Circuit-based induction motor drive reliability under different control schemes and safe-mode operation

This paper proposes a reliability modeling method for induction motor drives based on an equivalent dq 0 circuit model. The machine input voltages or currents are modeled as sources dependent on states and commands. This enables reliability analysis for different control strategies. The method utilizes failure modes and effects analysis, Monte Carlo simulations, and Markov models. Different controllers, in addition to faults in the machine, power electronics, and sensors, are addressed. The control methods considered in the analysis include constant volts per hertz (V/f), indirect field-oriented control (IFOC), direct torque control (DTC), and feedback linearization control (FLC). Direct line operation (DLO) is also considered. The proposed circuit-based reliability modeling method is versatile and easy to implement in commercial circuit simulators. The simulation model, which is experimentally verified, uses a reliability tool to evaluate the mean time to failure (MTTF) of the drive. Results show that, as expected, V/f control has the highest MTTF, due to its independence from sensor feedback, but poor dynamics. Among the closed-loop controllers, IFOC and DTC are better for fast dynamics, simplicity, and acceptable MTTF, but FLC is shown to have the longest MTTF. A safe-mode control scheme is proposed under which the drive is shown to have an even longer MTTF than V/f.