Real-time implementation of IPM motor protection using artificial neural network

This paper presents an online protection scheme for three-phase interior permanent magnet (IPM) motors using artificial neural network. The proposed protection scheme is developed and implemented in real-time using the ds1102 digital signal processor (DSP) board. In this work, a two-layer feed-forward neural network (FFNN) with sixteen inputs and single output is designed and trained off-line with experimental data using the back-propagation algorithm. An experimental setup is developed to accommodate the on-line testing and to carry out the protection of IPM motors. Three types of faults such as single line to ground (L-G) fault, line-to-line (L-L) fault, and single phasing fault are investigated. The technique is evaluated and tested on-line on the laboratory 1-hp and 5-hp IPM motors using the DSP board. The laboratory results show that the proposed technique is able to detect the faulted conditions with high accuracy.