A Novel Three-Phase Inverter Fault Diagnosis System Using Three-dimensional Feature Extraction and Neural Network

Abstract

In general, fault diagnosis and classification is concerned about monitoring a system, identifying the fault occurrence and pinpointing the exact fault location. Fault diagnosis and fault-tolerant systems are crucial for modern electrical and electronic system safety. Inverters are most frequently used DC–AC power converters. DC–AC three-phase inverter faults are very common to occur and need an effective and robust fault diagnosis system for protection and isolation. A new fault detection and classification technique is introduced in this paper. Feature extraction system considers three-phase voltage pattern plot in 3D, and the neural network are used to pinpoint the fault location. The testing process is conducted under simulation and physical environment to confirm its expediency. Proposed technique and experimental results are discussed in detail. Simulation experiment followed by practical implementation on SPWM three-phase inverter is included with results to ensure the system accuracy and reliability.