Offline Recursive Identification of Electrical Parameters of VSI-Fed Induction Motor Drives

Accurate estimation of electrical parameters of voltage-source-inverter-fed induction machine (IM) drives is very important while employing high-dynamic-performance control schemes, such as vector control. Parameter estimation schemes reported in the literature can estimate only four parameters ( $R_s$ , $L_{ss}$ , $\sigma$ , and $\tau _r$ ) independently out of five electrical parameters ( $R_s$ , $R_r$ , $L_{\sigma s}$ , $L_{\sigma r}$ , and $L_m$ ), when the core loss resistance is neglected. All the five parameters are not independently identified. This article proposes a parameter estimation method that independently identifies all the six electrical parameters ( $R_s$ , $R_r$ , $R_c$ , $L_{\sigma s}$ , $L_{\sigma r}$ , and $L_m$ ) of the IM, including the core loss resistance. Besides, the variation of core loss resistance with frequency is also discussed in this article. The Kalman filter algorithm is used to estimate machine parameters in the proposed method. In the proposed method, a sine-triangle pulsewidth modulation signal is used as input excitation, instead of pseudorandom binary sequence signals. The proposed estimation method is validated using both simulation and experimental results.