Performance Analysis of Induction-Based Reaction Spheres

Abstract

Induction-based reaction spheres have been presented in many references and their performances are normally investigated through experiments or numerical simulations which are time-consuming. Here, an analytical way is presented and it enables researchers to evaluate a new design quickly. In this article, the presented performance analysis is conducted through the classical equivalent circuit approach. Involved circuit parameters are determined through the magnetic flux density distribution which is a function of design variables. Based on this, the steady-state torque-speed curve and the achievable maximum driving torque T\ast are identified. T\ast deviates from the maximum torque obtained from numerical simulations by only 3%. For validation, the presented performance analysis method is applied to an experimental case. Mean absolute percentage errors of predicted torque-speed curves are within 23% and mainly caused by end effects. The presented performance analysis method is generally applicable to induction-based spherical actuators, not only limited to reaction spheres. Additionally, since influences of design variables of the actuator have been formulated analytically through the determined circuit parameters, performance optimizations could be greatly facilitated.