Sensorless control method with relatively low frequency signal injection for low acoustic noise

In this paper, we proposed the new sensorless control approach with relatively low frequency signal injection to reduce acoustic noise and vibration. Filters that are classically used in signal processing of currents are the main obstacle to high dynamic performance and even it make unstable the control system when the signal frequency become low enough that is close to the bandwidth of current controller. To overcome these problems, we proposed to use model-based estimation method instead of filters. It is verified by experiments that dynamic performance of the sensorless control is improved by proposed method compare to the conventional filter-based method.