A Fault-Tolerant Inverter Using a New Fuzzy-Based Detection Algorithm

With the increasing acceptance on the role of inverters in the drive control applications, the concern for the reliability of the system also has increased due to the addition of the switching device in the control system, which makes it more prone to disturbances. This renders the use of inverter based control systems unreliable for critical applications. The proposed model intends to nullify this unreliability. The following research work aims at designing a 3-phase inverter tolerant to the internal faults of the inverter and thus providing the smooth operation of the power system. A new strategy, based on the change pattern of the mean of the difference of the line currents, is proposed for detecting inverter switch open and switch short faults. A fuzzy-based change detection mechanism is designed with other current parameters taken as triggering parameters. Subsequently a phase redundant inverter is modelled based on the proposed fault detection technique. The inverter takes advantage of a redundant leg to maintain the required power capability in the emergence of a fault. The inverter is efficient in detecting both switch short and switch open faults, classifying the faulty switches and performing the fault tolerant measures to obtain the pre-fault parameters status.