Design of electric machines with an improved heuristic algorithm

Permanent Magnet Synchronous Motors (PMSM) are one of the most preferred electrical machines due to their superior features such as high efficiency, high power-torque density and compact structure. However, the motor structure needs to be optimized for situations such as increasing the energy efficiency of the PMSM, be maximum the output power, reducing the cogging torque and motor weight. In this study, it is aimed to find the optimal values of the parameters required for the optimal design of 15 kW PMSM. For this purpose, a new optimization algorithm is proposed. This proposed algorithm is an application of the fuzzy logic-based fitness distance balance design to the gradient-based optimization algorithm (GBO). The developed algorithm and various optimization algorithms with effective results in the literature were tested for the fitness function determined to find the optimal values of Embrace, Offset, Skew, Magnet Thickness and Stator slot bottom width Bs1 of PMSM. It is seen that the proposed fFDBGBO algorithm increases the efficiency by 0.2979% compared to the GBO algorithm. Compared to the initial design, the fFDBGBO motor design provides a significant increase in efficiency from 91.7788% to 93.0366%, the magnet weight is reduced from 2.14837 kg to 2.09475 kg, but the total motor weight increases from 53.2286 kg to 54.0436 kg.These results show that the fFDBGBO algorithm optimizes critical parameters such as weight and fill factor by keeping them within the limit values while increasing efficiency in motor design. In optimization algorithms, it is desired that the fitness value converges to zero. In this study, the fitness value has the best value in the proposed algorithm with 0.0699248. The results obtained show that the proposed algorithm is more effective compared to competing algorithms.