Predictive Functional Control in Fractional-Order Systems

Predictive Functional Control (PFC) has emerged as a compelling alternative to traditional proportional–integral–derivative schemes, distinguished by its capacity to forecast future process outputs and adjust inputs accordingly. When fused with fractional calculus—an extension of classical differentiation and integration to non-integer orders—this approach captures long-term memory effects and system inertia in a more nuanced manner. Fractional-order models imbue the controller with adjustable weighting of past behaviour, leading to improved robustness against disturbances and unmodelled dynamics. In practice, fractional-order PFC formulations define a cost function incorporating real-order derivatives of both output and control signals, optimised over a prediction horizon. This affords additional degrees of freedom in tuning and can alleviate issues of conservatism often encountered with purely integer-order formulations. Applications have spanned automotive cruise control, industrial furnaces and process loops, where time delays, nonlinearities and actuator faults pose significant control challenges. By integrating fractional dynamics directly into the prediction and control law, researchers have achieved enhanced tracking performance, smoother responses and systematic constraint handling. The global significance of this hybrid methodology lies in its versatile applicability across domains requiring fine-grained management of complex, memory-rich systems, from chemical reactors to renewable energy installations.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent improvements in fractional-order PFC have addressed the challenge of constraint handling through an implied closed-loop prediction mechanism. Rather than assuming constant future inputs, the controller infers future actuator dynamics within the predictive model, allowing more aggressive satisfaction of constraints without undue conservatism. Simulation studies demonstrate significantly tighter set-point tracking and more reliable disturbance rejection, along with reduced tuning complexity compared to conventional PFC.

Another line of work targets resilience to partial actuator failures in industrial processes. By embedding fractional-order derivatives into the PFC cost function and employing a population-based optimisation to select prediction horizon, fractional order and smoothing parameters, this strategy maintains control performance under both constant and time-varying actuator degradations. Comparative simulations on batch reactors and furnace flow processes confirm superior fault tolerance relative to integer-order PFC and other metaheuristic-tuned variants.

Predictive Functional Control in Fractional-Order Systems publication trend

The graph below shows the total number of articles in predictive functional control in fractional-order systems across all publications each year (not limited to Nature Index journals).

Technical terms

Fractional calculus: A generalisation of classical calculus in which differentiation and integration are defined for non-integer orders, enabling models to account for memory and hereditary properties of physical processes.

Predictive Functional Control (PFC): A control scheme that uses an explicit prediction model to forecast future outputs and computes control moves by minimising a performance index over a finite prediction horizon.

Prediction horizon: The future time window over which the control algorithm forecasts system outputs and evaluates performance, crucial for balancing responsiveness against computational complexity.

Cost function: An optimisation criterion defined over the prediction horizon, typically quantifying tracking error and control effort, which PFC seeks to minimise when determining control actions.

References

1. IMPROVED CONSTRAINT HANDLING APPROACH FOR PREDICTIVE FUNCTIONAL CONTROL USING AN IMPLIED CLOSED-LOOP PREDICTION. IIUM Engineering Journal (2021).
2. Fractional‐Order Predictive Functional Control of Industrial Processes with Partial Actuator Failures. Complexity (2020).