TL;DR: Advanced PID Control builds on the basics learned in PID Controllers but augments it through use of advanced control techniques, including auto-tuning, gain scheduling and adaptation.
Abstract: The authors of the best-selling book PID Controllers: Theory, Design, and Tuning once again combine their extensive knowledge in the PID arena to bring you an in-depth look at the world of PID control. A new book, Advanced PID Control builds on the basics learned in PID Controllers but augments it through use of advanced control techniques. Design of PID controllers are brought into the mainstream of control system design by focusing on requirements that capture effects of load disturbances, measurement noise, robustness to process variations and maintaining set points. In this way it is possible to make a smooth transition from PID control to more advanced model based controllers. It is also possible to get insight into fundamental limitations and to determine the information needed to design good controllers. The book provides a solid foundation for understanding, operating and implementing the more advanced features of PID controllers, including auto-tuning, gain scheduling and adaptation. Particular attention is given to specific challenges such as reset windup, long process dead times, and oscillatory systems. As in their other book, modeling methods, implementation details, and problem-solving techniques are also presented.
TL;DR: A general way to take into account, by an appropriate design of the controller, any discrepancy which can occur between the actual inputs of a process and the desired outputs of its controller, which yields to the so-called conditioned control algorithms.
TL;DR: A general procedure to avoid windup which admits a unification of the ideas is given and the results are illustrated on a number of examples.
Abstract: This paper describes the phenomenon of integrator windup and various ways of avoiding it. It first covers a number of ad hoc schemes. A general procedure to avoid windup which admits a unification of the ideas is given and the results are illustrated on a number of examples.
TL;DR: A software package developed in SIMULINK is discussed that enables the behavior of feedback control systems with actuator saturation and PID controllers to be evaluated and the performance of four different anti-windup implementations for PI or PID controllers is compared.
Abstract: This article describes software developed in SIMULINK that enables the behavior of feedback control systems with actuator saturation and PID controllers to be evaluated. The software, which is part command- and part menu-driven, allows a choice of four PID controllers using different integral wind-up strategies and transfer function entry of the actuator and plant dynamics. Most realistic control systems contain nonlinearities of some form. One nonlinearity commonly found in control systems is a saturating element. If integral control is applied to such a system to eliminate steady state error, an undesired side effect known as integrator windup may occur when large setpoint changes are made. This effect leads to a characteristic step response with a large overshoot and a very high settling time. To avoid this situation, many different anti-windup strategies have been suggested. This article discusses a software package that has been developed in the SIMULINK/MATLAB environment to investigate and compare the performance of four different anti-windup implementations for PI or PID controllers. The software is partially menu-driven and enables the user to easily enter his own actuator and plant transfer functions to study the performance with the different controllers. >
TL;DR: A new variable-structure (switching) method for the prevention of proportional-integral-derivative controller integrator windup in both continuous and discrete-time implementations is proposed.
Abstract: In this paper, the authors propose a new variable-structure (switching) method for the prevention of proportional-integral-derivative controller integrator windup in both continuous and discrete-time implementations. The method is easily implemented and tuned by practicing engineers. This new method is compared with several existing methods for the prevention of integrator windup.