Inertia Decoupling Identification Strategy Based on Disturbance Torque Adaptive Observation for PMSM Drives

The acceleration errors generated during the speed sampling process decrease the inertia identification accuracy of permanent magnet synchronous motor (PMSM) drives under low acceleration conditions. This article proposes an inertia decoupling identification strategy based on disturbance torque adaptive observation for PMSM drives. By revealing the impacts of acceleration errors, the relationship among the sampling windows, the inertia identification error, and the disturbance torque observation error is obtained. On this basis, the sampling windows for the inertia identification and the disturbance torque observation are adaptively adjusted under different speed and acceleration conditions, which expands the application range of inertia identification. Besides, according to the coupling characteristics between the inertia and the disturbance torque, an adaptive linear neuron network-based iterative mechanism is constructed. The inertia identification and disturbance torque observation can be iteratively updated, which improves the inertia decoupling identification accuracy. The stability and convergence are analyzed in detail. The experimental results verify the effectiveness of the proposed method.