Your conditions: Engineering Cybernetics
  • Method for simultaneous object identification and optimal control based on Model System

    Subjects: Information Science and Systems Science >> Control science and technology Subjects: Computer Science >> Computer Application Technology Subjects: Engineering and technical science >> Engineering Cybernetics submitted time 2022-12-07

    Abstract: Abstract: Objective In practice, more than 95 % of industrial process control problems can be solved by PID control algorithm. In this paper, a new method of general controller optimization setting is established on the basis of inheriting and innovatively using 2DOF PID internal model control technology. Methods Model System was introduced into the method, and optimization template was obtained after offline simulation of the Model System. In the process of the signal excitation of the actual control system, the PID internal model control parameters are set by a specific algorithm guided by the template. Results Without accurate object parameters, the identification of object parameters and optimization of control performance can be completed simultaneously after several cycles of iteration.
    Conclusions This method has high setting efficiency, is convenient for later online maintenance of the system, and reduces the requirements for the technology and experience of the implementation personnel. The controller inherits the characteristics of good control with large time delay and strong robustness of internal model control, and both "target tracking" and "interference suppression" performance optimization. The control algorithm is simple, intuitive, and easy to be upgraded in the original PID control or realized by embedded system software and hardware, which is easy to be popularized and applied in production.
    Limits and future work Restricted by the design requirements of traditional internal model control, for the controlled object with negative response characteristics, this method should adopt the pre compensation processing mechanism or find a new model system and study the improved algorithm.