Improvement of Servo-Hydraulic Pressure Control Performance Using PSO-Optimized Fuzzy-PID Control

Accurate pressure tracking in servo-hydraulic test benches is challenging under variations in actuator dynamics and feedback characteristics. This study compares conventional proportional–integral–derivative (PID), fuzzy self-tuning PID (Fuzzy-PID), and particle swarm optimization-based Fuzzy-PID (PSO-Fuzzy-PID) controllers using a common MATLAB/Simulink R2024b model. The proposed approach combines online Mamdani fuzzy gain adjustment with the offline PSO optimization of controller scaling and derivative-filter parameters. Performance is evaluated under nominal conditions and three single-parameter perturbations involving the servo-valve time constant, servo-valve gain, and pressure-sensor gain. The maximum percentage overshoot, 2% settling time, and integral of time-weighted absolute error (ITAE) are assessed. Across the four conditions, PSO-Fuzzy-PID reduces the average overshoot by 20.8%, settling time by 10.1%, and ITAE by 31.8% relative to conventional PID. Although the non-optimized Fuzzy-PID reduces the overshoot and ITAE, it increases the settling time, highlighting the importance of parameter optimization. Thirty independent PSO runs were also conducted to assess optimization repeatability. These findings demonstrate improved pressure-tracking performance and robustness within the investigated simulation conditions, rather than experimentally validated robustness. Future work will investigate broader simultaneous parameter uncertainties, measurement noise, and physical test-bench validation to assess practical applicability.