Switching model based on-line multi-parameter identification method for permanent magnet synchronous motor without injection
By establishing an IPMSM mathematical model that considers cross-saturation effects and iron losses, and using a switching model and RLS algorithm, the underrank problem in the traditional model is solved, achieving accurate identification of IPMSM electrical parameters, fast convergence, and anti-interference capabilities. It is applicable to fields such as industrial servo control, electric vehicles, and robot joint motors.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHONGQING UNIV
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing online parameter identification methods suffer from underrank problems, making it difficult for electrical parameter identification results to converge. Furthermore, the traditional IPMSM mathematical model does not consider magnetic circuit saturation, cross-saturation effects, and core losses, which affects the accuracy of motor parameter identification.
An IPMSM mathematical model considering cross-saturation effect and iron loss is established. Electrical parameters are identified using a switching model. Online parameter identification is achieved through seven-segment SVPWM modulation and RLS algorithm. The nonlinear characteristics of the magnetic circuit are described using dynamic inductance and cross-coupled inductance. The electrical parameters are accurately characterized by combining iron loss current and torque current models.
It achieves accurate characterization of motor nonlinear characteristics and iron loss, breaks through the underrank limitation, adapts to arbitrary initial values, requires no additional signals, converges quickly, has strong anti-interference ability, can simultaneously identify all key parameters, and dynamically track parameter changes.
Smart Images
Figure CN122419296A_ABST