Heating-up control with delay-free output prediction for industrial jacketed reactors based on step response identification

Twice-optimal control design with construction pruning for second order plus time delay systems

To effectively control a class of second-order plus time delay (SOPTD) systems, based on twice-optimal control (TOC) and construction pruning (CP) methods, an SOPTD-TOCCP controller is proposed, which can achieve strong robustness and excellent set-point tracking performance. The TOC controller of the SOPTD is designed based on a classical cascade controller and an extended state observer (ESO). A fast and accurate method is proposed to help engineers obtain the optimal time scale, which is the most critical parameter for regulating control performance. The influence of different parameter sensitivities on SOPTD is studied. In addition, a new robust enhancement method is proposed for SOPTD systems. A construction pruning method for SOPTD systems is proposed to further improve control performance, particularly robustness. Finally, a comparison with other control methods demonstrates that the SOPTD-TOCCP controller is simple, reliable, and versatile and can achieve better control performance.

Anti-windup ADRC design for temperature control systems with output delay against asymmetric input constraint

To tackle the asymmetric input constraint typically involved with industrial temperature control systems, e.g., there is a tuning range from 0 to 100% in the heating power, an anti-windup active disturbance rejection control (ADRC) is proposed for the practical application, in particular for dealing with output delay. Based on an artificial symmetric transformation for the asymmetric input constraint along with a delay-free output prediction, an anti-windup extended state observer (AESO) is firstly constructed to simultaneously estimate the delay-free system state and external disturbance for control design. When the input saturation constraint occurs, the proposed AESO could also offer anti-windup compensation. Then, a feedback controller is analytically derived by assigning repetitive-type poles for the characteristic equation of the closed-loop structure, followed by a set-point prefilter design to ensure no overshoot and output tracking deviation. Tractable linear matrix inequality based conditions are rigorously established to guarantee the asymptotic stability of the resulting closed-loop system under a specified initial condition. A benchmark example from a recent reference is used to verify the superiority of the proposed control method over some existing anti-windup control designs. Moreover, a real application to the temperature control of a jacketed crystallizer is shown in comparison with a recently developed control method with no anti-windup compensation.

Simplified filtered Smith predictor for high-order dead-time processes

This paper proposes a control structure suitable for high-order non-minimum phase (NMP) processes. In general, dead-time compensators (DTCs) firstly predict the process output with zero error at a steady-state and then a primary controller with an integrator is designed based on the delay-free model. The main advantage of the proposed structure is that the primary controller is only a state feedback gain with no integrators. This leads to fewer parameters to tune and lower order filters, while a robustness filter is used to reject disturbances and guarantee zero error at a steady state. Simulation results show better or equivalent performance compared to other recently published works, even kept the controller design simplicity.

Controlling industrial dead-time systems: When to use a PID or an advanced controller

This work presents a comparative analysis of PID, DTC and MPC strategies when used to control SISO processes with dead time considering characteristics commonly found in industry, such as noisy measurements in the process output and modeling error. For unconstrained processes, it is shown that the performance improvement obtained by using a more advanced control strategy instead of a PID is small or nonexistent for cases which require high robustness. However, for cases with well-known process models it is shown that the improvement obtained by using a more complex control structure is justified even for small delays. For constrained processes it was demonstrated that a PID with anti-windup is able to provide similar or even better results than MPC when robust solutions are considered.

New simple approach for enhanced rejection of unknown disturbances in LTI systems with input delay

This work proposes the extension of the simplified filtered Smith predictor (SFSP) for state-space systems to improve rejection of matched and unmatched unknown disturbances in LTI systems with input delay. The proposed structure is simpler than others recently proposed in the literature and can be applied to continuous-time or discrete-time systems. Furthermore, it allows improving rejection of both matched and unmatched disturbances, while also enhancing noise attenuation and robustness characteristics. Finite spectrum assignment (FSA) based implementation is used in order to guarantee the internal stability of the proposed controller. Simulation and experimental results are used to show the usefulness of the proposal.