PWM-Based Energy-Efficient Adaptive Control for Multichamber Hydraulic Servo Actuators

Digital multichamber hydraulic servo actuators (HSAs) guarantee good energy efficiency while delivering high force with a compact size-to-power ratio, even under multiple loads. However, the exponential growth of achievable force combinations as the number of chambers increases, together with the inherently hybrid behavior of digital HSAs, poses significant challenges in modeling and control. This article addresses these issues through a twofold contribution. First, we propose a novel multiple-input–single-output (MISO) average equivalent discrete-time model (AEDM) based on a pulsewidth modulation (PWM) strategy. The formulation establishes a direct relationship between the duty-cycle of the PWM signals governing each digital valve pair and the force generated by the piston rod, yielding a control-oriented representation suitable for force regulation. Second, a two-level adaptive control architecture, combining a decoupled proportional-integral (DPI) controller with a decentralized model reference adaptive control (MRAC), is developed for force tracking and pressure control, respectively. The proposed methodology is validated on a four-chamber digital HSA and compared against a model predictive control (MPC) architecture. Realistic simulations show that, while achieving a comparable peak force tracking error, the proposed approach significantly improves energy efficiency by streamlining the control structure and reducing computational burden. Moreover, the adaptive strategy effectively compensates for mechanical coupling effects and treats hydraulic parameters as uncertainties, highlighting its robustness and practical applicability.