Abstract

This article presents the analysis and demonstration of sensorless position estimation methodologies for separately excited synchronous electrostatic machines (SEMs). The torque density of the SEM comes at the cost of a high pole count (p >50) along with other enhancements. The high pole count makes the use of high-resolution position sensors necessary, driving up the cost of the system. The sensorless estimation of rotor position is seen as a solution for alleviating the need for high-resolution and high-cost position sensors. Key features of electrostatic machines that make them suitable for sensorless operation are identified. Back MMF (current) based rotor position sensing for medium to high speeds is investigated. High-frequency injection-based position estimation is investigated for zero and low-speed operating regions. The proposed methods are experimentally demonstrated using a current-source inverter-based drive platform.

Highlights

  • F LUID-FILLED synchronous electrostatic machines (SEMs) have the potential to power applications requiring low-speed high torque with very low losses

  • SEMs operate at a medium-voltage level, which creates a challenge for power electronics and control, as there is no off-the-shelf drive available for controlling the machine

  • The drive configuration for the study of sensorless control implementation is shown in Fig. 2, where the machine back MMF is shown as the current source

Read more

Summary

INTRODUCTION

F LUID-FILLED synchronous electrostatic machines (SEMs) have the potential to power applications requiring low-speed high torque with very low losses. While the high pole count provides significant improvements in machine torque production, it creates a rotor position measurement challenge. It is worth noting that the analysis in [5] points to the use of even higher pole counts for optimized torque production, making the need for high-resolution position sensors an imperative. Sensorless approaches offer system weight and volume reduction while increasing drive reliability since a high-resolution position sensor is no longer necessary for rotor position estimation [7]. These attributes are ideally suited for SEMs with their innately high pole counts that normally require expensive sensors

Background of Sensorless Position Estimation Methods
Platform for Demonstration of the Proposed Methods
SENSORLESS CONTROL OF ELECTROSTATIC MACHINES
Back MMF-Based Position Sensing
Signal Injection-Based Position Sensing
BACK MMF-BASED POSITION ESTIMATION
Machine Modeling and Back MMF Estimation
Experimental Demonstration of Back MMF-Based Position Estimation
HIGH-FREQUENCY INJECTION-BASED POSITION ESTIMATION
High-Frequency Machine Model
Experimental Demonstration of Injection-Based Rotor Position Estimation
Feedback Signal Resolution
Carrier Signal Separation
Findings
Inverter Nonlinearity and Machine Parameters
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call