A Novel Method for Accuracy Improvement of Variable Reluctance Linear Resolvers

Abstract

A novel technique for accuracy improvement of the variable reluctance linear resolvers (VR-L resolvers) is investigated in this article. A well-known structure and a calculation correcting factor are applied to propose a VR-L resolvers model with higher accuracy. A periodic finite-element analysis (FEA) is employed with a shorter processing time. The modeling is also performed for conventional and proposed improved structures by the mentioned FEA, where the extra teeth suppress the end effect in the model. The main reasons for the position error in linear actuators/sensors are the windings’ end effect and nonequal magnetic circuit. Two additional compensator teeth (CT)/slots with different heights are considered at both entrance and front parts of the mover, which introduces an improved structure. Furthermore, a new correction factor is employed to amend the demodulation process resulting in better performance. Compared to the conventional type, the proposed resolver is evaluated in healthy and faulty cases with asymmetric air-gap conditions. It is shown that the position error is significantly reduced in the improved structure. Finally, the effectiveness of the proposed design is proved by experimental results.