electronics

Article
Intelligent Vibration Monitoring System for Smart Industry
Utilizing Optical Fiber Sensor Combined with Machine Learning
Pradeep Kumar 1 , Guo-Liang Shih 1 , Cheng-Kai Yao 1 , Stotaw Talbachew Hayle 2 , Yibeltal Chanie Manie 1
and Peng-Chun Peng 1, *

1 Department of Electro-Optical Engineering, National Taipei University of Technology, Taipei 10608, Taiwan;
t111999408@ntut.org.tw (P.K.); t109650037@ntut.org.tw (G.-L.S.); t109658093@ntut.org.tw (C.-K.Y.);
yibeshmamaru@gmail.com (Y.C.M.)
2 Faculty of Natural and Computational Science, Debre Berhan University, Debre Berhan 445, Ethiopia;
stotaw19@gmail.com
* Correspondence: pcpeng@ntut.edu.tw; Tel.: +886-2-2771-2171 (ext. 4671)

Abstract: In this paper, we proposed and experimentally demonstrated the association of a fiber Bragg
Grating (FBG) sensing system with You Only Look Once V7 (YOLO V7) to identify the vibration
signal of a faulty machine. In the experiment, the YOLO V7 network architecture consists of a
backbone, three detection heads (Headx3), a path aggregation network (PAN), and a feature pyramid
network (FPN). The proposed architecture has an FBG sensor and the FBG interrogator employed
for collecting sensing vibration signals or vibration data when degradation or fault occurs. An FBG
interrogator collects vibration data independently, and then the YOLO V7 object detection algorithm
is the recognition architecture of the vibration pattern of the signal. Thus, the proposed vibration
recognition or detection is an assurance for detecting vibration signals that can support monitoring
the machine’s health. Moreover, this research is promising for ensuring a high accuracy detection
of faulty signals rate in industrial equipment monitoring and offers a robust system, resulting in
remarkable accuracy with an overall model accuracy of 99.7%. The result shows that the model can
identify the faulty signal more accurately and effectively detect the faulty vibration signal using the
detection algorithm.
Citation: Kumar, P.; Shih, G.-L.; Yao,
C.-K.; Hayle, S.T.; Manie, Y.C.; Peng,
Keywords: optical fiber; fiber sensor; fiber bragg grating (FBG); smart industry; signal detection;
P.-C. Intelligent Vibration Monitoring
YOLO V7; vibration signal
System for Smart Industry Utilizing
Optical Fiber Sensor Combined with
Machine Learning. Electronics 2023,
12, 4302. https://doi.org/10.3390/
electronics12204302 1. Introduction
The traditional method of monitoring machine health has relied on a particular opera-
Academic Editors: Nakkeeran
Kaliyaperumal and Suoda Chu
tor for handling the system. However, this manual approach lacked real-time capabilities,
consumed time, and needed human resources. To address these limitations, many studies
Received: 18 September 2023 have been conducted. Fiber sensors and artificial intelligence are two rapidly evolving
Revised: 8 October 2023 technologies. When fiber sensors are combined with machine learning, they offer solutions
Accepted: 15 October 2023 to many problems related to optical fiber sensors, such as motor machines in the industry
Published: 17 October 2023
generating vibration signals using different sensors. Sensors designed to measure various
strains address numerous issues, with vibrations being one of the challenges addressed
globally [1–5]. In many types of research, object detection is associated with the fiber
Copyright: © 2023 by the authors.
sensor and is a hot topic in vibration for daily applications, such as industrial motors,
Licensee MDPI, Basel, Switzerland. automatic machines, turbines, tunnels, elevators, and railway tracks problems to detect
This article is an open access article early faults and signals intrusion classification in structures to determine early warning.
distributed under the terms and Smart Industry employs various types of motors in different output machinery [6,7]. The
conditions of the Creative Commons primary importance of motor monitoring in various industries lies in its early fault de-
Attribution (CC BY) license (https:// tection capabilities, performance enhancement, and health assessment, especially in the
creativecommons.org/licenses/by/ vehicle sector. Safety concerns are a paramount consideration, hence the utilization of
4.0/). artificial intelligence algorithms for monitoring. Some studies of electronic sensing raised

Electronics 2023, 12, 4302. https://doi.org/10.3390/electronics12204302 https://www.mdpi.com/journal/electronics

Electronics 2023, 12, 4302 2 of 13

a lot of concerns, such as electromagnetic interference sensitivity and space constraints.

Some studies reflect that electromagnetic interference can influence electronic sensors to
cause inaccurate readings [8]. However, to overcome these limitations, an FBG vibration
sensor is used. FBG has the advantages of high sensitivity, long-distance transmission,
immunity to electromagnetic interference, high multiplexing, lightweight, and working
in a tiny space. This method also addresses challenges related to long distances, space
limitations, and high sensitivity while delivering accurate results. Some research highlights
the remarkable and unique properties of FBGs, and the interrogation technique involves
both wavelength-shifting and intensity-based methods to create adaptable sensors [9,10].
This detection technique works for real-time fault detection at an applied place for finding
the faulty site point or faulty machine with the association of artificial intelligence such as
object detection or machine learning [9–14].
In previous detection methods, various approaches are employed, including the
convolutional neural networks (CNNs), Faster R-CNN, and different versions of YOLO
ranging from YOLO V1 to YOLO V5 along with benefits of YOLO V7. Each version of
YOLO is tailored to achieve specific outcomes or improvements in the detection process.
Tasks related to non-linear problems have high complexity of processing and increment in
processing time. For machine learning or deep learning, the CNN method automatically
extracts features and understands the pattern for classifying the generated signal. These
CNN methods are classified into two categories: First is a one-dimensional convolutional
neural network (1-D CNN) that receives the transmit signal directly, and another is a two-
dimensional convolutional neural network (2-D CNN) that receives input as a converted
image directly from the signal. However, the standard method 2-D CNN model is only
capable of doing a signal target object detection in the picture, so during the multiple signal
generation scenarios, problems cannot be solved using this system [15–18]. Another is
a faster region-based convolutional neural network (Faster R-CNN), which is a popular
object detection algorithm. This solved multiple object detection problems instead of the
convolutional neural network [19]. This introduced the regional proposal network (RPN).
The RPN is a fully pre-trained convolutional neural network that is used on the feature and
generated by backbone networks, such as the visual geometry group (VGG-16) and residual
neural network (ResNet). VGG-16 is a convolutional neural network that has 16 layers in
the architecture, and ResNet is a model with a weight that learns residual function with
reference to input layers. However, this is a popular detection algorithm but has demanded
more computation than its predecessors because of the addition of the region proposal
network. In addition, the training process is time-consuming, and the network acquires
large computation resources during inference. Moreover, this Faster R-CNN is computing
single scaling processing and fixed anchor boxes process with limited ability to vary the
size of the detected object and lies on already defined anchor boxes, which is not ideal for
the size object ratio, respectively [20,21].
Some studies of object detection or signal detection have various kinds of result
variations, and their outcomes vary due to their different kinds of versions [22]. Every
year, the YOLO version is updated with some new advances in the model architecture [23].
In a series of YOLO, introduced in 2016 and the first real-time object detection algorithm,
YOLO V1 converts input images to the grid and predicts label boxes and probabilities
from the grid cells. The backbone of this version is Darknet-19; the number of layers of
the network is 19 convolutional layers. The main lack of this version is a lower value
result of mean average precision (MAP). In 2017, YOLO V2 was introduced, also known
as YOLO 9000, to improve the lack of YOLO v1; this version introduced the anchor boxes
and feature pyramid network (FPN) in various scales, and it improves the detection
accuracy [22–24]. This YOLO V2 used Darknet-19 and Darknet-53 as the backbone that was
used for small datasets and custom datasets in maximum count, respectively. The study
review reflects that complexity is increased in comparison to YOLO V1 because the number
of convolutional layers is 53, and the maximum layer needs more space for computation;
this is a struggle for small object detection compared to modern detection methods. In
Electronics 2023, 12, 4302 3 of 13

early 2018, YOLO V3 was introduced again to improve the detection accuracy and small
object detection. The version multi-scale detection approach was adopted with three kinds
of scales in the process of detection, but the Darknet-53 backbone did not change because
of the powerful feature extraction; although this provides benefits to the network, one
limitation is that it is more expensive and complex computationally [25,26]. Further, YOLO
V4 was launched in 2020, and developers improved the backbone Darknet-53 to cross-stage
partial (CSP) Darknet-53, which was the new backbone of this detection algorithm; CSP
Darknet-53 employs the cross-stage partial connection to increase the flow of data across
every step-in layer to layer path aggregation network (PANet) built in YOLO V4 to handle
the multi-scale features and improve the accuracy at the different quality of images. For
saving this multi-scale contextual data effectively, spatial pyramid pooling (SPP) is used.
Moreover, self-attention module (SAM) and pseudo-attention network (PAN) modules can
also be introduced to increase the network’s ability to approach the features in the image;
however, due to more computational demand than its predecessors, this affects the speed of
resource requirements such as particular power and memory specified. At the end of 2021,
YOLO V5 introduced some new features, such as implementing the PyTorch framework
because it was easy to use and implement, and the backbone was CSPDarknet-53, which
was referred to by the previous version, although the accuracy problem was occurring in
that version. In the latest research paper, improved methods are also being used to increase
the accuracy of detection objects [27,28].
In this paper, the FBG sensor is employed for sensing the vibration from the vibration
sources, such as various kinds of machinery factories, as shown in the schematic diagram,
Figure 1. However, in the experiment setup, only motors have been used to prove the
experiment on a minor stage with the motors. For detection, the four kinds of vibration
conditions are chosen here in the experiment. To detect signal and classification, the recent
version of YOLO is being employed. Although it is not the latest version, it offers improved
accuracy and speed compared to the earlier model networks. The evaluation metrics result,
such as MAP or latency of this model, is better than other detectors. YOLO V7 precision
detection outcomes in research are quite impressive. The overall structure is similar to
YOLO V5, like Backbone, PAN, FPN, and three different head scales, despite this module
changing a lot. For the feature extraction unit, Efficient Layer Aggregation Network (ELAN)
is being used here, with max pooling and convolutional with stride 2 in the down-sampling
method [29,30]. The following contribution in this paper is to identify the optical fiber
vibration signals.
A. Integrating FBG sensors and artificial intelligence for diverse motor conditions with
an advanced YOLO V7 architecture.
B. The Extended ELAN is implemented as a feature extraction unit and max-pooling,
and convolution in YOLO V7 along stride 2 is used in down-sampling. Besides, the
network contains a backbone, a head with three input branches, a path aggregation
network, and a feature pyramid network, which causes improved efficiency of
signal recognition.
C. YOLO V7 detection technique is implemented to accomplish the detection and
classification of vibration signals, which means improving the capability of signal
identification with high accuracy.
The rest of this paper is organized as follows: Section 2 describes the material and
methods in which we elaborate on data collection, label annotation methods, and proposed
YOLO V7 architecture and its theoretical foundation of the algorithm. Section 3 presents
the evaluation performance of the YOLO V7 model and analyses through the experiment
of vibration signal detection datasets. Finally, we provide a conclusion of this work in
Section 4.
Electronics
Electronics 2023, 12, x4302
FOR PEER REVIEW 4 4ofof 13
13

Figure 1. Conceptual schematic diagram of FBG integrated with YOLO V7 in the smart industry
Figure 1. Conceptual schematic diagram of FBG integrated with YOLO V7 in the smart industry with
with a distribution unit and a control unit as a central unit for the PCB Factory, electric vehicle
a distribution unit and a control unit as a central unit for the PCB Factory, electric vehicle factory, and
factory, and material factory. (PCB: Printed circuit board, FBG: Fiber Bragg grating).
material factory. (PCB: Printed circuit board, FBG: Fiber Bragg grating).
2.
2. Materials
Materials and and Methods
Methods
2.1.
2.1. Optical
Optical Fiber
Fiber Vibration
Vibration Sensing
Sensing System
System forfor Smart
Smart Industry
Industry
FBG
FBG vibration
vibration sensing
sensing technique
technique is aa widely
widely usedused approach.
approach. In In this
this experiment,
experiment, FBGsFBGs
are
are used
used forfor sensing
sensing thethe vibration
vibration of of machines
machines due due toto their
their high
high sensitivity
sensitivity and and capability
capability
through
through the thesingle-mode
single-mode fiber.
fiber. TheTheFBGFBG interrogator
interrogator is a wavelength
is a wavelength detector detector that
that measures
measures
the shiftedthe shifted wavelength
wavelength of thelight
of the reflected reflected
from light
sensors.from Assensors.
a result,As a result,
it is able toitmonitor
is able
to
themonitor
signal at the signal atpoints
different different points
of the stage.ofInthethestage. In the experimental
experimental setup, foursetup,motorfour motor
conditions
are used to generate the vibration. Figure 2 presents how to
conditions are used to generate the vibration. Figure 2 presents how to set up an experi- set up an experiment to
establish
ment an optical
to establish anfiber vibration
optical sensing system.
fiber vibration sensingThe process
system. Theofprocess
this sensing technique
of this sensing
consists ofconsists
technique severalofsteps.
severalFirstly,
steps.the attached
Firstly, FBG sensor
the attached FBG is at a different
sensor location
is at a different on
loca-
the machines
tion on the machines(motor), as shown
(motor), in Figure
as shown 1. For1.reference,
in Figure everyevery
For reference, sensor responds
sensor respondsat a
specific
at location,
a specific such such
location, as theasvibration sensingsensing
the vibration part of part
the smart
of theindustry. MultipleMultiple
smart industry. couplers
are deployed, such as coupler1, coupler2 and coupler3.
couplers are deployed, such as coupler1, coupler2 and coupler3. Each coupler has Each coupler has FBG sensors
FBG
connected to the printed circuit board factory, an electric vehicle
sensors connected to the printed circuit board factory, an electric vehicle factory, and a factory, and a material
factory, respectively.
material The whole
factory, respectively. The setup
wholetransmits a signal
setup transmits to the to
a signal central unit, unit,
the central and then
and
optical
then fiber fiber
optical is an is
aid
antoaidconnect and transmit
to connect and transmitthe reflected light signal
the reflected to thetowavelength
light signal the wave-
detector.
length The wavelength
detector. The wavelengthdetector is the is
detector FBGtheinterrogator,
FBG interrogator, and these shiftsshifts
and these are detected
are de-
by the interrogator and converted into vibration data(datasets),
tected by the interrogator and converted into vibration data(datasets), the process the process also known
also
known as data acquisition. One channel is detecting multiple vibrations by channeltoport
as data acquisition. One channel is detecting multiple vibrations by channel port the
source; these are the benefits of FBGs with FBG interrogators.
to the source; these are the benefits of FBGs with FBG interrogators.
Electronics2023,
Electronics 2023,12,
12,4302
x FOR PEER REVIEW 55 of 13
13

Figure 2.
Figure 2. The
The proposed
proposed experimental
experimental setup
setup of
of FBG
FBG Sensor,
Sensor, FBG
FBG Interrogator
Interrogator combined
combined with
with YOLO
YOLO
V7algorithm
V7 algorithmforfordetecting
detecting faulty
faulty motor
motor signal.
signal. (SL:(SL: Swept
Swept laser,laser, PD: Photodetector,
PD: Photodetector, FBG:Bragg
FBG: Fiber Fiber
Bragg grating, PC: Personal computer).
grating, PC: Personal computer).

In Figure
In Figure 2,
2, the
the experimental
experimental setup
setup is
is aa real
real experiment
experiment in in which
which vibration
vibration sources
sources
have two
have two kinds
kinds of
of independent
independentmotors
motorswithwithfourfourconditions
conditionsororfaulty
faultysignals,
signals,asas
shown
shown in
the data labeling section of the signal detection process. FBGs are used
in the data labeling section of the signal detection process. FBGs are used for collecting for collecting vi-
bration using
vibration usingananFBGFBGinterrogator
interrogatorwith
withaaswept
sweptlaser
laser(SL)
(SL)and
andphotodetector
photodetector (PD).
(PD). Single-
mode fiber cable connects the FBG
mode fiber cable connects the FBG sensor andsensor and the FBG interrogator, and the network
interrogator, and the network cablecable
transmits data
transmits data to
to the
the PC.
PC. The
The signal
signal detection
detection algorithm
algorithm collects
collects that
that data
data and
and processes
processes
it, further
it, further training
training the
the algorithm.
algorithm. YOLO
YOLO V7 V7 algorithm
algorithm is is embedded
embedded herehere for
for detecting
detecting the
the
faulty signal.
faulty signal. A trained model also can be used for real-time detection of test data. Below,
we have
we have described
described every
every parameter
parameter of of the
the signal
signaldetection
detectionsystem
systemof ofthis
thisexperiment.
experiment.

2.2.
2.2. Signal
Signal Detection
Detection System
System
This
This subsection
subsection concentrates
concentrateson onthe
the signal
signal detection
detectionsystem.
system. The
The first
first two
two subsections
subsections
describe
describe the data collection and annotation of the dataset, followed by an explanation of
the data collection and annotation of the dataset, followed by an explanation of
the
the vibration
vibrationdetection
detectionalgorithm
algorithm(YOLO
(YOLO V7).
V7).The
Thedataset is divided
dataset intointo
is divided a training set and
a training set
aand
validation set in
a validation seta in
specific ratio
a specific in the
ratio data
in the division
data process.
division process.Furthermore,
Furthermore, it provides
it providesa
detailed description of the experimental environment setup within the system.
a detailed description of the experimental environment setup within the system. Lastly, Lastly, the
subsequent segment
the subsequent segmentpresents a table
presents listing
a table the hyperparameters.
listing the hyperparameters.
2.2.1. Data Collection
2.2.1. Data Collection
Data collection is a method in which datasets are collected; here, data is collected
Data collection is a method in which datasets are collected; here, data is collected
through an experiment done using four different conditions of motor vibration: motor 1,
through an experiment done using four different conditions of motor vibration: motor 1,
motor 2, motor 1 abnormal, and motor 2 abnormal. Four different vibration signal data
motor 2, motor 1 abnormal, and motor 2 abnormal. Four different vibration signal data
are created to collect, as shown in the proposed architecture. After that, possible vibration
are created to collect, as shown in the proposed architecture. After that, possible vibration
signal data is gathered. Data is collected in the form of numerical values of amplitude or
signal data is gathered. Data is collected in the form of numerical values of amplitude or
power and time, and then collected data is converted to the form of an xlxs file extension for
power and time, and then collected data is converted to the form of an xlxs file extension
showing in rows and columns. Further number data were converted into video machine
for showing
learning in rows
libraries, and
such ascolumns. Further number
pandas, matplotlib, data were converted
matplotlib.animation, andinto
the video
DateTimema-
chine learning libraries, such as pandas, matplotlib, matplotlib.animation,
method. Generated videos are converted to images using Python libraries and open CV; and the
DateTime method. Generated videos are converted to images using Python libraries
this provides us with a dataset. The images generated are the y-axis as the wavelength of and
open
the CV; this
sensor andprovides
the x-axisus
aswith
time.a Each
dataset. The
class images
consists ofgenerated areand
600 images, thethe
y-axis
totalasnumber
the wave-of
length of the sensor and the x-axis as time. Each class consists of 600 images, and
images is 2400 images that are generated from the source video, as shown in Table 1. The the total
number of images is 2400 images that are generated from the source video, as shown in
Electronics 2023, 12, 4302 6 of 13

collected data is annotated or labeled; the below division has a description of labeling the
signal vibration image.

Table 1. Dataset.

Class Label Number of Images

Motor Normal 1 600
Motor Abnormal 1 600
Motor Normal 2 600
Motor Abnormal 2 600
Total 2400
Training size 1680 (70%)
Validation size 720 (30%)

2.2.2. Signal Annotations

YOLO needs label data to train a model of collected vibration datasets; then, vibra-
tion image data is labeled using labelImg. LabelImg is an open-source graphical image
annotation tool that is used to generate labels of the corresponding image for training an
objection detection model. In labelImg, four classes were created; the name annotation
of four classes are as follows: Motor 1 normal, Motor 1 Abnormal, Motor 2 normal, and
Motor 2 Abnormal. After classes are image labelled, each class generates 600 labels to their
corresponding class. The number of images used for training YOLO is 2400 images, of
which 70% were used for training and 30% for validation, as shown in Table 1.

2.2.3. YOLO V7
In the previous section, the data collection and annotation method are discussed,
followed by the description of the YOLO V7 algorithm, which was developed by a re-
searcher in early 2022. It incorporates a new efficient layer called Extended Efficient Layer
Aggregation Networks (E-ELAN), model scaling, and reparameterization for concatenation-
based models to achieve an appropriate balance between detection efficiency and precision.
YOLO is divided into four major modules: input, backbone, head, and prediction module.
Input: The first steps in the YOLO V7 in input images with their corresponding label
masking the object detection, the size is 416 × 416 as a color image as input of the backbone
layer network.
Backbone: The backbone consists of three main modules: E-ELAN, CBS, and MPI
modules. Convolution, batch normalization, and SiLU activation functions are the three
primary components of the CBS module. The E-ELAN module keeps the basic ELAN
design but improves the network’s learning capacity by directing discrete groups of feature
computational blocks to learn more diverse features while keeping the original gradient
path. The MPI module is organized using a combination of CBS and MaxPool operations
with upper and lower branches. In the upper branch, MaxPool is used to reduce the image’s
dimensions in half, both in length and width. A CBS operation with 128 output channels is
also used to reduce the image’s channel count in half. On the other hand, the lower branch
uses a CBS operation with a 1 × 1 kernel and stride to cut the number of channels in half.
Furthermore, the image’s length and breadth are cut in half using a CBS operation with a
3 × 3 kernel and 2 × 2 stride. Concatenation (Cat) is used to integrate the features extracted
from both branches. Max pool focuses on collecting the most valuable information from
localized locations, whereas CBS keeps a wide range of valuable information from these
tiny local areas. This combination method improves the network’s capacity to extract
meaningful features from input data greatly.
Head: This part of the YOLO V7 network is built using the feature pyramid network
(FPN) structure, which implements the PANs structure design. Batch normalization, CBS
block (SiLU activation), and many convolutional layers are composed together, coupled
with the convolutional spatial pyramid (CSP) and spatial pyramid pooling (SPP) module,
elongate E-ELAN and MaxPool 2 (MP2) that improves accuracy and aim to feature ex-
Electronics 2023, 12, 4302 7 of 13

traction and model optimization. The layer E-LAN-H composition of many feature layers
depends on E-ELAN. There are two MP blocks; both are the same, with a different number
of the output channel.
Prediction: This is a prediction network that has three rep design structures that
employ a number of picture channels for the head network’s output features. Here, 1 × 1
convolution is used to predict confidence, category, and anchor frame. VGG is the inspi-
ration for the Rep structure and, as a result, reduces the complexity of the model without
diminishing its predictive ability.
To train the YOLO V7 model, the following minimum system specifications were
employed: an experimental apparatus equipped with an Intel(R) Core (TM) i7-9700k CPU
@ 3.60 Hz, and 16 GB of installed RAM and NVIDIA GeForce RTX 2080Ti, CUDA 11.8.

2.2.4. Model Hyperparameter

The hyperparameters, as outlined in Table 2, play a pivotal role in determining the
effectiveness of YOLO V7. They aid in the quest for optimal results from the neural network.

Table 2. Experimental configuration.

Parameter Configuration
Learning Rate 1 × 10−5
Momentum 0.98
Weight Decay 0.001
Batch Size 16
Optimizer Adam and SGD
Image Size 416 × 416
Epochs 200

After the setup, the process was initiated by meticulously configuring the system and
then proceeding to train the algorithm with the predefined hyperparameters.

3. Results
In this paper, we propose an experimental setup run with defined parameters and
provide the results. Initially, we obtain predictions on the validation dataset using
the YOLO V7 algorithm, and subsequently, we evaluate the metrics as outlined in the
following subsections.

3.1. Prediction of YOLO V7 Model

In Figure 3a–d prediction results of YOLO V7, the bounding boxes highlight the target
signals generated from industrial machinery motors. The bounding box prediction is
dependent on both the anchor box size and the feature information outputs of the network.
The size of the anchor box also referred to as the a priori box, is established through prior
experience. The network’s prediction information adjusts the position of the preceding box
to yield the predicted bounding box. The process of creating prediction boxes using test
signals is illustrated in Figure 3. The four kinds of different color bounding boxes reflect the
results of different predicted classes. In Figure 3a inscribed by a light green color bounding
box, is the predicted result of the proposed model when motor 1 is in a normal condition.
On the other hand, in Figure 3b, inscribed by a blue color box, is the predicted result of the
proposed model when motor 1 is in an abnormal condition. In Figure 3c, inscribed by a
green color box, is the predicted result of the proposed model when motor 2 is in a normal
condition. Furthermore, Figure 3d inscribed by an orange color box, is the predicted result
of the proposed model when motor 2 is in an abnormal condition.
Electronics 2023, 12, x FOR PEER REVIEW 8 of 13

Electronics 2023, 12, 4302 8 of 13

color box, is the predicted result of the proposed model when motor 2 is in an abnormal
condition.

Figure 3. The predicted result of the proposed model is when the motors
motors are
are in
in different
different conditions.
conditions.
(a) motor
(a) motor 11 normal
normalcondition,
condition,(b)
(b)motor
motor1 1abnormal
abnormalcondition, (c)(c)
condition, motor 2 normal
motor condition,
2 normal andand
condition, (d)
motor
(d) 2 abnormal
motor condition.
2 abnormal condition.

3.2. Model Evaluation Metrics

To
To evaluate
evaluate the
the performance
performance of the
the proposed
proposed model,
model, we used different
different performance
metrics
metrics such as recall, precision,
precision, and intersection
intersection over the union (IoU) at a threshold of 0.5,
average precision (AP),
(AP), and
and mean
mean average
average precision
precision (MAP)
(MAP) [3,31].
[3,31].
𝐺𝑟𝑜𝑢𝑛𝑑Truth
Ground 𝑇𝑟𝑢𝑡ℎT⋂Detection
𝐷𝑒𝑡𝑒𝑐𝑡𝑖𝑜𝑛Result
𝑅𝑒𝑠𝑢𝑙𝑡 (1)
𝐼𝑜𝑈
IoU = = (1)
(1)
𝐺𝑟𝑜𝑢𝑛𝑑Truth
𝑇𝑟𝑢𝑡ℎ ⋃Detection
𝐷𝑒𝑡𝑒𝑐𝑡𝑖𝑜𝑛Result
𝑅𝑒𝑠𝑢𝑙𝑡
S
Ground
In Equation (1), the intersection over union (IoU) metric, is used to calculate the de-
In Equation (1), the intersection over union (IoU) metric, is used to calculate the degree
gree of overlap between the system-predicted and ground-truth bounding boxes in the
of overlap between the system-predicted and ground-truth bounding boxes in the original
original image. The intersection and concatenation ratio of the detection results to ground
image. The intersection and concatenation ratio of the detection results to ground truth
truth is calculated, where ground truth is actual data and the detection result is the output
is calculated, where ground truth is actual data and the detection result is the output
produced by
produced by the
the algorithm.
algorithm. The
The equation
equation below
below represents
represents the
the precision
precisioncalculation
calculation[31].
[31].
𝑻𝒓𝒖𝒆 𝑷𝒐𝒔𝒊𝒕𝒊𝒗𝒆
𝑷𝒓𝒆𝒄𝒊𝒔𝒊𝒐𝒏 = True Positive (2)
Precision = 𝑻𝒓𝒖𝒆 𝑷𝒐𝒔𝒊𝒕𝒊𝒗𝒆 + 𝑭𝒂𝒍𝒔𝒆 𝑷𝒐𝒔𝒊𝒕𝒊𝒗𝒆 (2)
True Positive + False Positive
𝑇𝑟𝑢𝑒 𝑃𝑜𝑠𝑖𝑡𝑖𝑣𝑒
𝑅𝑒𝑐𝑎𝑙𝑙 = (3)
Recall = 𝑇𝑟𝑢𝑒 𝑃𝑜𝑠𝑖𝑡𝑖𝑣𝑒 + 𝐹𝑎𝑙𝑠𝑒 𝑁𝑒𝑎𝑔𝑎𝑡𝑖𝑣𝑒
True Positive
(3)
True Positive + False Neagative
In Equations (2) and (3), where the true positive shows the predicted label is true and
the actual label is also true, the false positive points out the predicted label which is true and
the actual label is false and false negative means the predicted label is false, and the actual
Electronics 2023, 12, 4302 9 of 13

label is true. So, recall and precision reflect on possible true positive instances and ensure
that predicted predictions are valid and accurate to classify the vibration signals [3,31].

1 n
Z 1

k k∑
mAP = ( Precsion × Recall ) k (4)
=0 0

In Equation (4), mean average precision (mAP) is the most important graph commonly
used for object detection, and it is an incremental calculation of average precision in order
to calculate the mAP. First, calculate the average precision of a number of classes (k) and
then calculate the mean all average precision for measuring the mean average. In the next
section, vibration is here in order to measure all these values to show the graphs such as
confusion matrix, precision-recall curve, and so on [3,26].

3.3. Proposed Model Analysis on Experimental Vibration Datasets

The proposed algorithm evaluated the performance of the signal detection model
(YOLO V7) on the vibration dataset, as shown in Figures 4 and 5. The four classes are
accurately classified using YOLO V7, and each class’s separate accuracy is shown also at the
precision-recall relation curve, confusion matrix, and Loss graph. The result of each class is
as follows: motor 1 normal, motor 1 abnormal, motor 2 normal, and motor 2 abnormal is
shown in the confusion matrix, with the detection result of false negative and false positive
with respect to each class. Besides that, a precision-recall curve assesses the performance
of the detection model for correctly identifying the positive detection with fewer false
positives. The curve precision highlights the true positive to the predicted positive and,
meanwhile, the ratio of true positive to actual positive. The area under the curve (AUC)
reflects the performance. As a result, high precision shows the accurate positive vibration
signal and minimal false positive degradation signal. In Figure 6, the loss graph displays
the training and validation loss curve, with the blue curve representing the training loss and
the red curve representing the validation loss. Loss is a function that consists of bounding
box loss, confidence loss, and class loss. It is calculated by the addition of a bounding box,
confidence, and class loss. Bounding loss is a prediction of the accurate coordinates of the
box; it is also known as localization loss. Confidence loss is also known as objectness loss,
which is defined as the prediction of the object in the grid cell or anchor box, and class loss
predicts that object contained in the bounding box. The main focus of the three losses are
increased accuracy of classification, localization, and signal detection. In the experiment,
when we calculate the loss, it consistently decreases during the training and validation
of the signal detection model. The training and validation loss converg approximately at
75 epochs. In addition, during the training validation time, the model at 160 epochs was
totally stable, and loss was minimal. As Figure 6 shows, at every epoch after 75 epochs, the
loss curve was continuously in progress until optimal epochs. Our loss result was optimal
at 195 epochs. Along with a learning rate of 1 × 10−5 , momentum was 0.98, weight decay
was 0.001, batch size was 16, the optimizer was Adam, the image size was 416 × 416, and
total epochs were 200. When 191 epoch loss of validation 0.0005 approximately which was
the best loss and performance was quite impressive as shown in the figures below.
Electronics 2023, 12, x FOR PEER REVIEW 10 of 13
Electronics
Electronics 2023,
2023, 12,
12, x4302
FOR PEER REVIEW 1010ofof13
13

Figure
Figure4.4.The
Theconfusion
confusionmatrix
matrixofoflabels
labelsfor
fordifferent
differentclasses
classesororconditions
conditionsofofmotors.
motors.
Figure 4. The confusion matrix of labels for different classes or conditions of motors.

Theprecision-recall
Figure5.5.The
Figure precision-recallcurve
curverelation
relationfor
foreach
eachclass/condition
class/conditionofof themotor.
the motor.
Figure 5. The precision-recall curve relation for each class/condition of the motor.
Electronics 2023, 12, x FOR PEER REVIEW 11 of 13

Electronics 2023, 12, 4302 11 of 13

Figure 6.
Figure The proposed
6. The proposed model
model training
training and validation
validation losses
losses at
at various
various epoch
epoch numbers.

4. Conclusions
4. Conclusions
In this
In this paper,
paper, aa novel
novel approach
approach is is proposed
proposed to
to monitor
monitor the
the health
health of
of the
the industrial
industrial
motor in the smart industry to enhance the durability and capability of the sensing system
motor in the smart industry to enhance the durability and capability of the sensing system
as well as decrease the cost. This proposed model efficiently detects vibration signals
as well as decrease the cost. This proposed model efficiently detects vibration signals gen-
generated using an FBG interrogator, combined with the identification technology of YOLO
erated using an FBG interrogator, combined with the identification technology of YOLO
V7, making it an excellent performer with minimal equipment. The accuracy of four
V7, making it an excellent performer with minimal equipment. The accuracy of four con-
conditions “motor 1 normal is 99.7%”, “motor 1 abnormal is 99.2%”, “motor 2 normal
ditions “motor 1 normal is 99.7%”, “motor 1 abnormal is 99.2%”, “motor 2 normal is
is 99.6%”, “motor 2 abnormal is 99.7%”, and the average accuracy mean average of all
99.6%”, “motor 2 abnormal is 99.7%”, and the average accuracy mean average of all classes
classes are 99.5% at a threshold of 0.5. Besides, the proposed model accurately classifies the
are 99.5% at a threshold of 0.5. Besides, the proposed model accurately classifies the signal
signal generated and provides an excellent performance. Therefore, optical fiber sensors
generated and provides an excellent performance. Therefore, optical fiber sensors com-
combined with YOLO V7 easily output accuracy, robustness, sensitivity, and cost-effective
bined with YOLO V7 easily output accuracy, robustness, sensitivity, and cost-effective
apparatus setup for handling the problem of vibration.
apparatus setup for handling the problem of vibration.
Author Contributions: Conceptualization, P.K., G.-L.S., C.-K.Y. and P.-C.P.; methodology, P.K., G.-
Author Contributions:
L.S., C.-K.Y. and P.-C.P.;Conceptualization, P.K.,G.-L.S.
data preparation, P.K., G.-L.S., C.-K.Y.,
and and
C.-K.Y.; P.-C.P.; methodology,
software, P.K.,
P.K. and G.-L.S.; G.-
model
L.S., C.-K.Y., and P.-C.P.; data preparation, P.K., G.-L.S., and C.-K.Y.; software, P.K., and G.-L.S.;
validation, P.K. and G.-L.S.; formal analysis, P.K., G.-L.S., C.-K.Y., S.T.H., Y.C.M. and P.-C.P.; investiga-
model validation, P.K., and G.-L.S.; formal analysis, P.K., G.-L.S., C.-K.Y., S.T.H., Y.C.M., and P.-C.P.;
tion, P.K., G.-L.S., C.-K.Y., S.T.H., Y.C.M. and P.-C.P. All authors have read and agreed to the published
investigation, P.K., G.-L.S., C.-K.Y., S.T.H., Y.C.M. and P.-C.P. All authors have read and agreed to
version of the manuscript.
the published version of the manuscript.
Funding: This work was supported by the National Science and Technology Council, Taiwan, under
Funding: This work was supported by the National Science and Technology Council, Taiwan, under
Grant NSTC 112-2221-E-027-076-MY2.
Grant NSTC 112-2221-E-027-076-MY2.
Institutional Review Board Statement: Not applicable.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.
Conflicts of Interest: The authors declare no conflict of interest.
References
References
1. Li, W.; He, Y.; Geng, P.; Yang, Y. Research on Posture Sensing and Error Elimination for Soft Manipulator Using FBG Sensors.
1. Li, W.; He, 2023,
Electronics Y.; Geng, P.; Yang,
12, 1476. Y. Research on Posture Sensing and Error Elimination for Soft Manipulator Using FBG Sensors.
[CrossRef]
2. Electronics 2023, 12, 1476. https://doi.org/10.3390/electronics12061476.
Hayle, S.T.; Manie, Y.C.; Yao, C.K.; Yeh, T.Y.; Yu, C.H.; Peng, P.C. Hybrid of Free Space Optics Communication and Sensor System
2. Hayle, S.T.; Manie,
Using IWDM Y.C.; Yao,
Technique. C.K.;
J. Light. Yeh, T.Y.;
Technol. 2022,Yu,
40, C.H.; Peng, [CrossRef]
5862–5869. P.C. Hybrid of Free Space Optics Communication and Sensor
3. System Using IWDM Technique. J. Light. Technol. 2022, 40, 5862–5869.
Jiangwei, Z.; Jiaqing, M.; Xinrong, M.; Jincheng, H.; Fubao, S. Optical Fiber Vibration Signal Identification Method Based on
Improved YOLOv4. Sensors 2022, 22, 9259.
Electronics 2023, 12, 4302 12 of 13

4. Kashaganova, G.; Kozbakova, A.; Kartbayev, T.; Balbayev, G.; Togzhanova, K.; Alimseitova, Z.; Orazaliyeva, S. Research of a Fiber
Sensor Based on Fiber Bragg Grating for Road Surface Monitoring. Electronics 2023, 12, 2491. [CrossRef]
5. Janusz, J.; Patrycja, A.P. Fibre optic FBG sensors for monitoring of the temperature of the building envelope. Materials 2021,
14, 1207.
6. Tan, Y.; Tianliang, L.; Li, C.; Xiao, T. Mechanical structural damage detection based on distributed measurement. In Proceedings
of the 2014 International Conference on Innovative Design and Manufacturing (ICIDM), Montreal, QC, Canada, 13–15 August
2014; pp. 111–116.
7. Peng, P.C.; Feng, K.M.; Peng, W.R.; Chiou, H.Y.; Chang, C.C.; Chi, S. Long-distance fiber grating sensor system using a fiber ring
laser with EDWA and SOA. Opt. Commun. 2005, 252, 127–131. [CrossRef]
8. Sun, Y.; Li, H.; Fan, C.; Yan, B.; Chen, J.; Yan, Z.; Sun, Q. Review of a Specialty Fiber for Distributed Acoustic Sensing Technology.
Photonics 2022, 9, 277. [CrossRef]
9. Li, T.; Guo, J.; Tan, Y.; Zhou, Z. Recent advances and tendency in fiber Bragg grating-based vibration sensor: A review. IEEE Sens.
J. 2020, 20, 12074–12087. [CrossRef]
10. de la Torre, O.; Floris, I.; Sales, S.; Escaler, X. Fiber Bragg Grating Sensors for Underwater Vibration Measurement: Potential
Hydropower Applications. Sensors 2021, 21, 4272. [CrossRef] [PubMed]
11. Cheng, J.; Song, Q.; Peng, H.; Huang, J.; Wu, H.; Jia, B. Optimization of VGG16 Algorithm Pattern Recognition for Signals of
Michelson–Sagnac Interference Vibration Sensing System. Photonics 2022, 9, 535. [CrossRef]
12. Peng, P.C.; Tseng, H.Y.; Chi, S. Long-distance FBG sensor system using a linear-cavity fiber Raman laser scheme. IEEE Photonics
Technol. Lett. 2004, 16, 575–577. [CrossRef]
13. Feng, Z.; Shufeng, Z.; Hongzhen, Z.; Licheng, Y.; Qin, F.; Jiahuan, H. An intelligent optical fiber-based prewarning system for oil
and gas pipelines. Opt. Fiber Technol. 2022, 71, 102953.
14. Xu, G.; Gui, X.; Xiaodong, N.; Yutang, D.; Wang, J. Study of non-contact detection sensor based on fiber Bragg grating. In
Proceedings of the 2011 International Conference on Transportation, Mechanical, and Electrical Engineering (TMEE), Changchun,
China, 16–18 December 2011; pp. 990–993.
15. Permanasari, Y.; Ruchjana, B.N.; Hadi, S.; Rejito, J. Innovative Region Convolutional Neural Network Algorithm for Object
Identification. J. Open Innov. Technol. Mark. Complex. 2022, 8, 182. [CrossRef]
16. Yuan, W.; Yang, Q. A Novel Method for Pavement Transverse Crack Detection Based on 2D Reconstruction of Vehicle Vibration
Signal. KSCE J. Civ. Eng. 2023, 27, 2868–2881. [CrossRef]
17. Lin, S.; Wang, S.; Liu, T.; Liu, X.; Liu, C. Accurate Measurement of Bridge Vibration Displacement via Deep Convolutional Neural
Network. IEEE Trans. Instrum. Meas. 2023, 72, 5020016. [CrossRef]
18. Marshall, M.R.; Hellfeld, D.; Joshi TH, Y.; Salathe, M.; Bandstra, M.S.; Bilton, K.J.; Vetter, K. 3-D Object Tracking in Panoramic
Video and LiDAR for Radiological Source–Object Attribution and Improved Source Detection. IEEE Trans. Nucl. Sci. 2021, 68,
189–202. [CrossRef]
19. Liu, B.; Zhao, W.; Sun, Q. Study of object detection based on Faster R-CNN. In Proceedings of the Chinese Automation Congress
(CAC), Jinan, China, 20–22 October 2017; pp. 6233–6236. [CrossRef]
20. Zhang, K.; Houqiang; Huang, W. Defect Detection of Anti-vibration Hammer Based on Improved Faster R-CNN. In Proceedings of
the 7th International Forum on Electrical Engineering and Automation (IFEEA), Hefei, China, 25–27 September 2020; pp. 889–893.
21. Maity, M.; Banerjee, S.; Sinha, C.S. Faster R-CNN and YOLO based Vehicle detection: A Survey. In Proceedings of the 5th Interna-
tional Conference on Computing Methodologies and Communication (ICCMC), Erode, India, 8–10 April 2021; pp. 1442–1447.
22. Liu, K.; Tang, H.; He, S.; Yu, Q.; Xiong, Y.; Wang, N. Performance validation of YOLO variants for object detection. In Proceedings
of the 2021 International Conference on bioinformatics and intelligent computing, Harbin, China, 22–24 January 2021; pp. 239–243.
23. Hemmatirad, K.; Babaie, M.; Afshari, M.; Maleki, D.; Saiadi, M.; Tizhoosh, H.R. Quality Control of Whole Slide Images using the
YOLO Concept. In Proceedings of the IEEE 10th International Conference on Healthcare Informatics (ICHI), Rochester, MN, USA,
13–16 June 2022; pp. 282–287. [CrossRef]
24. Wahyutama, A.B.; Hwang, M. YOLO-Based Object Detection for Separate Collection of Recyclables and Capacity Monitoring of
Trash Bins. Electronics 2022, 11, 1323. [CrossRef]
25. Chen, D.; Ju, Y. SAR ship detection based on improved YOLOv3. In Proceedings of the IET International Radar Conference (IET
IRC 2020), Chongqing, China, 4–6 November 2020; pp. 929–934.
26. Yuanhong, L.; Zuoxi, Z.; Yangfan, L.; Zhi, Q. Real-Time Pattern-Recognition of GPR Images with YOLO v3 Implemented by
Tensorflow. Sensors 2020, 20, 6476.
27. Zhou, F.; Deng, H.; Xu, Q.; Lan, X. CNTR-YOLO: Improved YOLOv5 Based on ConvNext and Transformer for Aircraft Detection
in Remote Sensing Images. Electronics 2023, 12, 2671. [CrossRef]
28. Xiao, Y.; Chang, A.; Wang, Y.; Huang, Y.; Yu, J.; Huo, L. Real-time Object Detection for Substation Security Early-warning
with Deep Neural Network based on YOLO-V5. In Proceedings of the IEEE IAS Global Conference on Emerging Technologies
(GlobConET), Arad, Romania, 20–22 May 2022; pp. 45–50.
29. Wang, Y.; Wang, H.; Xin, Z. Efficient Detection Model of Steel Strip Surface Defects Based on YOLO-V7. IEEE Access 2022, 10,
133936–133944. [CrossRef]
Electronics 2023, 12, 4302 13 of 13

30. Hussain, M. YOLO-v1 to YOLO-v8: The Rise of YOLO and Its Complementary Nature toward Digital Manufacturing and
Industrial Defect Detection. Machines 2023, 11, 677. [CrossRef]
31. Kaiyue, L.; Qi, S.; Daming, S.; Lin, P.; Mengduo, Y.; Nizhuan, W. Underwater Target Detection Based on Improved YOLOv7. J.
Mar. Sci. Eng. 2023, 11, 677.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.