Analytical Modeling and Optimization of Partitioned Permanent Magnet Consequent Pole Switched Flux Machine With Flux Barrier

Abstract

Switched Flux Permanent Magnet Machine (SFPMM) encompass unique features of conventional direct current machine, permanent magnet (PM) synchronous machine and switch reluctance machine therefore, applicable for high-speed applications. However, conventional SFPMM exhibits demerits of high PM volume ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">V_PM</i> ), high torque ripples ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_rip</i> ), higher cogging torque ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_cog</i> ), lower torque density ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_den</i> ) and significant stator flux leakage. In this paper, a new topology of consequent pole (CP) SFPMM (CPSFPMM) is proposed having partitioned PM that improved flux modulation phenomena utilizing flux barriers. Moreover, due to non-linear behaviour of PM and complex stator structure alternate analytical sub-domain model is utilizes for initial design. However, initial design offers lower open-circuit phase flux linkage (Φ), average mechanical torque ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_avg</i> ) and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_den</i> . Aforementioned electromagnetic key performance indicator with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_cog</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_rip</i> , total harmonics distortion in Φ (Φ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>THD</i></sub> ), average power ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P_avg</i> ) and power density ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P_den</i> ) are refined utilizing Geometric-Based Deterministic Optimization (GBDO). Analysis reveals that proposed new topology of CPSFPMM with flux barriers reduces <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_cog</i> by 34.90%, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_rip</i> by 20.27%, Φ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>THD</i></sub> by 28.08% whereas it enhanced <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P_avg</i> by 17.79%, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_den</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P_den</i> by 34.38%.