Closed-loop from Step Response

  • 期刊论文

  • 发表日期：

    2026年3月13日

查看全文
  • Christiam Segundo Morales Alvarado,
  • Claudio Garcia,
  • Flavio S. C. da Silva

摘要

In the scientific community, there are many algorithms for computing the parameters of First-Order (FOPTD), Second-Order (SOPTD) and Integrating (IPTD) plus time delay processes in open-loop. There are, however, few methods available for identifying this type of process model in closed-loop, controlled by a Proportional-Integral-Derivative (PID) or a Model Predictive Control (MPC) controller. In this paper, it is proposed a novel method to perform identification of these three model structures in closed-loop. The proposed method uses the collected data of the controlled and manipulated variable from the step response in closed-loop. The proposed algorithm is formulated in two steps: (1) the direct identification of open-loop model using an ARX model on Laguerre orthonormal bases, looking for minimizing the error of real and predictive values and (2) the identification of the process based on graphical response. Several simulation results and an industrial real case, in the presence of different noise levels, demonstrated the effectiveness of the proposed method.