Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.

v1

Article
Wavelet Long Short-Term Memory to Fault Forecasting in
Electrical Power Grids
Nathielle Waldrigues Branco 1 *, Mariana Santos Matos Cavalca 1 , Stefano Frizzo Stefenon 2,3 , and
Valderi Reis Quietinho Leithardt 4,5

1 Department of Electrical Engineering, Santa Catarina State University, R. Paulo Malschitzki 200, 89219-710
Joinville, Brazil; (email: [email protected])
2 Fondazione Bruno Kessler. Via Sommarive 18, 38123 Trento, Italy; (email: [email protected])
3 Department of Mathematics, Informatics and Physical Sciences, University of Udine. Via delle Scienze 206,
33100 Udine, Italy
4 COPELABS, Lusófona University of Humanities and Technologies. Campo Grande 376, 1749-024 Lisboa,
Portugal; (email: [email protected])
5 VALORIZA, Research Center for Endogenous Resources Valorization. Instituto Politécnico de Portalegre,
7300-555 Portalegre, Portugal
* Correspondence: [email protected]

Abstract: The electric power distribution utility is responsible for providing energy to consumers in
a continuous and stable way, failures in the electrical power system reduce the reliability indexes of
the grid, directly harming its performance. For this reason, there is a need for failure prediction to
reestablish power in the shortest possible time. Considering an evaluation of the number of failures
over time, this paper proposes to perform a failure prediction during the first year of the pandemic in
Brazil (2020) to verify the feasibility of using time series forecasting models for fault prediction. The
Long Short-Term Memory (LSTM) model will be evaluated to obtain a forecast result that can be used
by the electric power utility to organize the maintenance teams. The Wavelet transform shows to be
promising in improving the predictive ability of the LSTM, making the Wavelet LSTM model suitable
for the study at hand. The results show that the proposed approach has better results regarding the
evaluation of the error in prediction and has robustness when a statistical analysis is performed.

Keywords: Electrical Power Grids, Fault Forecasting, Long Short-Term Memory, Time Series Fore-
casting, Wavelet Transform.

1. Introduction
For the electricity reach the consumers in a stable and continuous way, the electrical
power grid must be working independently of the weather conditions [1]. To keep the
Citation: Wavelet Long Short-Term
electrical distribution system running, it is necessary to evaluate the performance of the
Memory to Fault Forecasting in
electrical system’s equipment through simulation, then disturbance conditions present
Electrical Power Grids. Preprints 2022,
in the electrical power grid can be identified [2]. Disturbances that occur in the electrical
1, 0. https://doi.org/
power system can significantly affect the power supply, variations in voltage level, increased
Publisher’s Note: MDPI stays neutral
surface conductivity, or contact of conductors with the ground can result in faults, which
with regard to jurisdictional claims in
affect power quality [3].
published maps and institutional affil-
Time series forecasting can be used to identify the possibility of a failure occurring,
iations.
which is a promising way to assist the decision-making process for maintenance teams
in an electric power utility [4]. As the increase in failures has a strong relationship with
weather conditions, in rainy seasons there is a greater chance of a failure occurring, so the
Copyright:
c 2022 by the authors.
study of this variation in relation to a time series is an important aspect in this context [5].
Licensee MDPI, Basel, Switzerland. The use of Wavelet transform for noise reduction is an approach that is effective when
This article is an open access article there is high nonlinearity in the time series [6], using high-frequency bandwidth filters
distributed under the terms and there may be a loss of information considering that a high frequency might be related to the
conditions of the Creative Commons occurrence of a failure. Considering that the Wavelet transform evaluates the signal energy,
Attribution (CC BY) license (https:// high frequencies are not totally eliminated, thus maintaining the main signal characteristics
creativecommons.org/licenses/by/ [7]. Thus a hybrid approach that combines a deep learning model with Wavelet transform
4.0/). can be an interesting approach [8].

© 2022 by the author(s). Distributed under a Creative Commons CC BY license.

Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

2 of 13

Long Short-Term Memory (LSTM) is a model applied in deep learning that has been
widely used by researchers for time series forecasting [9–11], its units solve the vanishing
gradient problem partially since the LSTM units allow the gradients to flow unchanged
[12]. Based on the advantages of the Wavelet transform and the promising capabilities of
LSTM [13–16], this work proposes to use a combination of those techniques in a method
named Wavelet LSTM. For this purpose, a study will be conducted using the alarm data
obtained from a re-closer of a power utility company in the Serrana region of Santa Catarina,
Brazil.
The main contributions of this research are:
• Proposal of a hybrid Wavelet LSTM model, which has higher predictive capacity than
the standard LSTM model. The Wavelet LSTM shows to be a more stable model for
time series prediction that can be used in several applications.
• Evaluate a time series regarding the variation of the number of failures in distribution
networks with bare cables, due to the presence of contamination and contact of foreign
materials with the grid, resulting in disruptive discharges in the power grid.
• Presentation of a solution for evaluating the failure history based on a time series that
can be used in other works in which it is necessary to evaluate the number of failures
over time.
The continuation of this paper is organized as follows: Section 2 presents a review of
related works and the used data. Section 3 presents the proposed method. In Section 4 the
results are analyzed. Section 5 presents a conclusion and a discussion of possible future
works.

2. Related Work
In electrical distribution systems, an electrical fault is defined as an anomaly in a
particular equipment causing a forced interruption in the operation of the electrical power
grid [17]. There are two classes of faults: transient and permanent. Transient faults are
anomalies of short duration that disappear soon after the action of protective devices,
having as common causes atmospheric discharges, momentary contacts between conduc-
tors and ground, opening of an electric arc and materials without adequate insulation.
Permanent faults are faults that continue to exist until it is possible to replace the defective
component or equipment [18].
Through fault diagnosis it is possible to detect where the fault occurred, its size,
duration, and impact on the electrical power system [19]. Among the most current fault
diagnosis methods are Bayesian Networks [20], Fuzzy Logic [21], Kalman Filter [22], and
other mathematical models based on artificial intelligence. The use of artificial intelligence
techniques for fault identification has been growing over the years, becoming nowadays a
hot topic, especially for the electric power system [23]. Deep learning models have being
increasingly used to improve the ability to identify faults in the electrical grid [24–26].
However, as these models have a large number of layers, they require more computational
effort, making the choice of the appropriate model a challenge [27]. From the image
processing of failed components, it is possible to identify patterns and thus improve their
identification in the field [28]. Several researchers are using object detection and image
classification based on convolutional neural networks (CNNs) models [29–31]. The CNNs
can be specially applied to improve the ability to identify faulty components, as shown
by Liu et al. [32] and Sadykova et al. [33] using You Only Look Once (YOLO), Li et al. [34]
with an improved Faster R-CNN, and Wen et al. [35] using Exact R-CNN. As presented by
Sadykova et. [33] the YOLO model is a promising alternative to identify insulators during
power grid inspections, being able to handle large datasets, where data augmentation
techniques can also be applied to avoid early overfitting. In this context, CNN super-
resolution can perform the reconstruction of the blurred images to perform the expansion
of the dataset [36].
The YOLO model has been updated and variations in its structure can result in
significant performance improvements. According to Liu et al. [32] the YOLOv3-dense
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

3 of 13

model proposed by them reaches up to 94.47% for insulator identification using varied
image backgrounds, in comparison for the same dataset YOLOv3 reaches 90.31% accuracy.
Previous versions such as YOLOv2 reach a maximum of 83.43%, making it clear that in
many situations the use of non-standard models can be a promising alternative.
Variations of the YOLO model have proven to be very efficient for locating insulators
on transmission lines. According to Liu et al. [37], MTI-YOLO has a higher average
precision than YOLO-tiny and YOLO-v2. Liu et al. [38] proposed an improved YOLOv3
model that is better than YOLOv3 and YOLOv3-dense models. Hu and Zhou [39] show
that YOLOv4 can reach an accuracy of 96.2% for insulator defect detection, also using
YOLOv4 Xing and Chen [40] had a precision of 97.78% for insulator identification.
When the distribution power system does not have insulation on the medium voltage
conductors, trees might touch the conductors resulting in discharges to the ground [41].
This type of fault is common in rural power grids that are close to wooded areas. To
prevent these faults, the electric power utility performs pruning of trees that are close to
the network, thus reducing the chance of discharges to the ground [42].
Insulators installed outdoors are exposed to environmental variations, such as dust
accumulation on their surface [43]. When contamination accumulates on insulators their
surface conductivity increases, generating leakage current until a discharge occurs [44],
when there is high humidity in the air the conductivity increases even more, consequently
increasing the chance of faults in the grid [45]. One type of contamination that has a
significant impact on the conductivity of insulators is salt contamination, which can be
measured by the equivalent salt deposit density [46].
Considering all these kind of possible faults [47], time series forecasting comes as
an alternative to prepare maintenance teams in advance to an event based the historical
knowledge of data variation over time [48]. A forecast with many steps ahead is challenging
as each step ahead contains the accumulated forecast error of the previous step [49], so
time series forecasting needs to take into account how many steps ahead can be considered
to obtain acceptable assertiveness [50].
Among algorithms for time series forecasting, ensemble learning models in general
have high performance and lower computational effort [51], and may be promising ap-
proaches for failure prediction. Various ways of combining the weak learners can be used
to create a model that has greater capacity, such as bagging, boosting, and stacking [52].
Further optimized models, such as the bayesian optimization-based dynamic ensemble
proposed by Du et al. [53] can be used, and are even applied with nonlinear data [54].
Many variations of ensemble models for time series forecasting can be found such
as efficient bootstrap stacking presented by Ribeiro et al. [55], extreme gradient boost-
ing proposed by Sauer et al. [56]. Especially for power system failure prediction, the
Wavelet transform combined with ensemble models becomes a superior approach to well-
established models such as the adaptive neuro-fuzzy inference system [57]. Therefore,
ensemble models are successful approaches for multi-step forward prediction [58], which
is equivalent to what is being evaluated in this paper.
Due to the existing features in the structure of the LSTM, it is one of the best qualified
model to handle chaotic time series, since it has the ability to remember distant values
and interpret order of dependencies, which are essential characteristics for a prediction
models. Abbasimehr and Paki [59] used the attention mechanism to have a enhanced LSTM
model. Related to the power system using LSTM, Guo et al. [60] and Ko et al. [61] presented
a research about wind power forecasting. Specially for fault prediction, Guo et al. [62]
proposed an modified LSTM version to improve safe and reliable operation of mechanical
equipments.

2.1. Faults
The vast majority of faults that occur in the electrical power distribution system with
naked cable are caused by direct contact with the network, this occurs mainly when the
weather conditions are bad (rain and intense wind), increasing the likelihood of the contact
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

4 of 13

of trees with the power grid. Another failure that can occur frequently is when insulators
lose their insulating capacity due to contamination or when the insulators are damaged.
To perform the time series evaluation all failures that occurred on the same day are
added up, to obtain a daily failure rate over time and thus evaluate the influence of the
change of season in relation to the increase of failures in the electrical power grid. These
failures are evaluated in relation to the alarms registered by the electric power utility
company during the evaluated period, some examples of alarms are presented in Table 1.

Table 1. Example of alarms that have been registered in the considered period.

Day Time Failure Record

06/01/2020 11:09:41 Current Phase B
06/01/2020 11:09:50 Current Phase A
06/01/2020 17:10:32 Current Phase C
26/01/2020 13:57:37 Recloser Communic. Failure
06/04/2020 10:04:23 Relay 50/51 (Neutral)
01/06/2020 17:24:51 Current Phase A
30/06/2020 14:11:56 Phase Voltage C
27/08/2020 10:00:54 Neutral Protection
27/08/2020 11:58:48 Current Phase C
11/09/2020 03:06:12 Current Phase A
29/12/2020 13:56:32 Relay 50/51 (Phase A)

Since failures generally occur in a non-linear pattern, this evaluation is based on a

statistical analysis, and it is not possible to determine exactly when a failure will occur,
however it is possible to evaluate in which period of the year there is a greater chance of
the highest number of failures occurring.
In this paper, the evaluation of the history of recorded faults is in relation to the
year 2020 (from January 1 to December 31), this history corresponds to the sum of all the
faults of the distribution branches in the Lages region. In total there are 366 days recorded
considering that the year 2020 was a leap year, Figure 1 presents the sum of the alarms
regarding faults per day in this period.

Figure 1. Failures registered in the power grid in 2020 (Lages region), data provided by Centrais
Elétricas de Santa Catarina (CELESC).

3. Wavelet LSTM
The Wavelet LSTM method is a combination of the Wavelet transform and the Long
Short-Term Memory. This approach has been widely used for fault diagnosis, as presented
in the work of Sabir et al. [63] and Jalayer, Orsenigo, and Vercellis [64] for electrical machines,
and specially for rolling bearing in the work of Tan et al. [65].
To apply the Wavelet LSTM here, initially the time series passes through the Wavelet
filter to reduce noise and non-linearities, after the signal is decomposed and reconstructed,
the LSTM receives the filtered signal and performs the prediction. The complete structure
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

5 of 13

of this approach is presented in Figure 2 and will be explained in this section. The values of
the original signal (shown in Figure 1) are transformed into an initial time series, that is
used as input to the model.

Figure 2. Structure of the Wavelet Long Short-Term Memory model.

To use the Wavelet transform first the signal is decomposed using the Wavelet packet
method. The decomposition may be denoted by:
Z +∞
t−b
 
1
WΨ,x ( a, b) = √ x (t)Ψ ∗ dt, a 6= 0 (1)
a −∞ a
where x (t) is the signal to be decomposed, Ψ is the time-based function (mother Wavelet), a
and b are the scale and the displacement parameters respectively [66]. Given a discretization
the high-pass filter g(n) is:
g(n) = h(2N − 1 − n). (2)
where h(n) is the low-pass filter. Thereby, the mother Wavelet and the scaling function (Φ)
are given by:
N −1
Ψ(n) = ∑ g(i )Φ(2n − i ), (3)
i =0

N −1
Φ(n) = ∑ h(i )Φ(2n − i ). (4)
i =0

After reconstructing the filtered signal, a time series is obtained that is used for LSTM
forecast evaluation.
LSTM is a recurrent neural network, that has feedback allowing the model to remember
distant values. For the time series forecasting starting from D samples,

x ( t − ( D − 1) ∆ ), . . . , x ( t − ∆ ), x ( t ) (5)

to predict future value,

x ( t + P ), (6)
where P are the steps forward, and ∆ is the period of the samples. In this paper, ∆ is equal
to one day, where all faults of the same day were summed.
The LSTM is capable of understanding order dependence in problems that require
sequence prediction, making it promising for time series forecasting [67]. In an LSTM
algorithm, each cell is divided into three gates, the input (it ), output (ot ) and forgetting
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

6 of 13

( f t ) gates [68]. The f t controls how much information will be forgotten and how much
will be remembered. The useful information for the states is added through the it , and ot
determines how much of the current state must be assigned to the output [69]. The LSTM
can be defined by the following equations:

it = σg (Wi xt + Ri ht−1 + bi ),
f t = σg (W f xt + R f ht−1 + b f ), (7)
ot = σg (Wo xt + Ro ht−1 + bo ).
in which b is the polarization matrix, R and W are earnings matrices, and σg is the activation
function [70]. The LSTM has input activation function G and output activation function H,
that are used to update the cell and the hidden state, as given in the equations:

c̃t = G (Wc xt + Rc ht−1 + bc ),

ct = f t ◦ ct−1 + it ◦ c̃, (8)
h t = o t ◦ H ( c t ).
To perform the predicted values of future time steps, the training responses sequences
are shifted by a time step. Thus, at each time-step in the input sequence, the net learns to
forecast the following time-step value.

3.1. Considered Measures

In this paper the root-mean-square error (RMSE), mean absolute error (MAE), and
coefficient of determination (R2 ) were considered, given by:
s
1 n
n i∑
RMSE = (yi − ŷi )2 , (9)
=1
n
1
MAE =
n ∑ |yi − ŷi |, (10)
i =1
2
∑in=1 (yi − ŷi )
R2 = 1 − 2
, (11)
∑in=1 (yi − ȳi )
where yi is the observed value, ŷi is the predicted output, and ȳi is the average of the
observed value [71]. The final statistical evaluation is performed with 50 runs using the
same parameters configuration, where the mean, median, and standard deviation were
evaluated. The simulations were computed using an Intel Core I5-7400, 20 GB of RAM,
with MATLAB software.

4. Analysis of Results
The first evaluation is regarding the analysis of the time series forecast performed in
relation to the percentage of data used for training and testing of the neural network. The
evaluation of this parameter is important because it can be used to define the minimum
amount of data needed for training the model. The evaluation results are shown in Table 2,
considering a training ratio from 50 to 90 percent. The test uses the difference equivalent
percentage to complete the data set, in which the validation stage is not considered.

Table 2. Evaluating the influence of the training/testing relationship.

Time
Train / Test RMSE MAE R2
(s)
50 / 50 8.02×10−3 3.03×10−3 0.1516 17.21
60 / 40 6.24×10−3 7.47×10−4 0.1681 18.53
70 / 30 4.73×10−3 3.59×10−4 0.0325 18.64
80 / 20 3.60×10−3 1.18×10−3 0.2779 20.47
90 / 10 4.29×10−3 5.49×10−5 0.0884 19.68
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

7 of 13

Using 80% of the data for training and 20% of the data for testing gives the best RMSE
and R2 values, hence this ratio was used in the further analysis. As can be seen in Figure 3,
there is a major difficulty in predicting the data due to the nonlinearities in the time series,
considering that in some cases there were several failures in a short period of time.

Figure 3. Preliminary analysis of fault prediction capability.

The failures that occurred after the middle of the year were due to the rainy season
that starts after winter in the southern hemisphere. The greater presence of bad weather
conditions favors the development of faults in the electrical power distribution system. In
the following analysis, the optimizer and the number of hidden units are evaluated (see
Table 3).

Table 3. Assessment of the number of hidden units (HU) using different optimizers.

H Time
Optimizer RMSE MAE R2
U (s)
50 3.49×10−3 9.12×10−4 0.1957 17.74
100 3.50×10−3 9.12×10−4 0.2035 18.29
SGDM 200 3.51×10−3 1.17×10−3 0.2089 19.58
500 3.44×10−3 9.18×10−4 0.1624 25.06
1000 3.44×10−3 9.37×10−4 0.1623 36.66
50 4.79×10−3 1.77×10−3 - 17.94
100 7.69×10−3 4.58×10−3 - 19.94
ADAM 200 3.98×10−3 6.51×10−4 0.5554 19.21
500 3.96×10−3 8.29×10−4 0.5469 25.55
1000 3.94×10−3 1.52×10−3 0.5242 35.22
50 6.22×10−3 3.01×10−3 - 21.58
100 5.65×10−3 1.22×10−3 - 19.14
RMSprop 200 3.93×10−3 6.64×10−4 0.5212 21.64
500 3.35×10−3 1.00×10−3 0.1074 27.51
1000 3.16×10−3 2.96×10−5 0.0190 36.50

In this evaluation, the SGDM optimizer had more stable results for the coefficient
of determination, presenting a smaller variance in relation to the change of hidden units.
Comparing all the models the best results occurred using 200 hidden units considering the
coefficient of determination. In some cases, it was not possible to measure the coefficient
of determination due to the high intensity of variation in the prediction using the ADAM
and RMSprop optimizers. Considering that there was a large variation in the values, a
statistical analysis was performed and will be presented using 200 hidden units.
The inclusion of the Wavelet transform for noise reduction is added for the following
analysis, this transform should be used with caution as it can result in a loss of features of
the signal. Figure 4 shows the result of the Wavelet transform relative to the original signal
using 1 node, and Figure 5 shows this comparison using 2 nodes.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

8 of 13

Figure 4. Evaluation of the Wavelet transform with 1 node.

The use of 2 nodes considerably alters the response of the transform, hindering a
practical application. When 3 nodes or more are used, the signal loses its characteristics
and is not considered in this paper.

Figure 5. Evaluation of the Wavelet transform with 2 nodes.

The complete analysis of the Wavelet transform depth variation is presented in Table
3. Considering that because there is an error value that makes the prediction not suitable
for analysis, the use of 2 nodes is disregarded after this evaluation, as mentioned earlier
this can also be observed when the Wavelet transform was compared in this configuration
in relation to the original signal (see Figure 5).

Table 4. Assessment of depth using different 1 and 2 nodes.

Time
Nodes Depth RMSE MAE R2
(s)
1 2.19×10−3 4.54×10−4 0.3375 23.14
2 2.16×10−3 4.69×10−4 0.3547 22.13
1 3 2.22×10−3 5.28×10−4 0.3132 23.61
4 2.16×10−3 3.53×10−4 0.3509 22.76
5 2.17×10−3 3.70×10−4 0.3493 21.71
1 2.22×1010 1.24×109 0.6888 21.34
2 2.13×1010 1.24×109 0.7155 22.75
2 3 2.45×1010 6.79×109 0.6214 18.05
4 2.32×1010 1.90×109 0.6601 18.74
5 2.44×1010 6.58×109 0.6244 18.69

The best coefficient of determination was reached using a depth equal to 2 in the
Wavelet transform, getting close to the best MAE value that happened using the depth
equal to 4. Considering these results, the depth equal to 2 was used for statistical analysis
which is presented in Table 5.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

9 of 13

Table 5. Statistical evaluation.

Model Optimizer Mean Median Std Dev.

SGDM 3.50×10−3 3.49×10−3 2.62×10−5
LSTM ADAM 7.87×10−3 8.04×10−3 2.63×10−3
RMSprop 4.90×10−3 4.95×10−3 6.41×10−4
SGDM 2.19×10−3 2.19×10−3 2.76×10−5
Wavelet
ADAM 1.79×10−3 1.27×10−3 2.27×10−3
LSTM
RMSprop 1.67×10−3 1.54×10−3 7.15×10−4

The Wavelet LSTM model was superior in all comparative analyses to the LSTM model
with respect to RMSE. Even varying the optimizer, the Wavelet LSTM showed promise for
the analysis in question. The best average RMSE result was obtained using the RMSprop
optimizer in the Wavelet LSTM model. The comparison between the prediction result and
the original signal is presented in Figure 6.

Figure 6. Comparison of the prediction using the Wavelet LSTM model to the observed values.

The prediction results using the Wavelet LSTM in comparison to the original signal
showed that this model is suitable for the application in question. As can be seen, after
the highest accumulated value of the number of failures there was an oscillation in the
prediction, something that is expected due to this abrupt variation in the time series.

5. Conclusion
The prediction of faults in the electrical distribution system is necessary to ensure
the operation of the power grid. Analyzing the variation of a time series it is possible to
verify the presence of a higher number of failures during a certain time and thus define
a more effective correction strategy. Based on a time series forecasting, the electric utility
may know when there are higher chances to have faults before it happened, and so have a
better define a strategy to deal with that.
It is noticeable that there is a difficulty in this prediction due to the large variation
in the number of failures in some seasons of the year, mainly related to the rainy season.
Using traditional models the forecast results are ineffective, so it is necessary to combine
algorithms to create a hybrid model to meet the needs of the problem.
The Wavelet LSTM model showed better results in all analyses compared to the stan-
dard LSTM model, including better results in the statistical analysis, being an appropriate
model for the evaluation presented in this paper. Using this model it is possible to have
failure prediction indicators that can help the organization of maintenance teams, thus
reducing the response time when a disruptive failure occurs.
Future work can be done regarding the type of failure, the failures can vary, for
instance, as a result of direct contact with the grid and/or leakage current. Specific analysis
on which type of failure that occurs more frequently and how to avoid this type of failure
is promising work to be done in the future.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

10 of 13

Author Contributions: Writing—original draft, methodology, Nathielle Waldrigues Branco; Writing—

review and editing, supervision, Mariana Santos Matos Cavalca; Writing—review and editing, soft-
ware, formal analysis, Stefano Frizzo Stefenon; Project administration, supervision, Valderi Reis
Quietinho Leithardt.
Funding: This work was supported by the National Funds through the Fundação para a Ciên-
cia e a Tecnologia, I.P. (Portuguese Foundation for Science and Technology) by the Project “VAL-
ORIZA—Research Centre for Endogenous Resource Valorization” under Grant UIDB/05064/2020
and Grant UIDB/04111/2020, and in part by the Instituto Lusófono de Investigação e Desenvolvi-
mento (ILIND) under Project COFAC/ILIND/COPELABS/3/2020.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data used in this paper were provided by Centrais Elétricas de
Santa Catarina, regarding the alarms of the power distribution grids in Lages region, Brazil, from
January 1 to December 31 of 2020. These records are available at: https://github.com/SFStefenon/
FailuresPowerGrid2020 (Accessed on September 21, 2021).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Araya, J.; Montaña, J.; Schurch, R. Electric Field Distribution and Leakage Currents in Glass Insulator Under Differ-
ent Altitudes and Pollutions Conditions using FEM Simulations. IEEE Latin America Transactions 2021, 19, 1278–1285.
doi:10.1109/TLA.2021.9475858.
2. Sun, J.; Yang, Q.; Cui, H.; Ran, J.; Liu, H. Distribution Line Fault Location With Unknown Fault Impedance Based on Electromag-
netic Time Reversal. IEEE Transactions on Electromagnetic Compatibility 2021, 63, 1921–1929. doi:10.1109/TEMC.2021.3097105.
3. Liu, Z.; Chen, H.; Hu, Z.; Li, Y.; Wu, X.; Peng, H. Fault Detection System for 500 kV AC Fault Current Limiter Based on
High-Coupled Split Reactor. IEEE Transactions on Applied Superconductivity 2021, 31, 1–7. doi:10.1109/TASC.2021.3101745.
4. Stefenon, S.F.; Bruns, R.; Sartori, A.; Meyer, L.H.; Ovejero, R.G.; Leithardt, V.R.Q. Analysis of the Ultrasonic Sig-
nal in Polymeric Contaminated Insulators Through Ensemble Learning Methods. IEEE Access 2022, 10, 33980–33991.
doi:10.1109/ACCESS.2022.3161506.
5. Medeiros, A.; Sartori, A.; Stefenon, S.F.; Meyer, L.H.; Nied, A. Comparison of artificial intelligence techniques to failure
prediction in contaminated insulators based on leakage current. Journal of Intelligent & Fuzzy Systems 2021, 42, 3285–3298.
doi:10.3233/JIFS-211126.
6. Rhif, M.; Ben Abbes, A.; Farah, I.R.; Martínez, B.; Sang, Y. Wavelet Transform Application for/in Non-Stationary Time-Series
Analysis: A Review. Applied Sciences 2019, 9, 1345. doi:10.3390/app9071345.
7. Ameid, T.; Menacer, A.; Talhaoui, H.; Azzoug, Y. Discrete wavelet transform and energy eigen value for rotor bars fault detection in
variable speed field-oriented control of induction motor drive. ISA Transactions 2018, 79, 217–231. doi:10.1016/j.isatra.2018.04.019.
8. Stefenon, S.F.; Kasburg, C.; Nied, A.; Klaar, A.C.R.; Ferreira, F.C.S.; Branco, N.W. Hybrid deep learning for power generation
forecasting in active solar trackers. IET Generation, Transmission & Distribution 2020, 14, 5667–5674. doi:10.1049/iet-gtd.2020.0814.
9. Chandra, R.; Goyal, S.; Gupta, R. Evaluation of Deep Learning Models for Multi-Step Ahead Time Series Prediction. IEEE Access
2021, 9, 83105–83123. doi:10.1109/ACCESS.2021.3085085.
10. Hu, Y.; Sun, X.; Nie, X.; Li, Y.; Liu, L. An Enhanced LSTM for Trend Following of Time Series. IEEE Access 2019, 7, 34020–34030.
doi:10.1109/ACCESS.2019.2896621.
11. Ma, C.; Dai, G.; Zhou, J. Short-Term Traffic Flow Prediction for Urban Road Sections Based on Time Series Analysis and LSTM
BILSTM Method. IEEE Transactions on Intelligent Transportation Systems 2022, 23, 5615–5624. doi:10.1109/TITS.2021.3055258.
12. Zhang, S.; Wang, Y.; Liu, M.; Bao, Z. Data-Based Line Trip Fault Prediction in Power Systems Using LSTM Networks and SVM.
IEEE Access 2018, 6, 7675–7686. doi:10.1109/ACCESS.2017.2785763.
13. Kim, W.H.; Kim, J.Y.; Chae, W.K.; Kim, G.; Lee, C.K. LSTM-Based Fault Direction Estimation and Protection Coordination for
Networked Distribution System. IEEE Access 2022, 10, 40348–40357. doi:10.1109/ACCESS.2022.3166836.
14. Qiao, M.; Yan, S.; Tang, X.; Xu, C. Deep Convolutional and LSTM Recurrent Neural Networks for Rolling Bearing Fault Diagnosis
Under Strong Noises and Variable Loads. IEEE Access 2020, 8, 66257–66269. doi:10.1109/ACCESS.2020.2985617.
15. Stefenon, S.F.; Freire, R.Z.; Meyer, L.H.; Corso, M.P.; Sartori, A.; Nied, A.; Klaar, A.C.R.; Yow, K.C. Fault detection in insulators
based on ultrasonic signal processing using a hybrid deep learning technique. IET Science, Measurement & Technology 2020,
14, 953–961. doi:10.1049/iet-smt.2020.0083.
16. Ma, Y.; Oslebo, D.; Maqsood, A.; Corzine, K. DC Fault Detection and Pulsed Load Monitoring Using Wavelet Transform-Fed LSTM
Autoencoders. IEEE Journal of Emerging and Selected Topics in Power Electronics 2021, 9, 7078–7087. doi:10.1109/JESTPE.2020.3019382.
17. Furse, C.M.; Kafal, M.; Razzaghi, R.; Shin, Y.J. Fault Diagnosis for Electrical Systems and Power Networks: A Review. IEEE
Sensors Journal 2021, 21, 888–906. doi:10.1109/JSEN.2020.2987321.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

11 of 13

18. Li, B.; Cui, H.; Li, B.; Wen, W.; Dai, D. A permanent fault identification method for single-pole grounding fault of overhead
transmission lines in VSC-HVDC grid based on fault line voltage. International Journal of Electrical Power & Energy Systems 2020,
117, 105603. doi:10.1016/j.ijepes.2019.105603.
19. Wadi, M.; Elmasry, W. An Anomaly-based Technique for Fault Detection in Power System Networks. In Proceedings of the 2021
International Conference on Electric Power Engineering; ICEPE: Palestine, 2021; pp. 1–6. doi:10.1109/ICEPE-P51568.2021.9423479.
20. Wu, J.; Zhang, L.; Bai, Y.; Reniers, G. A safety investment optimization model for power grid enterprises based on System
Dynamics and Bayesian network theory. Reliability Engineering & System Safety 2022, 221, 108331. doi:10.1016/j.ress.2022.108331.
21. Sadi, M.A.H.; AbuHussein, A.; Shoeb, M.A. Transient Performance Improvement of Power Systems Using Fuzzy
Logic Controlled Capacitive-Bridge Type Fault Current Limiter. IEEE Transactions on Power Systems 2021, 36, 323–335.
doi:10.1109/TPWRS.2020.3003294.
22. Rigatos, G.; Serpanos, D.; Zervos, N. Detection of Attacks Against Power Grid Sensors Using Kalman Filter and Statistical
Decision Making. IEEE Sensors Journal 2017, 17, 7641–7648. doi:10.1109/JSEN.2017.2661247.
23. Sopelsa Neto, N.F.; Stefenon, S.F.; Meyer, L.H.; Bruns, R.; Nied, A.; Seman, L.O.; Gonzalez, G.V.; Leithardt, V.R.Q.; Yow, K.C. A
Study of Multilayer Perceptron Networks Applied to Classification of Ceramic Insulators Using Ultrasound. Applied Sciences
2021, 11, 1592. doi:10.3390/app11041592.
24. Haj, Y.E.; El-Hag, A.H.; Ghunem, R.A. Application of Deep-Learning via Transfer Learning to Evaluate Silicone Rubber Material
Surface Erosion. IEEE Transactions on Dielectrics and Electrical Insulation 2021, 28, 1465–1467. doi:10.1109/TDEI.2021.009617.
25. Stefenon, S.F.; Singh, G.; Yow, K.C.; Cimatti, A. Semi-ProtoPNet Deep Neural Network for the Classification of Defective Power
Grid Distribution Structures. Sensors 2022, 22, 4859. doi:10.3390/s22134859.
26. Zhao, M.; Barati, M. A Real-Time Fault Localization in Power Distribution Grid for Wildfire Detection Through Deep Convolu-
tional Neural Networks. IEEE Transactions on Industry Applications 2021, 57, 4316–4326. doi:10.1109/TIA.2021.3083645.
27. Mantach, S.; Lutfi, A.; Moradi Tavasani, H.; Ashraf, A.; El-Hag, A.; Kordi, B. Deep Learning in High Voltage Engineering: A
Literature Review. Energies 2022, 15, 5005. doi:10.3390/en15145005.
28. Corso, M.P.; Perez, F.L.; Stefenon, S.F.; Yow, K.C.; García Ovejero, R.; Leithardt, V.R.Q. Classification of Contaminated Insulators
Using k-Nearest Neighbors Based on Computer Vision. Computers 2021, 10, 112. doi:10.3390/computers10090112.
29. Wu, H.; Hu, Y.; Wang, W.; Mei, X.; Xian, J. Ship Fire Detection Based on an Improved YOLO Algorithm with a Lightweight
Convolutional Neural Network Model. Sensors 2022, 22, 7420. doi:10.3390/s22197420.
30. Vieira, J.C.; Sartori, A.; Stefenon, S.F.; Perez, F.L.; de Jesus, G.S.; Leithardt, V.R.Q. Low-Cost CNN for Automatic Violence
Recognition on Embedded System. IEEE Access 2022, 10, 25190–25202. doi:10.1109/ACCESS.2022.3155123.
31. Hou, L.; Chen, C.; Wang, S.; Wu, Y.; Chen, X. Multi-Object Detection Method in Construction Machinery Swarm Operations
Based on the Improved YOLOv4 Model. Sensors 2022, 22, 7294. doi:10.3390/s22197294.
32. Liu, C.; Wu, Y.; Liu, J.; Sun, Z. Improved YOLOv3 Network for Insulator Detection in Aerial Images with Diverse Background
Interference. Electronics 2021, 10, 771. doi:10.3390/electronics10070771.
33. Sadykova, D.; Pernebayeva, D.; Bagheri, M.; James, A. IN-YOLO: Real-Time Detection of Outdoor High Voltage Insulators Using
UAV Imaging. IEEE Transactions on Power Delivery 2020, 35, 1599–1601. doi:10.1109/TPWRD.2019.2944741.
34. Li, X.; Su, H.; Liu, G. Insulator Defect Recognition Based on Global Detection and Local Segmentation. IEEE Access 2020,
8, 59934–59946. doi:10.1109/ACCESS.2020.2982288.
35. Wen, Q.; Luo, Z.; Chen, R.; Yang, Y.; Li, G. Deep Learning Approaches on Defect Detection in High Resolution Aerial Images of
Insulators. Sensors 2021, 21, 1033. doi:10.3390/s21041033.
36. Chen, H.; He, Z.; Shi, B.; Zhong, T. Research on Recognition Method of Electrical Components Based on YOLO V3. IEEE Access
2019, 7, 157818–157829. doi:10.1109/ACCESS.2019.2950053.
37. Liu, C.; Wu, Y.; Liu, J.; Han, J. MTI-YOLO: A Light-Weight and Real-Time Deep Neural Network for Insulator Detection in
Complex Aerial Images. Energies 2021, 14, 1426. doi:10.3390/en14051426.
38. Liu, J.; Liu, C.; Wu, Y.; Xu, H.; Sun, Z. An Improved Method Based on Deep Learning for Insulator Fault Detection in Diverse
Aerial Images. Energies 2021, 14, 4365. doi:10.3390/en14144365.
39. Hu, X.; Zhou, Y. Insulator defect detection in power inspection image using focal loss based on YOLO v4. In Proceedings of the
International Conference on Artificial Intelligence, Virtual Reality, and Visualization (AIVRV 2021). SPIE, 2021, Vol. 12153, pp. 90
– 95. doi:10.1117/12.2626694.
40. Xing, Z.; Chen, X. Lightweight algorithm of insulator identification applicable to electric power engineering. Energy Reports 2022,
8, 353–362. doi:10.1016/j.egyr.2022.01.209.
41. Stefenon, S.F.; Seman, L.O.; Pavan, B.A.; Ovejero, R.G.; Leithardt, V.R.Q. Optimal design of electrical power distribution grid
spacers using finite element method. IET Generation, Transmission & Distribution 2022, 16, 1865–1876. doi:10.1049/gtd2.12425.
42. Yang, C.; Chen, T.; Yang, B.; Zhang, X.; Fan, S. Experimental study of tree ground fault discharge characteristics of 35 kV
transmission lines. In Proceedings of the 2021 IEEE Sustainable Power and Energy Conference (iSPEC); , 2021; Vol. 1, pp.
2883–2891. doi:10.1109/iSPEC53008.2021.9735502.
43. Stefenon, S.F.; Neto, C.S.F.; Coelho, T.S.; Nied, A.; Yamaguchi, C.K.; Yow, K.C. Particle swarm optimization for design of insulators
of distribution power system based on finite element method. Electrical Engineering 2022, 104, 615–622. doi:10.1007/s00202-021-
01332-3.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

12 of 13

44. Salem, A.A.; Abd-Rahman, R.; Al-Gailani, S.A.; Kamarudin, M.S.; Ahmad, H.; Salam, Z. The Leakage Current Compo-
nents as a Diagnostic Tool to Estimate Contamination Level on High Voltage Insulators. IEEE Access 2020, 8, 92514–92528.
doi:10.1109/ACCESS.2020.2993630.
45. Stefenon, S.F.; Seman, L.O.; Sopelsa Neto, N.F.; Meyer, L.H.; Nied, A.; Yow, K.C. Echo state network applied for clas-
sification of medium voltage insulators. International Journal of Electrical Power & Energy Systems 2022, 134, 107336.
doi:10.1016/j.ijepes.2021.107336.
46. Cao, B.; Wang, L.; Yin, F. A Low-Cost Evaluation and Correction Method for the Soluble Salt Components of the Insulator
Contamination Layer. IEEE Sensors Journal 2019, 19, 5266–5273. doi:10.1109/JSEN.2019.2902192.
47. Stefenon, S.F.; Corso, M.P.; Nied, A.; Perez, F.L.; Yow, K.C.; Gonzalez, G.V.; Leithardt, V.R.Q. Classification of insulators using
neural network based on computer vision. IET Generation, Transmission & Distribution 2021, 16, 1096–1107. doi:10.1049/gtd2.12353.
48. Hou, C.; Wu, J.; Cao, B.; Fan, J. A deep-learning prediction model for imbalanced time series data forecasting. Big Data Mining
and Analytics 2021, 4, 266–278. doi:10.26599/BDMA.2021.9020011.
49. Taieb, S.B.; Atiya, A.F. A Bias and Variance Analysis for Multistep-Ahead Time Series Forecasting. IEEE Transactions on Neural
Networks and Learning Systems 2016, 27, 62–76. doi:10.1109/TNNLS.2015.2411629.
50. Duan, J.; Kashima, H. Learning to Rank for Multi-Step Ahead Time-Series Forecasting. IEEE Access 2021, 9, 49372–49386.
doi:10.1109/ACCESS.2021.3068895.
51. Stefenon, S.F.; Ribeiro, M.H.D.M.; Nied, A.; Yow, K.C.; Mariani, V.C.; Coelho, L.S.; Seman, L.O. Time series forecasting using
ensemble learning methods for emergency prevention in hydroelectric power plants with dam. Electric Power Systems Research
2022, 202, 107584. doi:10.1016/j.epsr.2021.107584.
52. Ribeiro, M.H.D.M.; Coelho, L.S. Ensemble approach based on bagging, boosting and stacking for short-term prediction in
agribusiness time series. Applied Soft Computing 2020, 86, 105837. doi:10.1016/j.asoc.2019.105837.
53. Du, L.; Gao, R.; Suganthan, P.N.; Wang, D.Z. Bayesian optimization based dynamic ensemble for time series forecasting.
Information Sciences 2022, 591, 155–175. doi:10.1016/j.ins.2022.01.010.
54. Kim, D.; Kim, C. Forecasting time series with genetic fuzzy predictor ensemble. IEEE Transactions on Fuzzy Systems 1997,
5, 523–535. doi:10.1109/91.649903.
55. Ribeiro, M.H.D.M.; da Silva, R.G.; Moreno, S.R.; Mariani, V.C.; Coelho, L.S. Efficient bootstrap stacking ensemble learning
model applied to wind power generation forecasting. International Journal of Electrical Power & Energy Systems 2022, 136, 107712.
doi:10.1016/j.ijepes.2021.107712.
56. Sauer, J.; Mariani, V.C.; Coelho, L.S.; Ribeiro, M.H.D.M.; Rampazzo, M. Extreme gradient boosting model based on improved
Jaya optimizer applied to forecasting energy consumption in residential buildings. Evolving Systems 2021, pp. 1–12.
57. Stefenon, S.F.; Ribeiro, M.H.D.M.; Nied, A.; Mariani, V.C.; Coelho, L.D.S.; Leithardt, V.R.Q.; Silva, L.A.; Seman, L.O.
Hybrid Wavelet Stacking Ensemble Model for Insulators Contamination Forecasting. IEEE Access 2021, 9, 66387–66397.
doi:10.1109/ACCESS.2021.3076410.
58. da Silva, R.G.; Ribeiro, M.H.D.M.; Moreno, S.R.; Mariani, V.C.; Coelho, L.S. A novel decomposition-ensemble learning framework
for multi-step ahead wind energy forecasting. Energy 2021, 216, 119174. doi:10.1016/j.energy.2020.119174.
59. Abbasimehr, H.; Paki, R. Improving time series forecasting using LSTM and attention models. Journal of Ambient Intelligence and
Humanized Computing 2022, 13, 673–691.
60. Yu, R.; Gao, J.; Yu, M.; Lu, W.; Xu, T.; Zhao, M.; Zhang, J.; Zhang, R.; Zhang, Z. LSTM-EFG for wind power forecasting based on
sequential correlation features. Future Generation Computer Systems 2019, 93, 33–42. doi:10.1016/j.future.2018.09.054.
61. Ko, M.S.; Lee, K.; Kim, J.K.; Hong, C.W.; Dong, Z.Y.; Hur, K. Deep Concatenated Residual Network With Bidirec-
tional LSTM for One-Hour-Ahead Wind Power Forecasting. IEEE Transactions on Sustainable Energy 2021, 12, 1321–1335.
doi:10.1109/TSTE.2020.3043884.
62. Guo, J.; Lao, Z.; Hou, M.; Li, C.; Zhang, S. Mechanical fault time series prediction by using EFMSAE-LSTM neural network.
Measurement 2021, 173, 108566. doi:10.1016/j.measurement.2020.108566.
63. Sabir, R.; Rosato, D.; Hartmann, S.; Guehmann, C. LSTM Based Bearing Fault Diagnosis of Electrical Machines using Motor
Current Signal. In Proceedings of the 2019 18th IEEE International Conference On Machine Learning And Applications. ICMLA,
2019, pp. 613–618. doi:10.1109/ICMLA.2019.00113.
64. Jalayer, M.; Orsenigo, C.; Vercellis, C. Fault detection and diagnosis for rotating machinery: A model based on convolutional LSTM,
Fast Fourier and continuous wavelet transforms. Computers in Industry 2021, 125, 103378. doi:10.1016/j.compind.2020.103378.
65. Tan, W.; Sun, Y.; Qiu, D.; An, Y.; Ren, P. Rolling Bearing Fault Diagnosis Based on Single Gated Unite Recurrent Neural Networks.
Journal of Physics: Conference Series 2020, 1601, 1–9. doi:10.1088/1742-6596/1601/4/042017.
66. Stefenon, S.F.; Kasburg, C.; Freire, R.Z.; Silva Ferreira, F.C.; Bertol, D.W.; Nied, A. Photovoltaic power forecasting using wavelet
Neuro-Fuzzy for active solar trackers. Journal of Intelligent & Fuzzy Systems 2021, 40, 1083–1096. doi:10.3233/JIFS-201279.
67. Casado-Vara, R.; Martin del Rey, A.; Pérez-Palau, D.; de-la Fuente-Valentín, L.; Corchado, J.M. Web Traffic Time Series Forecasting
Using LSTM Neural Networks with Distributed Asynchronous Training. Mathematics 2021, 9, 421. doi:10.3390/math9040421.
68. Fernandes, F.; Stefenon, S.F.; Seman, L.O.; Nied, A.; Ferreira, F.C.S.; Subtil, M.C.M.; Klaar, A.C.R.; Leithardt, V.R.Q. Long
short-term memory stacking model to predict the number of cases and deaths caused by COVID-19. Journal of Intelligent & Fuzzy
Systems 2022, 6, 6221–6234. doi:10.3233/JIFS-212788.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 3 October 2022 doi:10.20944/preprints202210.0004.v1

13 of 13

69. Sagheer, A.; Kotb, M. Time series forecasting of petroleum production using deep LSTM recurrent networks. Neurocomputing
2019, 323, 203–213. doi:10.1016/j.neucom.2018.09.082.
70. Liu, Y.; Guan, L.; Hou, C.; Han, H.; Liu, Z.; Sun, Y.; Zheng, M. Wind Power Short-Term Prediction Based on LSTM and Discrete
Wavelet Transform. Applied Sciences 2019, 9, 1108. doi:10.3390/app9061108.
71. Sopelsa Neto, N.F.; Stefenon, S.F.; Meyer, L.H.; Ovejero, R.G.; Leithardt, V.R.Q. Fault Prediction Based on Leakage Current in
Contaminated Insulators Using Enhanced Time Series Forecasting Models. Sensors 2022, 22, 6121. doi:10.3390/s22166121.