Pole-zero cancellation speed and acceleration filtering technique with disturbance observer for servo drive applications without model parameter information

This paper is concerned with the problem of robustly estimating the speed and acceleration of servo drives without the system model parameter information, depending on only the order of the open-loop system. The proposed filtering solution incorporates the disturbance observer (DOB) into the second-order pole-zero cancellation (PZC) observer. The main contributions fall into three parts. First, the model-free Luenberger observer as the first subsystem specifies its gain structure in the nonlinear form to invoke the second-order PZC. Second, the first-order nonlinear DOB as the second subsystem estimates the high-frequency disturbance to yield the compensation term for the second-order PZC observer. Third, the combination of these two subsystems eventually results in the diagonalized estimation error dynamics, which makes the performance adjusting process more convenient for field engineers. The experimental study confirms the effectiveness of the proposed filtering solution using a 500-W brushless DC (direct current) motor-based servo drive under various operation modes.