Application of fuzzy-logic to design fuzzy compensation controller for speed control system to reduce vibration of CBШ-250T drilling machine in mining industry
90
Journal of Mining and Earth Sciences, Vol 61, Issue 6 (2020) 90 - 96
Application of fuzzy-logic to design fuzzy compensation
controller for speed control system to reduce
vibration of CBШ-250T drilling machine in mining
industry
Dung Ngoc Le 1,*, Chi Van Dang 2
1 Dong Nai University of Technology ,Vietnam
2 Hanoi University of Mining and Geology, Vietnam
ARTICLE INFO
ABSTRACT
Article history:
Received
Accepted
The paper introduces fuzzy compensation control algorithm based on
fuzzy logic to control the rotation speed in CБШ-250T drilling machine.
The proposed solution uses an artificial neural network instead of a
vibration measuring sensor to identify the amplitude and vibration
frequency on a rotary drill. The vibration amplitude, vibration frequency
and setpoint of the drilling speed are the input variables for the fuzzy
logic unit. The fuzzy tool will diagnose the compensatory parameter δα
with the target to reduce the vibration of the drilling equipment. The
results were tested through modeling using Simulink_matlab tool. It can
be applied to control system to improve control quality, and reduce
vibration for CБШ-250T drilling machine, which is being operated on
mines in Quang Ninh area.
Available online
Keywords:
Drilling machine CBШ-250T,
Fuzzy compensation control,
Fuzzy logic,
Neural network.
Copyright © 2020 Hanoi University of Mining and Geology. All rights reserved.
law or algorithm to adjust the drilling mode
parameters (rotation speed and pressure) in
complex geological conditions and specific
mining environments in Vietnam to reduce
vibration. Many scientists in the field of mining
are interested in research.
Some previous studies of scientists in
Vietnam also mentioned optimal control of
drilling mode parameters based on object
modeling (B.Y. Lee, H.S. Liu, Y.S. Tarng, 1998);
(Claude E. Aboujaoude, 1991), which improves
the control scheme of the system (Nguyen Thac
Khanh, 2003) for 2 channels to control the
1. Introdution
Nowadays, CБШ-250T type drilling
machines are being widely used on mines in
Quang Ninh – Vietnam. During the drilling
process, the countersink is constantly in contact
with the rock, it has different hardness and
geological structure. The study found a suitable
_____________________
* Corresponding author
E - mail: dangvanchi@humg.edu.vn
DOI: 10.46326/JMES.2020.61(6).10
Dung Ngoc Le and Chi Dang Van/Journal of Mining and Earth Sciences 61 (6), 90 - 96
91
rotation speed and axial pressure. In the doctoral
thesis, the author Ngo Duc Thao proposed the
solution to automate the drilling process based
on the physical and mechanical properties of
rock (Ngo Duc Thao, 1971). Currently, no
research projects have been done to improve the
control system to reduce vibration for rotary
drilling machines in Vietnam.
control systems to reduce vibrations for drilling
machines (Alexei A. Zhukovsky, 1982); (Claude E.
Aboujaoude, 1991). Current studies in Viet Nam,
due to technological limitations and the means to
directly measure the hardness of rock and soil,
have many technical disadvantages. The idea
proposed to indirectly apply artificial neural
networks to identify soil hardness through
measuring the critical parameters of the process
such as rotation speed and pressure to promise
to bring positive results.
Based on the neural network's prediction
information, it is possible to develop a fuzzy
compensation algorithm (δα) to automatically
compensate for the opening angle α in the
thyristor control system to adjust the rotation
speed to suit the properties of the rock. The
proposed solution is checked through modeling
the control system on the simulation software.
The results confirm that the control system
adapts and responds well to the mining process,
reducing machine vibration, improving the
control system's quality while ensuring a good
working efficiency for the drilling equipment.
The topics related to the field of study with
the goal of reducing vibration for drilling
machines have been in research all over the
world. Jerome Rajnauth and Tennyson Jagai
performed vibration measurement on drilling
equipment, then the vibration signal was sent
through FFT spectrum analysis to assess the
vibration spectrum (Jerome Rajnauth and
Tennyson Jagai, 2012) proposing a suitable
algorithm to be embedded in the control system
to reduce the vibration of the machine at the
exploitation wells to a depth of hundreds of
meters. In another study (Edward A.
Branscombe, 2010) on PH120A (Rotary
Blasthole Drill), Edward A. Branscombe applied
DATAQ DI718 device to receive vibration
measurement signals from an accelerometer
sensor. This sensor set On the drill rod to
experiment measure vibration of the machine in
different geological conditions in iron ore mine.
The measurement signal was also via FFT
spectrum analysis. The author has established
the relationship between amplitude, spectrum in
different geological conditions and depths, and
other drilling process relationships such as
rotation speed, drill motor current, and drill bit
depth. In another study, Stuart Jardine, Dave
Malone and Mike Sheppard proposed an
algorithm to control vibration reduction by
measuring feedback signals from process
parameters: voltage, current, and drill motor
speed. Parameters put into the microprocessor
to predict the speed feedback compensation
signal. The difference between the set value and
the speed compensation feedback value is used
to adjust the PID controller's parameters (Stuart
Jardine, Dave Malone and Mike Sheppard; 1994).
In general, many previous authors' studies
show that there are many different approaches
and solutions that can be embedded in existing
2. Proposing rotation speed control system for
drilling machine
2.1. Diagram of the proposed principle for
the rotation speed control system
Diagram of the principle of controlling the
rotation speed on CБШ-250T drilling machines
as shown in Figure 1 (Эkcплуатационная
документация, 2003); (Nguyen Chi Tinh and
others. 2013).
In the control system of drilling machine
CБШ-250T, the signal setpoint Uđk is set directly
by the driver in the cabin, through the controller
ĐK to change the opening angle α. In the
proposed system, the opening angle α will be
compensated by the amount of δα through 2
devices (2 blocks) including:
+ Vibration sensor block: in the proposed
modeling, it is replaced by Neural Network
(Nguyen Phung Quang, 2004);(with the function
of recognizing amplitude and vibration
frequency after successful network training).
92
Dung Ngoc Le and Chi Dang Van/Journal of Mining and Earth Sciences 61 (6), 90 - 96
3 phase power
drill body
Neural Network
Card
MyRio
Labview
software
motor
excitation
circuit
Thyristor
controller
Setpoint
Drill drive
control angle α
Udk
ĐK
Ware form
Motor
+
δα
fast fourier transform
FFT
control angle α
Fuzzy
inference
vibration frequency
spectrum
amplitude and
frequency of vibration
Fuzzy compensation control
countersink
Figure 1. Principle diagram of rotation control of CБШ-250T drilling machine.
+ Fuzzy logic block (Nguyen Phung Quang, more knowledgeable and become smarter. That
2004): This function receives amplitude and
frequency signals from the vibration
measurement sensor to determine the
appropriate compensation angle (δα) to reduce
vibration.
In the control system of drilling machine
CБШ-250T, the signal setpoint Uđk is set directly
from the hand controlled by the driver in the
cabin, through the controller to change the
opening angle α. In the proposed system, the
opening angle α will be compensated by the
amount of δα through 2 devices (2 blocks)
including:
is the basis for building and developing an
intelligent tool to predict the hardness and
properties of rock and soil in reality, thereby
evaluating the vibration ability of the machine.
Developing a neural network depends on the
quality and number of samples in the training
process. Drilling process variables such as speed,
force, and torque are important and are selected
as inputs to the neural network. The output
signal is amplitude and frequency of vibration.
+ Table of input and output data for network
training, see table 1 (Le Ngoc Dung and Dang Van
Chi, 2018).
+ Vibration sensor block: in the proposed
modeling, it is replaced by Neural Network (with
the function of recognizing amplitude and
vibration frequency after successful network
training).
+ Fuzzy logic block: This function receives
amplitude and frequency signals from the
vibration measurement sensor to determine the
appropriate compensation angle (δα) to reduce
vibration.
+ Network design and training
The network is built based on the
programming in m-file, including the network
structure: the number of neurons layers, the
number of neurons in the layers, the transfer
function, deviations, etc... Perform the training
process, training results are neural network
diagram and deviation graph as shown in Figure
2, Figure 3 and Figure 4.
The result of checking the input and output
data sets of the 3-layer network model [16 x 36 x
2] shows that the identification data sets are
closely following the sample data sets. The newly
established neural network had learned the set
of input and output signals. The difference
between the real value and the target value
achieved after 652 generations (Epochs) training.
2.2. Building a neural network block to
identify frequency and vibration amplitude
Neural networks are a very useful tool for
identifying and controlling non-linear systems.
The ability to self-study and update knowledge is
an advantage that makes the network more and
Dung Ngoc Le and Chi Dang Van/Journal of Mining and Earth Sciences 61 (6), 90 - 96
Table 1. Data for neural network training.
93
Rock
Spectrum
(FFT)
Speed of
drilling
(vòng/ph)
Drilling
Torque
Mc
Amplitude
STT
hardness
force (F)
fc
13
12
11.5
11
10.5
10
9
8.5
8
7
6.5
6
5
(rad/s)
(Hz)
0.16
0.48
0.8
(m/s2)
0.3
(ton)
30
27.5
25
24
23
20
17
15
13
12
10
9
(Nm)
260
218
185
183
172
165
156
153
134
121
102
91
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1
3
5
50
63
70
75
78
84
90
96
102
107
110
123
132
138
145
150
0.65
0.35
0.15
0.23
0.2
0.75
0.25
0.2
10
15
18
26
31
35
40
55
60
82
100
120
140
1.6
2.4
2.88
4.16
4.96
5.6
6.4
8.8
9.6
13.12
16
0.15
0.1
0.02
0.05
0.03
0.05
0.03
83
82
75
67
8
7
6
5
4.5
4
3
19.2
22.4
2.3. Application of fuzzy-logic to design fuzzy
compensation controller (δα)
+ Defining input and output linguistic
variables:
Input parameters:
1. Frequency f of the vibration signal, using 5
fuzzy sets: from (0.08 - 22.4) Hz.
2. Vibration amplitude, using 5 fuzzy sets:
from (0.003 - 1.14) m/s2.
Figure 2. Reduced structure of 3 layers of the
network.
3. Control angle α, using 5 fuzzy sets: from
(53.2o - 88.2o).
Output: compensating angle δα, using 5 sets
of fuzzy words (-35o →+35o).
The structure diagram for the fuzzy
inference set in the Matlab is shown in Figure 5.
Figure 3. Input layer shortening structure of the
network.
Figure 5. Schematic structure for the fuzzy
inference.
Figure 4. Deviations in neural network training.
94
Dung Ngoc Le and Chi Dang Van/Journal of Mining and Earth Sciences 61 (6), 90 - 96
Alpha_
compensation = 10.5
Frequency=9.58
Amplitude_A=0.76
Alpha_angle=77.9
Figure 6. Test results of fuzzy inference output offset δα.
+ Composition law and defuzzification methods
Non-linear transmission relation of fuzzy
prediction system with three input variables and
one output variable, it is collected according to
expert data, and data table 1 has a total of 125
clauses constituted by law:
If Freqf=Freqfi and Ampl_A=Ampl_Ai and
Alpha=Alphai then Alpha_comp= Alpha_compj
The fuzzy inference set is installed with the
Max-Min component device. The inference is
performed according to the Min law. The fuzzy
inference is performed according to the Max law,
defuzzification average method focus.
After successfully developing two sets of
Neural network and Fuzzy logic tools, it will be
saved in Simulink Matlab's library for research
and modeling. From the proposed principle
diagram (Figure 1), the control system
modeling was implemented, in which the
motor model was developed (Nguyen Chi Tinh
al. 2013), linking blocks together and running.
(Figure 7).
Test results in operating conditions with
different hardness soils, systems with and
without compensation, results achieved with
control quality and vibration reduction
objectives for the device. Figure 8 and Figure 9
show the results of model tests at different
depths. The red is the amplitude and frequency
of vibration with the current controller, blue is
the amplitude and frequency of vibration to
compensate fuzzy controller. Observing the
+ Simulation results in the Matlab as shown
in Figure 6.
3. Applying Neural network and Fuzzy logic
to model the rotary channel control system
on drilling machine CБШ-250T
Figure 7. Simulation of the rotating channel on the CБШ-250T drilling machine.
Dung Ngoc Le and Chi Dang Van/Journal of Mining and Earth Sciences 61 (6), 90 - 96
95
peak amplitudes at different frequencies of 0.1
Hz, 1.5Hz, 3.7 Hz, 5.2 Hz, 7.85 Hz (at a depth of
3 m) and 0.2 Hz, 0.4 Hz, 3.8 Hz, 4.2 Hz all show
the reduced effect vibration level ranges from
20% - 60%.
control rotation speed and reduce vibration,
including:
+ Training a neural network to determine
the properties of rock by amplitude and
vibration frequency.
+ Developing fuzzy logic to determine the
complementary value.
4. Conclusion
+ Summarizing and model the rotation
speed control system to apply the fuzzy
compensation controller, comparing and
evaluating with the current controller in use.
The paper presents research and
development of two tools, neural network and
fuzzy logic to build a fuzzy controller
embedded into the current control system to
Vibration signal
Frequency (Hz)
Vibration reduction controls
Frequency (Hz)
Figure 8. Test results on the model at a depth of 3 m.
Vibration signal
Frequency (Hz)
Vibration reduction controls
Frequency (Hz)
Figure 9. Test results on the model at a depth of 6 m.
96
Dung Ngoc Le and Chi Dang Van/Journal of Mining and Earth Sciences 61 (6), 90 - 96
+ The research results were tested on the Le Ngoc Dung, Dang Van Chi, 2018,
simulation model, evaluated by the control
system's quality criteria, the vibration reduction
criteria on the machine. The results allow the
controller's application to the actual drill
operation.
"Application of Matlab to study and analyze
vibration frequency spectrum for CБШ –
250T drilling machine in the mining
industry", Proceedings of the National
Conference of Science earth and resources
+ The research results confirm that the
application of neural networks and fuzzy logic to
improve the quality of control and reduce
vibration for drilling machines is a reasonable
solution in non-linear power transmission
systems.
+ Proposing to continue evaluating the fuzzy
compensation control system's stability and
sustainability through simultaneous control of
force and rotation speed.
with
sustainable
development,
transportation publisher, Hanoi
Nguyen Chi Tinh et al. 2013, "Modeling the
automatic control system of the rotation
speed of CБШ-250T rotary drilling rig".
Summary report on basic research 2013,
University of Mining and Geology, Hanoi
Nguyen Thac Khanh, 2003, "Research and
improve the diagram of control system of
rotary drilling machine CБШ-250T in open
mines in Vietnam". Master of Engineering
thesis, Hanoi University of Mining and
Geology, Hanoi
References
Alexei A. Zhukovsky, 1982, “Rotary Drilling
Automatic Control system”. United States
Patent.
Ngo Duc Thao, 1971, "Researching and
proposing the automation system of drilling
boreholes for open pit mining", Doctoral
Thesis of Engineering, Moscow Mining
University.
B.Y. Lee, H.S. Liu, Y.S. Tarng, 1998, “Modeling
and optimization of drilling process”,
Journal of Materials Processing Technology,
74 (1998) 149–157.
Nguyen Phung Quang, 2004, “Matlab & Simulik
for automatic control engineers”, Publisher
of Science & Engineering, Hanoi.
Claude E. Aboujaoude, 1991, “Modeling,
Simulation and Control of Rotary Blasthole
Drills”, Masters of Engineering, Department
of Electrical Engineering McGill University,
Montreal.
Stuart Jardine, Dave Malone, Mike Sheppard;
1994. “Putting a Damper on Drilling’s Bad
Vibrations”, Sugar Texas – USA, Oil field
Review
Edward A. Branscombe,
2010,
“Investigation of Vibration Related Signals
for
Monitoring
of Large Open‐
Queen’s University
Thai Duy Thuc, 2001, "Theoretical basis for
automatic electric transmission". Publisher
Transport - Hanoi.
Pit Rotary Electric”
Kingston, Ontario, Canada
Jerome Rajnauth, Tennyson Jagai, 2012,
“Reduce Torsional Vibration and Improve
Drilling Operations “, International Journal
of Applied Science and Technology, Vol. 2
No. 7.
Эkcплуатационная документация; 2003;
ВБІПРЯМИТЕЛБ
ТПЕ-200-460-Y2.1.
(Technical document on a rotary drilling
machine - provided by Cao Son Coal Joint
Stock Company).
Bạn đang xem tài liệu "Application of fuzzy-logic to design fuzzy compensation controller for speed control system to reduce vibration of CBШ-250T drilling machine in mining industry", để tải tài liệu gốc về máy hãy click vào nút Download ở trên
File đính kèm:
- application_of_fuzzy_logic_to_design_fuzzy_compensation_cont.pdf