The permanent-magnet-motor operating mechanism has the advantages of simple and flexible control and

at the same time

has the shortcomings of long opening and closing time

easy vibration and noise

therefore the motor needs design optimization. Usually

the optimization of motor parameters requires much finite element simulation calculation

which consumes much time and computer resources. In view of above problems

the simulation model of permanent magnet motor is firstly set up in this paper in order to reduce the opening and closing time of the circuit breaker as well reduce the vibration of the motor and

at the same time

perform fast speed optimization of the motor. The samples are accumulated based on the method of center compound experiment. Then

the polar arc coefficient

permanent magnet thickness

stator notch width and spherical gap length are taken as input parameters

and the opening and closing time as well as groove torque are taken as output parameters to train the BP neural network model so that the neural network can predict the opening and closing time and torque better. Finally

the NSGA-II algorithm is used to optimize the structural parameters of the drive motor by way of combining BP neural network with genetic algorithm. The opening and closing time of the the motor after optimization is reduced by 12.94 ms

and the groove torque is reduced by 59.2%

which significantly improves performance of the motorized operating mechanism.