To realize a high-precision online identification of multiple electrical parameters of a permanent magnet synchronous linear motor (PMSLM), a dual-model-based recursive least squares online identification algorithm for electrical parameters is proposed in this paper. First, according to the dq-axis voltage equation of the motor, the first model for identifying the stator resistance and permanent magnet flux linkage and the second model for identifying the q-axis inductance and d-axis inductance are established, respectively, and the two identification models are combined cyclically. Second, based on the dual-model structure, the recursive least squares algorithm is used to realize the online identification of electrical parameters, and a piecewise variable forgetting factor with saturation characteristics is proposed for the characteristics of a large number of dynamic processes during the operation of PMSLM. Meanwhile, the error voltage caused by the power switch non-ideal factor is compensated, which further improves the identification accuracy. Finally, simulation and experimental results prove the effectiveness of the identification algorithm, indicating that it has advantages such as a fast convergence speed, a high accuracy of identification results, and applicability under multiple working conditions.