Professional Documents
Culture Documents
Stator Fault Analysis of Three-Phase Induction Motors Usinginformation Measures and Artificial Neural Networks
Stator Fault Analysis of Three-Phase Induction Motors Usinginformation Measures and Artificial Neural Networks
Stator Fault Analysis of Three-Phase Induction Motors Usinginformation Measures and Artificial Neural Networks
a r t i c l e i n f o a b s t r a c t
Article history: The three-phase induction motors are considered one of the most important elements of the industrial
Received 21 May 2016 process. However, in this environment, these machines are subject to electrical and mechanical faults,
Received in revised form which may cause significant financial losses. Thus, the purpose of this paper is to present a pattern
23 September 2016
recognition method for the detection of stator windings short circuits based on measures of mutual
Accepted 29 September 2016
information between the phase current signals. In order to validate the proposed patterns, feature vectors
Available online 26 October 2016
obtained from normal and faulty motors are applied to two topologies of artificial neural networks.
The classification results presented accuracies over 93% even when the motors were subject to several
Keywords:
Three-phase induction motor
conditions of load torque and power supply voltage unbalance.
Artificial neural networks © 2016 Elsevier B.V. All rights reserved.
Information measures
Stator fault
1. Introduction [10], these measures are also used for rotor eccentricity faults along
with stator winding inter-turn short circuit. In [11], the authors
The three-phase induction motor (TIM) is one of the most com- showed that Park vector modules combined with Hilbert Trans-
mon machines used to transform electrical energy into mechanical form contain relevant information for detecting short circuit faults
energy in industrial processes. This is mainly due to the low acqui- in TIM stator windings. In addition, in [12], wavelet transform was
sition and maintenance costs, reliability, robustness to aggressive used for feature extraction of current signals for the identification of
environments, as well as being easily adapted to several load con- fault severity, even with the machine subject to different operating
ditions [1–3]. Despite these favorable characteristics, TIMs can conditions.
present electrical or mechanical faults that may be caused by A more recent approach of pattern recognition for motor fault
prolonged activity times, harsh operating conditions, voltage or detection is the use of information theory (IT) measures such as
current unbalance, among other factors. Even incipient faults can entropy. One example is presented in [13], in which TIM stator
affect the motor performance, which may result in financial losses currents are analyzed using DWT and entropy estimations for fault
[4,5]. Therefore, the search for adequate diagnostic methods is a diagnosis. Also, in [14], multiscale entropy estimations from cur-
recurrent research topic in the past few years, aiming to detect the rent and vibration signals are used as patterns for stator fault
faults in an early stage and avoid unwanted production stops [6]. detection. Similar entropy measures obtained from the stator cur-
Among the electrical failures, up to 36% of the cases are due to rent are presented in [15], where they are applied to an ANN for
stator winding faults [7]. Specifically regarding this kind of fault, in fault prediction. Entropy is also employed in [16], along with sup-
[8], Discrete Wavelet Transform (DWT) is used to obtain patterns port vector machines (SVMs) and ANNs, for the detection of TIM
from high frequency components that are classified using Artifi- faulty bearings and broken rotor bars. Finally, in [17], bearings faults
cial Neural Networks (ANN). Also, in [9], statistical measures of the are also the object of analysis, but using relative entropy and DWT
stator current, such as mean, variance, skewness, and kurtosis, are of the TIM current signals.
used as inputs to a multilayer perceptron (MLP) ANN. Similarly, in Another IT measure that has recently being applied in fault
diagnosis is mutual information (MI). It is a similarity measure
used in feature extraction methods that has applications in many
∗ Corresponding author. areas, such as in speech intelligibility prediction [18]. For motor
E-mail address: prscalassara@utfpr.edu.br (P.R. Scalassara). fault detection, in [19], MI measurements of vibration signals
http://dx.doi.org/10.1016/j.epsr.2016.09.031
0378-7796/© 2016 Elsevier B.V. All rights reserved.
348 G.H. Bazan et al. / Electric Power Systems Research 143 (2017) 347–356
in time-frequency domain are used to evaluate the condition of considered to diagnose stator faults. The TIMs were driven by three
helicopter bearings and shafts. Similarly, in [20], MI is used to different types of frequency inverters and the currents amplitudes,
assess the correlation of diesel engine signals in order to obtain in time domain, were used to classify stator fault severity. Also, in
the most informative ones for fault diagnosis. One final example [23], the authors proposed an evaluation of several pattern classifi-
is [21], where the authors employed MI in the feature extraction cation methods, namely naive Bayes, k-nearest neighbors, support
step to select the most relevant and less redundant measurements vector machines, ANNs, and decision trees, for multi-fault classi-
from TIM current signals for eccentricity fault identification. fication (rotor, stator, and bearings faults) using the TIM currents
Considering this recent application of IT measures, the aim of amplitudes in time domain.
this work is to present an alternative approach to the diagnosis It is noteworthy that intelligent systems methods are widely
of TIM stator windings short circuit. This method is based on the used for fault identification of electrical machines, mainly due to
analysis of MI values obtained from delayed line currents of two their ability to classify features without requiring complex math-
TIM phases. Using this tool, it is possible to evaluate the common ematical models. The most usual methods are ANNs [22,40–42],
information of the signals, as well as verify the changes of MI values SVMs [31,43], fuzzy logic [44,45], decision trees [46], and hybrid
due to different TIM operating conditions. systems [30,47]. Thus, based on this, the features obtained with MI
The efficiency of the proposed feature extraction method is measures of the TIM current signals are classified using ANNs in
assessed using two ANN topologies, multilayer perceptron (MLP) this paper.
and radial basis function (RBF), which are proved to be adequate
for motor fault detection as shown in [9,10,22,23]. For that, we
3. Mutual information
performed 814 experimental tests with two TIMs of 0.74 kW and
1.48 kW in steady state and under several operating conditions
MI is a measure of how much information a random variable
that are common in an industrial environment, such as varia-
(RV) has over another one. In other words, it is the uncertainty
tions of load torque and power supply voltage unbalance. It is
reduction of a RV brought by information of the other [48]. This
worth pointing out that the technique has a characteristic of
quantity describes the similarity between time series collected
being low cost, therefore the current signals are acquired nonin-
simultaneously from a system under study [49,50].
vasively, using Hall-effect sensors, and no frequency domain tools
Thus, by estimating the MI between the TIM current signals, it
are employed.
is possible to quantify the association between them, allowing an
This paper is structured as follows. Section 2 presents aspects
efficient monitoring of the machine operating conditions. Consid-
related with TIM stator winding faults. In Section 3, the main
ering two RV X and Y, the MI can be estimated using (1), where x
concepts of MI are addressed. Section 4 presents aspects of the pro-
and y are realizations of the RVs, p(x) and p(y) are the probability
posed methodology and experimental setup. The results are shown
density functions (PDFs) of X and Y, respectively, and p(x, y) is the
in Section 5, and finally, the conclusions in Section 6.
joint PDF of the them.
2. Stator faults
p(x, y)
I(X, Y ) = p(x, y)log2 dxdy (1)
Y X
p(x)p(y)
In this section, aspects related to TIM stator winding faults are
presented, which are one of the main problems found in elec- In practice, the PDFs are approximated by the signals normalized
tric motors [4,7,24,25]. Also shown are brief descriptions of other histograms and the expected values are estimated by time averages.
researches addressing the identification of these faults at an early Since the winding faults tend to present asymmetrical effects
stage, aiming to avoid unscheduled downtime and reduce mainte- in the TIM line current signals and MI is a similarity measure, this
nance costs. quantity should reflect such variations. In this work, we analyzed
The most common defects associated with TIM stator are phase the delayed MI of two TIM phase current signals. This quantity is
to ground, phase to phase, and coil to coil short circuits [26]. When dependent on a time shift of samples that is applied to one of the
there is a short circuit involving just a few coil turns, the stator signals before estimation using (1). Delayed MI has been used in
insulation starts to deteriorate; this could rapidly develop into a dynamic structural analysis of complex systems, due to its capacity
more serious fault and affect the motor operation [4,7,27]. of quantifying the dependence between RVs, seeking the similarity
In order to monitor the TIM operating conditions and identify between them relative to the time delay [51–53].
these faults, one must use a proper signal processing technique In order to obtain a reasonable analysis of the TIM current vari-
for the extraction and selection of relevant features of the signals. ations caused by the stator faults, several MI values are estimated
The motor current signature analysis is a commonly used method with shifted versions of the phase B current signal. As an example,
because it is simple, noninvasive, and can be automated [28]. This Fig. 1(a) presents a part of a three-phase current acquired during 3 s
approach is focused on the current spectral analyses and has been on a normal TIM. The MI values obtained using the phase A current
widely used for TIM fault diagnosis [29–31]. signal and 150 delayed versions of the phase B current signal are
In the literature of fault detection, there are other examples shown in Fig. 1(b).
of signal processing techniques based on spectral decomposi- In the next section, a detailed description of the methodology
tion using short-time Fourier transform and Wavelet transform employed in this work is presented.
[8,12,32–34]. In other papers, applications using these transforms
with the Hilbert transform are presented [11,35]. Other authors
propose methods for TIM fault diagnosis based on the analysis of 4. Proposed approach
vibrations [5,36], magnetic signals [5,37,38], active and reactive
power [7], temperature [39], among other variables. As mentioned in the previous sections, the purpose of this work
One disadvantage of most of these methods is the requirement is to identify stator short circuit faults by using MI values of the
for high-priced sensors, which increase the cost of the diagnostic TIM current signals. Fig. 2 presents a diagram of the proposed fault
system. In order to avoid that, some authors are looking for other detection method explained in this section. In order to validate
noninvasive methods, in which measurements are provided by low this approach, experimental tests were performed. Next, details are
cost sensors, and using time domain strategies. In [22], an alterna- presented about the current signals acquisition, and feature analy-
tive approach to the traditional intelligent systems methods was sis and classification using ANN.
G.H. Bazan et al. / Electric Power Systems Research 143 (2017) 347–356 349
Fig. 1. (a) Part of a three-phase current acquired on a normal TIM showing phases A (solid line), B (dashed line), and C (dotted line). (b) Estimated MI values obtained using
phase A and 150 delayed versions of phase B (using the whole 3 s signals).
the short circuits. Both machines have stators with six coils, 306 Total 110 66
turns per phase for the 0.74 kW TIM and 780 turns for the 1.48 kW
TIM. For this latter, the minimum possible short circuit level was
3%. the 0.74 kW TIM were varied as 0%, −2%, and −4% of phase A, and
Fig. 3 presents the experimental workbench used for signal 0%, +2%, and +4% of phase B. For the 1.48 kW TIM, it was 0%, −2%,
acquisition. This system was designed to monitor TIM electrical and −4% voltage variations of phase A.
and mechanical characteristics, namely voltage, current, vibration, The combinations of the load torque and power supply volt-
torque and speed. In addition, it allows emulation of several oper- age unbalance (including the balanced condition) resulted, for each
ating conditions such as variations of load torque and power supply short circuit fault level, in 110 experiments for the 0.74 kW TIM
unbalance. More details are presented in [17,22,23]. and 66 for the 1.48 kW TIM, as shown in Table 1. The sampling fre-
The database consists of 814 experiments, 550 performed on the quency was 15.5 kHz, amplitude quantization of 16 bits, and 5 s of
0.74 kW TIM and 264 on the 1.48 kW one, both operating in steady acquisition time.
state under various conditions commonly found in the indus-
trial production process as mentioned before. The machines were 4.2. Feature analysis
connected directly to the power network and tested under condi-
tions ranging from 20% to 120% of their nominal load (0.7–4.9 Nm, The feature extraction method consists of estimating the MI of
in intervals of 0.2 Nm, for the 0.74 kW TIM, and 0.9–9.3 Nm, in inter- the TIM phase A current signal and delayed versions of the phase B
vals of 0.4 Nm, for the 1.48 kW TIM). The power supply voltages for signal, considering delays from 0 to 150 samples. From the analysis
350 G.H. Bazan et al. / Electric Power Systems Research 143 (2017) 347–356
of the MI versus delay curves of the acquired signals for all the in a slight shift right but with peak decrease compared with the
operating conditions described, one can notice that each of these reference curve. But, the same variations of phase B voltage causes
conditions presents a characteristic signature. In order to illustrate a shift left and also a peak decrease of the curves.
this, Figs. 4 and 5 show the curves obtained for the 0.74 kW TIM, The same analysis can be extended to TIMs with short circuits in
with no short circuit, but subject to variations of load torque and the stator windings. Figs. 6 and 7 illustrate these situations for the
power supply voltage unbalance, respectively. 0.74 kW TIM operating with 10% of stator short circuit of phase A,
From the analysis of Fig. 4, one can observe that an increase in variations of load torque and power supply voltage unbalance are
load level leads to a slight shift right and peak increase of the MI respectively considered in each figure. The behavior of the curves
characteristic curve in relation to the reference curve, that is, the for the faulty TIM is similar to what is observed for the normal
one with minimum torque and no voltage unbalance. Regarding operating condition. Load torque increase caused a larger shift right
Fig. 5, it is noticeable that variations of phase A voltage also results of the curves, as observed in Fig. 6. Also, considering Fig. 7, the same
Fig. 4. MI versus delay curves obtained for the 0.74 kW TIM with no short circuit, balanced power supply voltage, and variations of load torque.
Fig. 5. MI versus delay curves obtained for the 0.74 kW TIM with no short circuit, 0.7 Nm of load torque, and variations of power supply voltage.
Fig. 6. MI versus delay curves obtained for the 0.74 kW TIM with 10% of stator short circuit of phase A, balanced power supply voltage, and variations of load torque.
352 G.H. Bazan et al. / Electric Power Systems Research 143 (2017) 347–356
Fig. 7. MI versus delay curves obtained for the 0.74 kW TIM with 10% of stator short circuit of phase A, 0.7 Nm of load torque, and variations of power supply voltage.
Table 2
Details of the ANN topologies.
Number of layers 2 2 2 2 2 2
Neurons on the hidden layer 10 20 50 3 5 7
Neurons on the output layer 2 2 2 2 2 2
Training algorithm BP BP BP SO SO SO
Activation function (hidden layer) Sig Sig Sig G G G
Activation function (output layer) Linear Linear Linear Linear Linear Linear
Learning rate 0.1 0.1 0.1 0.01 0.01 0.01
Momentum 0.9 0.9 0.9 – – –
Epochs 526 688 856 – – –
kind of curve shifts is observed when there are voltage variations, stage. Differently, the RBF topology consists of an input layer, only
but with an increase of MI values for unbalance of phase B voltage. one middle layer, and an output layer. Besides that, the neurons
Fig. 8 presents a comparison of the MI curves obtained for the of the middle layer have Gaussian activation functions, instead of
0.74 kW TIM under normal and faulty operating conditions with the logistical or hyperbolic tangent functions used for the MLP.
0.7 Nm load torque and balanced power supply voltage. The sta- The RBF training has two stages, the first is associated with the
tor winding short circuit levels are 1%, 3%, 5%, and 10%. The faults adjustment of the synaptical weights of the middle layer using
caused a shift left of the MI curves proportional to the short circuit an unsupervised learning method that depends only on the input
levels, a decrease of MI values, and an enlargement of the curve. data. The second stage is related to the adjustment of the output
The same behavior shown in Figs. 4–8 for the 0.74 kW TIM was layer weights using the generalized delta learning technique [54].
observed for the 1.48 kW TIM, but omitted here for simplicity. The ANN training and testing were performed in the software
WEKA [55], using cross-validation with the samples randomly
4.3. Feature classification using ANN divided in ten groups. Using this procedure, the network is evalu-
ated ten times, in each time, one group is selected as validation and
For each of the 814 experimental tests, a feature vector of 150 the rest is used for training. At each iteration, the misclassification
MI values was created and used as the input to two ANN topologies: rate of the test set is calculated, and, at the end, the performance is
MLP and RBF networks, which are briefly described next. estimated by the classifier average errors.
The ANNs are mathematical models based on the nervous sys- The details of the ANN topologies are shown in Table 2. In
tems of living creatures. They are defined as a set of processing the next section, the results of the ANN validation stage are pre-
units, the artificial neurons, and have the capacity for acquiring sented.
and maintaining knowledge from information. According to [54],
the essential ANN characteristics are the adaptation through expe- 5. Results and discussion
rience, learning capability, generalization ability, data organization,
fault tolerance, and distributed storage. The proposed ANN architectures were trained according to the
Several ANN topologies are built using one or many neuron details discussed in the previous section. Tables 3 and 4 present
layers, examples are the MLP and RBF networks that are used in the classification errors and accuracies achieved by the MLP and
this work. The former consists of at least one hidden layer and one RBF networks, respectively for the feature vectors obtained from
output layer. This topology uses the backpropagation algorithm, the current signals of the 0.74 kW TIM. The kappa index is also
which has two stages: feedforward and backward. The first is a shown, which is a measure of consistency of the network results.
supervised stage where only the current synaptical weights and According to [22], kappa values higher than 0.8 represent the ideal
thresholds are considered. These values are adjusted in the second concordance of the obtained success rates.
G.H. Bazan et al. / Electric Power Systems Research 143 (2017) 347–356 353
Fig. 8. MI versus delay curves obtained for the 0.74 kW TIM with 0.7 Nm of load torque, balanced power supply voltage, and variations of short circuit levels of phase A.
Table 3 Table 5
Classification results of the MLP network for the 0.74 kW TIM features. Confusion matrices of the MLP topologies for the 0.74 kW TIM.
Classification error 2/550 3/550 3/550 Normal Faulty Normal Faulty Normal Faulty
Classification accuracy (%) 99.64 99.45 99.45
Kappa index 0.989 0.983 0.983 Normal 108 2 107 3 107 3
Mean absolute error 0.0096 0.0095 0.0082 Faulty 0 440 0 440 0 440
Root mean squared error 0.0596 0.0726 0.0661
Relative absolute error 2.999 2.956 2.547
Root relative squared error 14.909 18.158 16.517
Recall 0.996 0.995 0.995 Similarly, for the RBF networks, the reached accuracies were
F-Measure 0.996 0.995 0.995 over 93%, regardless of the number of neurons in the interme-
ROC 1 0.999 1 diate layer. The network with 7 neurons (RBF3) obtained results
Time taken to build model (s) 9.39 18.83 44.41
slightly better than the others, with 95.82% of classification accu-
racy, as shown in Table 4. Despite that the RBF networks presented
low computational costs, their accuracy are relatively lower than
For the 0.74 kW TIM, both ANN topologies achieved accuracies
the MLP networks. Furthermore, other statistical indices, such as
over 93%, but the MLP networks had a better performance than the
the relative absolute error and root relative squared error, which
RBF ones. The MLP network with 10 neurons in the hidden layer
present results over 15% and 46%, respectively, confirmed that, for
(MLP1) reached the highest classification rate of 99.64%. This result
the 0.74 kW TIM, the RBF networks are less suitable in comparison
is slightly higher then those of the other MLP networks with 20 and
with the MLP ones.
50 neurons, which were 99.45%. But the greatest advantage of MLP1
The Kappa indices above 0.8 obtained by the two ANN topolo-
is that it has a lower computational weight as shown in Table 3 for
gies confirm their potential for classification of the TIM condition:
the time taken to build the model.
normal or faulty (stator short circuit). Tables 5 and 6 present the
Other statistical indices are also presented in the tables, such
confusion matrices of the MLP and RBF topologies results, respec-
as mean absolute error, root mean squared error, relative abso-
tively.
lute error, root relative squared error, recall, F-measure, and region
Based on these tables, for the MLP networks, only normal sam-
of convergence (ROC). For the MLP network with 10 neurons, the
ples were misclassified (false positives). This kind of error leads to
values of these indices were 0.0096, 0.0596, 2.999, 14.909, 0.996,
programmed interruption of the industrial process for an unnec-
0.996, and 1, respectively, thus confirming that it is the most ade-
essary maintenance. On the other hand, the false negative would
quate topology for the 0.74 kW TIM.
probably result in an unexpected stop due to the TIM collapse,
which could have a greater financial loss depending on the pro-
Table 4 cess costs. Despite the fact that the RBF network presented false
Classification results of the RBF network for the 0.74 kW TIM features. negative classifications, the accuracy rate and the Kappa index indi-
Type RBF1 RBF2 RBF3 cate that this topology had similar results to the MLP networks, but
with lower processing time.
Classification error 34/550 28/550 23/550
Classification accuracy (%) 93.82 94.91 95.82
Kappa index 0.803 0.843 0.863
Mean absolute error 0.0878 0.0638 0.0493 Table 6
Root mean squared error 0.2258 0.2004 0.1869 Confusion matrices of the RBF topologies for the 0.74 kW TIM.
Relative absolute error 27.378 19.879 15.363
Root relative squared error 56.448 50.109 46.72 RBF1 RBF2 RBF3
Recall 0.938 0.949 0.958
Normal Faulty Normal Faulty Normal Faulty
F-measure 0.938 0.949 0.958
ROC 0.945 0.974 0.967 Normal 90 20 98 12 98 12
Time taken to build model (s) 0.18 0.38 0.31 Faulty 14 426 16 424 5 435
354 G.H. Bazan et al. / Electric Power Systems Research 143 (2017) 347–356
Table 7 Table 9
Classification results of the MLP network for the 1.48 kW TIM features. Confusion matrices of the MLP topologies for the 1.48 kW TIM.
Classification error 1/264 1/264 1/264 Normal Faulty Normal Faulty Normal Faulty
Classification accuracy (%) 99.62 99.62 99.62
Kappa index 0.989 0.989 0.989 Normal 65 1 65 1 65 1
Mean absolute error 0.0043 0.0037 0.0038 Faulty 0 198 0 198 0 198
Root mean squared error 0.0429 0.0409 0.0493
Relative absolute error 1.136 0.993 1.012
Root relative squared error 9.908 9.442 11.374 Table 10
Recall 0.996 0.996 0.996 Confusion matrices of the RBF topologies for the 1.48 kW TIM.
F-measure 0.996 0.996 0.996
RBF1 RBF2 RBF3
ROC 1 1 1
Time taken to build model (s) 4.52 9.08 22.10 Normal Faulty Normal Faulty Normal Faulty
Normal 64 2 65 1 64 2
Table 8 Faulty 5 193 0 198 0 198
Classification results of the RBF network for the 1.48 kW TIM features.
Table 11
Comparison summary of some studies reporting stator fault diagnosis.
Proceedings of the 2012 International Conference on Electrical Machines, [47] M. Seera, C.P. Lim, D. Ishak, H. Singh, Offline, Online fault detection and
IEEE, Marseille, France, 2012, pp. 1538–1544. diagnosis of induction motors using a hybrid soft computing model, Appl. Soft
[39] Z. Gao, T.G. Habetler, R.G. Harley, R.S. Colby, A sensorless adaptive stator Comput. 13 (12) (2013) 4493–4507.
winding temperature estimator for mains-fed induction machines with [48] T.M. Cover, J.A. Thomas, Elements of Information Theory, 2nd ed., Wiley, New
continuous-operation periodic duty cycles, IEEE Trans. Ind. Appl. 44 (5) Jersey, 2006.
(2008) 1533–1542. [49] A. Dionisio, R. Menezes, D.A. Mendes, Mutual information: a measure of
[40] M.-Y. Chow, R.N. Sharpe, J.C. Hung, On the application and design of artificial dependency for nonlinear time series, Physica A: Stat. Mech. Appl. 344 (1–2)
neural networks for motor fault detection – Part I, IEEE Trans. Ind. Electron. 40 (2004) 326–329.
(2) (1993) 181–188. [50] R. Steuer, J. Kurths, C.O. Daub, J. Weise, J. Selbig, The mutual information:
[41] M.-Y. Chow, R.N. Sharpe, J.C. Hung, On the application and design of artificial detecting and evaluating dependencies between variables, Bioinformatics 18
neural networks for motor fault detection – Part II, IEEE Trans. Ind. Electron. (Suppl. 2) (2002) S231–S240.
40 (2) (1993) 189–196. [51] J.M. Nichols, Examining structural dynamics using information flow, Probab.
[42] S.R. Kolla, S.D. Altman, Artificial neural network based fault identification Eng. Mech. 21 (4) (2006) 420–433.
scheme implementation for a three-phase induction motor, ISA Trans. 46 (2) [52] J.F. Alonso, M.A. Mañanas, D. Hoyer, Z.L. Topor, E.N. Bruce, Evaluation of
(2007) 261–266. respiratory muscles activity by means of cross mutual information function at
[43] P. Konar, P. Chattopadhyay, Bearing fault detection of induction motor using different levels of ventilatory effort, IEEE Trans. Biomed. Eng. 54 (9) (2007)
wavelet and support vector machines (SVMs), Appl. Soft Comput. 11 (6) 1573–1582.
(2011) 4203–4211. [53] U. Melia, M. Guaita, M. Vallverdú, C. Embid, I. Vilaseca, M. Salamero, J.
[44] F. Zidani, M.E.H. Benbouzid, D. Diallo, M.S. Nait-Said, Induction motor stator Santamaria, Mutual information measures applied to EEG signals for
faults diagnosis by a current Concordia pattern-based fuzzy decision system, sleepiness characterization, Med. Eng. Phys. 37 (3) (2015) 297–308.
IEEE Trans. Energy Convers. 18 (4) (2003) 469–475. [54] S. Haykin, Neural Networks and Learning Machines, 3rd ed., Pearson, New
[45] P.V.J. Rodríguez, A. Arkkio, Detection of stator winding fault in induction Jersey, 2008.
motor using fuzzy logic, Appl. Soft Comput. 8 (2) (2008) 1112–1120. [55] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann, I.H. Witten, The
[46] V.T. Tran, B.-S. Yang, M.-S. Oh, A.C.C. Tan, Fault diagnosis of induction motor WEKA data mining software: an update, ACM SIGKDD Explor. Newsl. 11 (1)
based on decision trees and adaptive neuro-fuzzy inference, Expert Syst. Appl. (2009) 10–18.
36 (2) (2009) 1840–1849.