Interturn Fault Detection in PMSMs: Two Adaptive Observer-Based Noise Insensitive Solutions

In this article, we address the problem of online detection of interturn short-circuit faults (ITSCFs) that occur in interior- and surface-mounted permanent magnet synchronous motors (PMSMs). We propose two solutions to this problem: 1)a very simple linear observer and 2) a generalized parameter estimation-based observer, that incorporates a high performance estimator—with both observers detecting the short-circuit current and the fault intensity. Although the first solution guarantees the detection of the fault exponentially fast, the rate of convergence is fully determined by the motor parameters that, in some cases, may be too slow. The second observer, on the other hand, ensures <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">finite convergence time</i> (FCT) under the weakest assumption of interval excitation (IE). To make the observers <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">adaptive</i>, we develop a parameter estimator that, in the case of surface-mounted motors, estimates online (exponentially fast) the resistance and inductance of the motor. It should be underscored that, in contrast with existing observers (including the widely popular Kalman filter) that provide <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">indirect</i> information of the fault current, our observers provide an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">explicit</i> one—namely the amplitude of the fault current. An additional advantage of the observers is that they <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">do not require</i> the knowledge of the motor currents, making them insensitive to current measurement noise. The performance of both observers, in their linear and generalized parameter estimation-based versions, is illustrated with realistic simulation studies.