Distributed Fault Diagnosis System based on Wireless Sensor Networks

Abstract

This article presents the development of a distributed fault diagnosis and monitoring system whose remote nodes are responsible for data collection and distributed analysis to identify problems that could lead to critical faults in industrial processes or systems. The developed intelligent remote node was implemented with MCU LPCXpresso54114 connected to a ZigBee protocol wireless sensor network through XBee communication module. The gateway node is a Raspberrry PI with HTTP communication and JSON format to the PI System industrial monitoring system database. Motor Current Signature Analysis (MCSA) was implemented and validated to identify interturn faults of induction motors. The developed platform is a tool to perform comparison and validation of analysis techniques, indicators, and fault classification, because there are different combinations that can be applied to improve diagnosis reliability, fault observability, differentiation between fault conditions, classification accuracy, tolerance to transients, sensitivity, among others.

Keywords

• distributed analysis
• fault diagnosis
• induction motor
• motor-current signature analysis
• stator current
• wireless sensor networks
• zigbee