Induction Machine Stator Fault Tracking Using the Growing Curvilinear Component Analysis

Detection of stator-based faults in Induction Machines (IMs) can be carried out numerous ways. In particular, the shorted turns stator windings IM are among most common industry. As a matter fact, IMs come with pre-installed current sensors for purpose control and protection. At this aim, using only fault detection has become recent trend nowadays as it is much cheaper than installing additional sensors. The three-phase signatures have been used study to observe effect inter-turn respect healthy condition IM. pre-processing faulty done via in-built DSP module dSPACE after which, these passed into MATLAB ® software further analysis AI techniques. authors present Growing Curvilinear Component Analysis (GCCA) neural network that capable detecting follow evolution signature, making online possible. For purpose, topological manifold evolution, which fundamental step calibrating GCCA network. effectiveness proposed method verified experimentally.