Analysis and Design of Position and Velocity Estimation Scheme for PM Servo Motor Drive with Binary Hall Sensors

Abstract

with the increasing demand of low-cost, highefficiency, high performance for AC motor drive system, the permanent-magnet synchronous motor (PMSM) with binary Hall sensors begins to be adopted in many fields. Compared with sensorless control, the usage of binary Hall sensors is a guarantee for the drive to achieve moderate control performance, and it is in smaller volume and more costeffective compared with other types of position sensors. In this paper, a solution is provided to realize fully-closed loop control with low-resolution position sensors, by treating the position and speed estimators as separate systems. Results reveal that the model-based methods can take advantage of model information and model-free methods can smoothly process the quantized Hall position signal. Extensive experiment results are provided demonstrating the position control performance and basic servo performance for a PMSM drive using 3 bitper-pole-pair sensing system.

A New Position and Speed Estimation Scheme for Position Control of PMSM Drives Using Low-Resolution Position Sensors

“…Thus, the estimated speed of average speed method with MA filter acts as a low pass filter whose effects on the velocity are presented in [22]. Due to the delay of the estimated velocity in the low-speed region and the integration error in estimated position, the speed control loop may become slower and inefficient, and sometimes even unstable.…”

“…And Δt is much bigger than Ts almost in the whole speed range when using lowcount encoder. Therefore, Δt must be considered, and this is the reason for the existence of bumps [22].…”

“…8, the next Hall signal will be produced at the position (θh(k)+ Δθ). In addition, in order to gain a better fitting accuracy, a constraint condition is used for the coming Hall sector with θf bounded by: To realize a smooth and steady performance at the start-up process, the real-time reconstruction strategy is utilized [22]. As shown in Fig.…”

“…However, the two opposite goals in position and speed estimation cannot be achieved in one estimation at the same time. Since nonmodel-based methods can smoothly process the quantized Hall position signal and the model-based methods feature good dynamics in speed estimation by taking advantage of model information, it is reasonable to design the position and speed estimators separately by taking the advantages of those nonmodel-based and model-based methods [13,21,22].…”

Development of a Resolver-to-Digital Converter Based on Second-Order Difference Generalized Predictive Control

“…Encoder and resolver are the most used angular position sensors in industry [7][8][9][10]. Resolver is most used for harsh applications as it can resist higher temperatures, shocks, and vibrations than encoders [9,10].…”