The invention discloses a permanent
magnet wind driven generator
inductance parameter identification method based on a JA model and data fusion, and the method comprises the following steps: decomposing the
magnetization intensity of a ferromagnetic material into a reversible component and an irreversible component, deducing the nonlinear relation between
inductance and current through combining the
Ampere loop law, introducing a cross saturation coefficient to correct the inter-axis
coupling effect, and identifying the
inductance parameter of the permanent
magnet wind driven generator. Establishing an inductance basic model in a magnetic saturation state; integrating the influences of current, temperature and frequency to form a composite inductance model; establishing a state equation based on the
voltage equation and the
flux linkage equation;
dynamic prediction and updating of states and parameters are realized through extended Kalman filtering, parameter correction is driven by using
flux linkage errors in combination with a model reference
adaptive system, and finally, weights are adjusted based on disturbance errors of rotational
inertia and friction coefficients, so that fusion identification is realized. According to the invention, through physical mechanism modeling and multi-field
coupling correction, high-precision tracking of inductance parameters under a wide working condition is realized, and the identification precision and generalization ability of a magnetic saturation region are significantly improved.