Parameter Estimation of PMSM Based on Improved Secretary Bird Optimization Algorithm

ABSTRACT Accurate and efficient parameter estimation is a critical prerequisite for achieving high‐performance control of PMSMs in electric aircraft propulsion.To overcome the trade‐off between accuracy and computational cost in existing methods, this paper proposes an improved secretary bird optimization algorithm (ISBOA). Key enhancements encompass Tent chaotic mapping for population initialization, stochastic differential mutation to maintain solution diversity, random interaction migration for escaping local optima, and adaptive step‐size adjustment to balance exploration and exploitation. Moreover, a non‐intrusive parameter estimation framework is introduced, enabling ISBOA to be seamlessly embedded within a standard motor control system as a low‐overhead concurrent process. This allows online parameter estimation to proceed without disrupting real‐time control tasks. Experimental validation on a TMS320F28335 DSP platform demonstrates that ISBOA achieves parameter estimation errors below 2% while exhibiting accelerated convergence compared to benchmark algorithms.