Flux Estimation In Switched Reluctance Motors Using a Two Dimensional Least Square Method Approximation

Switched reluctance (SR) motors are among the most robust motors founnd in industry. The adoption in practice is limited by the difficulty of the drive electronics and control. This is due to the highly non linear characteristics of SR motors. Existing control methods used in SR motors are often based on look-up tables of the torque and the flux characteristics that are derived from magnetic finite element analysis (FEA). This paper proposes a new method of control that will not rely upon look-up tables of the flux values. It is replaced by a two-dimensional least squares polynomial estimate of the magnetic flux linkage. The variables are phase current and rotor position. To obtain the polynomial coefficients, magnetostatic simulations are run for several blocked rotor positions. The phase current is increased until saturation values are achieved. During runtime operation, the rotor position and the current values will be evaluated by the algorithm which results in flux linkage values that are subsequently used by the motor control scheme.