Dual three-phase permanent magnet synchronous generator (DTP-PMSG) offers high efficiency and power density, and is suitable for high-reliability applications. To improve the voltage response of traditional linear active disturbance rejection voltage regulators in noisy environments, a voltage control method based on cascaded linear active disturbance rejection controllers is proposed. A low-pass filter is often used to filter noise in bus voltage signals, resulting in phase and amplitude losses. Therefore, in the proposed method, the output and filtering losses of the low-pass filter are defined as new system state variables. Subsequently, a cascaded linear active disturbance rejection controller is introduced to estimate the system state variables and compensate for filtering loss. Compared to the traditional linear active disturbance rejection voltage controller, the proposed method can effectively shorten the voltage recovery time and improve the anti-interference performance. Simulation and experimental results demonstrate the feasibility and effectiveness of the proposed control strategy.

Existing sensorless control methods are constrained by the convergence rate and estimation accuracy, and is difficult to be directly applied to high dynamic electro-mechanical servo systems. To improve the estimation accuracy of sensorless control under high dynamics, this paper proposes a novel sensorless control method for the permanent magnet synchronous motor based on amplitude-phase adaptive square wave injection. By injecting high-frequency square wave voltage signals and demodulating the high-frequency response current of the q-axis, high dynamic rotor position estimation is achieved. At the same time, the amplitude and phase of the injected voltage are adaptively adjusted according to the real-time operating status of the motor. Therefore, the estimation errors caused by inverter nonlinearity and motor cross-saturation effect are effectively reduced. Finally, the performance of the method proposed is experimentally verified on a 2 kW servo motor platform. The experimental results show that the proposed method can enhance the accuracy of rotor position estimation, while achieving high dynamic characteristics of sensorless control.