Fault Detection and Classification in Power Electronic Circuits Using Wavelet Transform and Neural Network

Abstract

Problem statement:The identification of faults in any analog circuit is highly required to ensure the reliability of the circuit. Early detection of faults in a circuit can greatly assist in maintenance of the system by avoiding possibly harmful damage borne out of the fault. Approach: A novel method for establishing a fault dictionary using Wavelet transform is presented. The Circuit Under Test (CUT) is three phase single level inverter. The transform coefficients for the fault free circuit as well as for the simulated faults of CUT are found. The Wavelet transform is applied to the output of CUT and Standard Deviation (SD) of the transform coefficients are extracted. Using the transform coefficients, fault dictionary has been formed. In order to identify the type of fault, a neural network classifier has been utilized. Results: The compatibility of wavelet analysis with the various classification techniques for fault diagnosis has been illustrated in this study. The results of the study demonstrate the suitability and viability of wavelet analysis in fault diagnosis of power electronic circuits. Conclusion: The proposed approach is found to be more reliable in accurate identification and isolation of faults using fault dictionary. Moreover, the neural classifier improves the efficiency of the system as neural networks do not require prior knowledge as they are capable of learning and evolving through a number of learning algorithms.

Fault diagnosis in analog electrical circuits: data-driven method

“…-the simulation of a circuit for different faults to generate training data for an artificial neural network is presented in [2,15]; -the method of establishing a fault dictionary using Wavelet transform is developed in [13]; -in [6,7] the probabilistic graphical models are used; here faulty components are identified by looking for high probabilities for values of characteristic magnitude that deviate considerably from the nominal values; -the probabilistic model-based approach presented in [12] is formally founded and based on Bayesian network and arithmetic circuits; -the method based on diagnostic observers of states developed in [5,[20][21][22] considers linear and nonlinear circuits as dynamic systems; it is necessary to stress that the method suggested in [21] can be used for diagnosis in electrical circuits containing non-smooth nonlinearities such as hysteresis and saturation.…”

Wavelet-based sensor validation: Differentiating abrupt sensor faults from system dynamics

“…In [8] threshold level is computed in the assumption of chi-square distribution of wavelet coefficients. In [11] neural network is applied to fault diagnosis taking standard deviations of wavelet coefficients as inputs. Another approach is to analyze the difference between wavelet coefficients for various decomposition levels [12], [13].…”

Multiresolution Analysis Based Effective Diagnosis of Induction Motors

“…Many methods, widely studied in literature, have been adopted for the monitoring of electric motors and especially induction motors, as well as the diagnosis of their defaults such as artificial intelligence based methods (Kanthalakshmi and Manikandan, 2011;Prasannamoorthy and Devarajan, 2010;Bouzid et al, 2008;Martins et al, 2007;Tallam et al, 2007), signal processing based methods (Prasannamoorthy and Devarajan, 2010;Kia et al, 2007;Jung et al, 2006), automatic and control based methods (Kanthalakshmi and Manikandan, 2011;Angelo et al, 2009) and a combination of them (Prasannamoorthy and Devarajan, 2010).…”

“…In fact, wavelet technique is a very useful, powerful and efficient tool for monitoring and diagnosis machines purpose because of its capabilities to perform signal content analysis in both time and frequency domains (Prasannamoorthy and Devarajan, 2010;Cusido et al, 2008;Ukil and Zivanovic, 2005;Truchetet and Laligant, 2004;Chow and Hai, 2004;Lee et al, 2004). This is of a great importance for the detection of changes starting from the motor signals and especially abrupt and time localised changes caused by defaults occurrence.…”