To assess demagnetization, a line-to-line short-circuit fault study is carried out with constant input torque and field-oriented control (FOC) for the speed of a permanent magnet synchronous generator (PMSG). For the study, a cosimulation between finite element analysis (FEA), power electronics simulation (Simplorer) and model-based control design software (Matlab Simulink) is used. The analyzed system was previously validated in the experimental set-up with DS1103 card. The initial short circuit currents, although higher than nominal, did not present a risk of demagnetization for the generator in the analyzed working conditions, namely magnets temperature and speed. Nevertheless, the acceleration during the fault, due to constant input torque, and the control scheme with a constant speed reference, provided a greater demagnetizing current while breaking the machine speed back to the pre-fault reference. The cosimulation results bring higher fidelity to the modeling and analysis of the behavior of the PMSG under FOC.
In the analysis of robots and mechanisms, the computation of velocities and accelerations is frequently needed. In other cases, higher-order kinematic formulas such as the jerk and the jounce/snap—the third and fourth-order time derivative of the position vector, respectively—are also needed. While procedures for computing velocity and accelerations can be found elsewhere in the literature, the computations of the jerk and especially of the jounce/snap are not so common. In this study, a novel formulation to compute all kinematic quantities up to the fourth order, based on the dual number is presented. In dual number formalism, kinematic analysis equations are computed automatically once the position vector is constructed. This provides a concise and efficient method to conduct kinematic analysis. To prove the validity of the implemented formulation, we compute the velocity, acceleration, jerk, and jounce/snap for the end effector of the RC robot manipulator. In addition, we compute the jerk and the jounce/snap for the coupler point of the spherical 4 R mechanism.
13 January 2024 in Journal of Computational Science