A model-free adaptive control algorithm design suitable for minimum quantity lubrication systems

To address the instability in atomization and the limited precision in cooling parameter adjustments in minimum quantity lubrication (MQL) systems, a novel MQL system was developed, employing an improved model-free adaptive control algorithm. The model-free adaptive control algorithm (MFAC) faces challenges such as insufficient dynamic performance and difficulty in parameter tuning when applied in practical scenarios. This study proposes a novel model-free adaptive control algorithm (MFA-DTSMC) that synergizes data-driven methodology with sliding mode control, integrated with an enhanced differential whale optimization algorithm. This innovative approach effectively addresses two critical limitations of conventional model-free adaptive control (MFAC): inadequate dynamic performance and empirical dependence in parameter tuning. The technical framework involves: (1) Transforming nonlinear systems into pseudo gradient data models through dynamic linearization modeling, (2) designing discrete terminal sliding mode surfaces with exponential reaching laws to significantly enhance steady-state accuracy and dynamic response speed without requiring prior system models. Furthermore, to resolve controller parameter sensitivity issues, we develop a hybrid whale optimization algorithm incorporating differential evolution mutation strategies. This advancement overcomes the global search capability constraints inherent in traditional whale optimization algorithms, enabling autonomous adaptation of critical control parameters. Simulation studies were conducted using MATLAB to verify the effectiveness of the proposed approach. The results showed that under disturbance conditions, the maximum steady-state error rates for air pressure and flow rate decreased by 2.07% and 5.14%, respectively, compared to the traditional model-free adaptive control (MFAC) controller. Additionally, the maximum overshoot for air pressure and flow rate decreased by 8.12% and 21.3%, respectively, compared to controllers with manually tuned parameters. Experimental atomization tests demonstrated that the MFA-DTSMC-controlled MQL oil mist exhibited more stable fluctuations within the desired particle size range (5–10 μm), achieving a desired particle size ratio of 40.20% with a standard deviation of 1.68%. In contrast, the MQL system without the control algorithm showed inferior performance, with an average desired particle size ratio of 37.89% and a standard deviation of 5.26%. The proposed algorithm offers a new strategy for enhancing the stability of atomization in MQL systems.