Inter-turn Short Circuit Fault Detection in Permanent Magnet Synchronous Motors Through Machine Parameter Identification

AC motors, specifically Permanent Magnet Synchronous Motors (PMSM) have been gaining popularity in industrial applications due to their high efficiency, high power density and high torque generation capabilities. Apart from industrial applications, the market share of these motors has seen significantly increasing use in emerging technologies such as Electric vehicles (EVs), Hybrid Electric Vehicles (HEVs) as traction motors, and use in electric ships, and electric airplanes. To ensure safety and reliability of operation in safety critical applications, continuous health condition monitoring and fault diagnosis of these motors is essential. A robust and accurate fault diagnosis and condition monitoring reduces downtime and unplanned catastrophic failures, shutdowns, and maintenance, which saves both time, money, and labor. Through the available drive sensors such as current and voltage sensors, current and voltage information can be acquired and used to diagnose different faults in the motor. This dissertation presents condition monitoring of stator Inter-Turn Short Circuit (ITSC) faults in PMSM motors. ITSC faults make up more than 40% of all PMSM electrical faults, mainly caused by extreme mechanical, thermal, and electrical stresses. Other factors such as high dv/dt and chemicals accelerate insulation material aging and result in ITSC faults. ITSC fault causes high circulating currents to be induced in the stator of the motors, and if not properly managed, can result in local overheating which leads to further winding degradation and melting, which can result in complete shutdown of the machines operation. Thus, to avoid unexpected machine shutdowns, ITSC fault must be monitored, diagnosed, and managed. This dissertation employs continuous monitoring of machines electrical parameters such as the stator resistance (Rs) and both d-axis and q-axis inductances (Ld and Lq) to detect any changes which may be a result of ITSC faults. Upon detecting changes in parameters, ITSC fault is diagnosed by isolating it from other closely similar faults such as eccentricity faults. The techniques employed in this dissertation eliminates the use of extra sensors in the drive, and only uses current sensors to acquire feedbacks, and uses voltages inside the micro-controller to estimate different parameters for diagnosis purposes.