FPGA-based entropy neural processor for online detection of multiple combined faults on induction motors

For industry, a faulty induction motor signifies production reduction and cost increase. Real-world induction motors can have one or more faults present at the same time that can mislead to a wrong decision about its operational condition. The detection of multiple combined faults is a demanding task, difficult to accomplish even with computing intensive techniques. This work introduces information entropy and artificial neural networks for detecting multiple combined faults by analyzing the 3-axis startup vibration signals of the rotating machine. A field programmable gate array implementation is developed for automatic online detection of single and combined faults in real time.

► Multiple combined faults in induction motors are detected.
► Information entropy and ANN are applied on 3-axis startup vibration signals.
► An FPGA implementation is developed to detect single and combined faults.
► The proposed approach is highly efficient detecting single and combined faults.