1,721,096 research outputs found
Neural networks aided on-line diagnostics of induction motor rotor faults
No full textAn improvement of induction machine rotor fault diagnosis based on neural network approach is presented. A neural network can substitute in a more effective way the faulted machine models used to formalize the knowledge base of the diagnostic system with suitably chosen inputs and outputs. Training the neural network by data achieved through experimental tests on healthy machines and through simulation in case of faulted machines, the diagnostic system can discern between `healthy' and `faulty' machines. This procedure substitutes the statement of a trigger threshold, needed in the diagnostic procedure based on the machine models
CLASSIFICATION OF DIAGNOSTIC INDEXES FOR FIELD ORIENTED INDUCTION MOTOR DRIVES
No full textNoninvasive diagnosis of incipient faults of mains supplied electric machine is a wide spread technology, that allows cost-savings. Inverter-fed machines are becoming commonly available, thanks to technology advances. The traditional MCSA fails for closed loop machines, because of regulators impact on electrical quantities. The paper includes a study on the stator faults impact on controlled variables and machine input variables in function of drive operating frequency. This study is aimed at the development of new diagnostics indexes fitted to closed loop machines faults
Survey of neural network approach for induction machine on-line diagnosis
No full textFault detection and diagnosis is currently a very important problem in induction machine management. Both model-based method and expert systems have been suggested to solve the problem. Recently neural networks have been advocated as a possible technique to handle diagnostic tasks providing them with an effective improvement. Neural networks can be applied autonomously or can integrate with existing diagnostic tools. Several architectures for fault diagnosis are studied. In this paper the attention is focused to the multilayer perceptron and to the self organizing map networks which, with different features, seem best suited for induction machine diagnostic tasks. The application of the different neural architectures to specific problem by practical examples is discussed. In particular it will be shown that the synergy between the two mentioned neural architectures provides a global diagnosis approach: NNs specifically trained for dealing with certain tasks are the basic elements. A first level NN classifies the fault, then several second level fault specific NNs (FS-NNs) for stator short-circuits, bearing damages, rotor bar breakages etc. evaluate the fault severity
Neural network aided on-line diagnostics of induction machine stator faults
No full textA Multy Layer Perceptron Neural Network able to recognize inter-turn short-circuits in the stator of induction machines is presented. The network is inserted in an on-line diagnostic system which utilizes the machine voltages and currents as input signals. The current computed by a faulted machine models are analyzed with the aim to evidence the variables more suitable to characterize the short circuit's effects. These variables, properly normalized, are the input data sets for the learning process of the network, which is therefore applicable to a class of induction machines
Monitoring of induction machines load torque disturbances: An alternative NN-based method
No full textThis paper addresses the problem of the real time rebuilding of the load torque disturbances in asynchronous machines. Since the load pattern modifies the motor's supply current, it should be possible to use the current pattern to rebuild torque pattern, utilizing the machine itself as a torque sensor. In the paper the problem is studied utilizing both relationships developed under simplifying assumptions and a more complex model of the machine. The results obtained are compared with the experimental ones. Reference is made to low frequency torque disturbances, that cause a quasi-stationary machine behavior. It is shown that a Neural Network approach can be an alternative and efficient method for the torque pattern recognition
- …
