Neural network approach for stator current sensor fault detection and isolation for vector controlled induction motor drive

Kamil Klimkowski, Mateusz Dybkowski

2016 · 16 citations · 14 references

Concepts

Abstract

The paper deals with the fault tolerant Direct Field Oriented Control of an induction motor drive. Influence of the stator current sensor general failures to the properties of the drive system is presented. To the sensor fault detection the artificial neural network approach is proposed. After the fault detection, drive topology is changed based on the sensor or control topology redundancy. Drive is tested in simulation and in the laboratory set-up with 1.1kW machine and DS1202 card under faulted and post-fault conditions, in wide speed reference changes.

References

14