This thesis presents an optimal design of 12-slot/8-pole surface-mounted permanent magnet synchronous machines (SMPMSMs) by using subdomain model (SDM) and genetic algorithm (GA) during open-circuit and on-load conditions. The effect of magnet pole-arc, αp and slot opening angle, boa during open-circuit, and on-load conditions are taken into consideration. The model of 12-slot/8-pole SMPMSM with single segment per rotor pole are modelled and simulated by using OPERA 2D and AutoCAD software. Finite element analysis (FEA) in OPERA 2D is deployed to validate the predicted analytical results of the SMPMSMs by using SDM that include only for non-overlapping winding in radial magnetization (RM) and parallel magnetization (PaM) patterns. Besides that, a genetic algorithm is developed to optimize motor performances of the 12-slot/8- pole SMPMSMs with a single segment per rotor pole. The computing equations produced
the solutions for the objective function in order to obtain the optimum motor performances such as lower cogging torque, lower total harmonic distortion (THDv), lower electromagnetic torque ripple and higher average electromagnetic torque. The objective function is then employed with GA to determine the optimum settings of 12- slot/8-pole PM motor, such as, the magnet pole-arc and the slot opening of the PM machine. The optimal design of 12-slot/8-pole SMPMSMs with a single segment per rotor pole is further developed by using the combination of SDM and GA. After the optimization process, the optimal design of PM motor demonstrates lower cogging torque, THDv and reasonable electromagnetic torque as compared with that of the initial
design. The comparison of parameters and motor performances between the initial and the optimal designs of 12-slot/8-pole PM motors with the computing framework are validated in open-circuit and on-load conditions using SDM. Thus, the framework of SDM and GA is verified in reducing the usage of magnet materials and the motor performances of SMPMSM, which is highly viable for industrial applications. Therefore, the proposed framework of GA and SDM can determine the optimal settings of geometry design for SMPMSMs in order to produce optimum motor performances.
