Adaline Neural Network Based Online Parameter Estimation for Surface-mounted Permanent Magnet Synchronous Machines

An Adaline neural network (NN) based estimation strategy is proposed for estimating the winding resistance,inductance and rotor flux linkage of surface-mounted permanent magnet synchronous machines (SPMSM).The proposed method does not need any machine design parameter information for estimation and only needs to sample the stator currents,voltages and rotor speed.In the proposed estimation,the stator winding resistance value is firstly estimated at PMSM standstill and the estimated winding resistance value is then used for estimating the rotor flux linkage and inductance when the PMSM is started.Further,the estimated rotor flux linkage value is used for online estimating the variation of stator winding resistance.The validity of proposed method is verified by experiments which show that it is effective in estimating the winding resistance,inductance and rotor flux linkage.In addition,it is also effective in online tracking the stator winding resistance variation when the PMSM is with load.