Automatic Control Systems and Process Identification

Automatic control systems ensure the stability and performance of dynamic processes by continuously adjusting inputs based on feedback. Central to this field is the creation of accurate mathematical models through process identification, which infers system dynamics from measured signals. Techniques range from classic relay feedback tests to modern event-based sampling and infinite-dimensional modelling, allowing controllers to adapt in real time and cope with delays, nonlinearities and network constraints. Process identification underpins auto-tuning of proportional–integral–derivative (PID) controllers, enabling rapid deployment across industries such as chemical processing, energy management and robotics. Recent advances have emphasised robustness against communication uncertainties, minimal data exchange via event triggers and high-fidelity approximation of complex thermal or mechanical systems. The interplay between identification algorithms and controller design has driven global adoption of self-optimising loops, energy-efficient network protocols and autonomous calibration routines in smart factories.

Research from Nature Portfolio

A recent study has explored relay-based identification of an infinite-dimensional model representing a heat-exchanger process. By analysing the pole loci of a delayed quasi-polynomial characteristic function, researchers derived a low-order surrogate model whose dominant poles can be assigned via a single saturated relay experiment. This approach yields multiple parameter estimates in one test and refines them through numerical optimisation and autotune relay variations. The method demonstrates accurate control performance for complex thermal plants while maintaining computational simplicity, thereby facilitating real-world implementation in industrial heat-exchange loops.

Research from all publishers

An auto-tuning procedure for PI controllers operating under a symmetric-send-on-delta sampling strategy has been developed, employing two iterative frequency-response estimates to apply a simple robust tuning rule. Simulation studies confirm its effectiveness across typical industrial dynamics while guarding against limit-cycle oscillations induced by packet delays. Another investigation has characterised the robustness of networked event-based control systems utilising regular quantisation with hysteresis sampling. By defining a margin for limit-cycle emergence in the frequency domain, this work evaluates standard PID designs under quantised communication and validates findings experimentally. A further contribution introduces an online method to identify second-order plus dead time (SOPDT) models via a single relay experiment. By measuring four key quantities during the relay oscillation, the algorithm computes model parameters directly for immediate PI/PID tuning, as demonstrated in both simulations and laboratory apparatus.

Automatic Control Systems and Process Identification publication trend

The graph below shows the total number of articles in automatic control systems and process identification across all publications each year (not limited to Nature Index journals).

Technical terms

Automatic control system: A feedback loop that adjusts inputs to maintain a desired output in a dynamic process.

Process identification: The procedure of deriving a mathematical model of a system from measured input–output data.

Relay feedback experiment: A test where an on/off relay replaces the controller to induce self-sustained oscillations, revealing frequency-response points.

Event-based sampling: A data-acquisition strategy that transmits signals only when certain conditions or thresholds are met, reducing communication load.

Transfer function: A mathematical representation relating a system’s output to its input in the frequency domain.

Limit cycle oscillation: A self-sustained periodic output caused by nonlinear elements or sampling strategies within a control loop.

References

1. Auto-tuning method for PI controllers under Symmetric-Send-on-Delta sampling strategy. Journal of the Franklin Institute (2024).
2. Robustness Study for Networked Event-Based Control System Under Regular Quantization With Hysteresis Sampling. IEEE Transactions on Control of Network Systems (2024).
3. Parameter identification of a delayed infinite-dimensional heat-exchanger process based on relay feedback and root loci analysis. Scientific Reports (2022).
4. Online SOPDT Model Identification Method Using a Relay. Applied Sciences (2023).