PMDC Motor Parameter Estimation Using Bio-Inspired Optimization Algorithms

The precise estimation of the motor parameter is essential to design the appropriate controller. The main goal of this paper is to estimate the parameters of permanent magnet dc (PMDC) motor used in a wheelchair, applying standard as well as a dynamic particle swarm optimization (PSO), ant colony optimization (ACO), and artificial bee colony (ABC) along with experimental methods. The electromechanical, mechanical, and electrical parameters, such as torque constant, back-emf constant, moment of inertia, viscous friction coefficient, armature inductance, and resistance are estimated using both the experimental and optimization methods. The motor is modeled in Matlab/Simulink R2015a using the estimated motor parameters and studied the performance with different loading conditions starting from no-load to full-load. The simulated results of motor performance with estimated parameters are compared with the experimental load test results. The results showed that the PMDC motor parameters estimated from dynamic PSO with varying inertia weight as well as ABC algorithm have comparatively very less speed and current error than standard PSO, dynamic PSO with constant inertia weight, and ACO algorithms. Furthermore, parameters from dynamic PSO with varying inertia weight showed speed as well as current error less than 0.5%, and the ABC algorithm shown current error slightly more than 0.5%. However, the analysis of variance tests shown no significant difference in current and speed performance with parameter estimated from ABC and dynamic PSO with varying inertia weight. Furthermore, ABC algorithm convergence is faster than dynamic PSO with varying inertia weight. But parameters estimated from dynamic PSO with varying inertia weight are precise and may be appropriate for the design of the motor controllers.