Parameters identification BLDC motor: Instrumentations and transfer functions
MATEC Web of Conferences
Parameters identification BLDC motor: Instrumentations and transfer functions
Izza Anshory 0 1
Imam Robandi 1
Wirawan Wirawan 1
0 Universitas Muhammadiyah Sidoarjo, Department of Electrical Engineering , 601271 , Indonesia
1 Institut Teknologi Sepuluh Nopember (ITS), Department of Electrical Engineering , Surabaya, 60111 , Indonesia
One of the steps that must be taken in optimizing the speed control of the Brushless DC motor (BLDC) Motor is to mathematically model. The purpose of this research is to get the mathematical modeling of BLDC motor with 350 watts of power in the form of transfer function. The method used to obtain the transfer function is to identify the input and output data. The identification is done by designing and placing current sensors, voltages and motor speeds in the prototype BLDC motor speed control. The measured input data includes current and voltage, while the output data is the engine speed (RPM). By adjusting the pulse width modulation value (PWM), a change in the speed of the BLDC motor is affected. From the process of research in the form of instrumentation on hardware prototype and simulation with application program Matlab System Identification Toolbox (SIT), successfully obtained mathematical modeling in the form of transfer function.
Optimization is a method used to find the best possible
solution of a problem with certain algorithms . One of
the studies that need to be done is the optimization of
motor speed setting of Brushless Direct Current
(BLDCM) as an electric bicycle drive, so that the
optimal stability is obtained, such as steady state error
reduction and fast transient response . To do the
design and optimization, one of the steps that must be
done by the researcher is to mathematically model the
BLDC motor in the form of transfer function.
Information on input and output data from system
measurement results , is needed to help identify
controlled motor systems without risk of damage. There
are two approaches in modeling that is modeling based
on the theory of knowledge and experimental modeling
is done by collecting data such as input and output
signals on the system. Modeling with the experimental
approach is divided into two, namely parametric and
non-parametric approach. The parametric approach is the
approach to find out the structure of the model of a
system through identification using numerical
algorithms. There are four types of model structures used
in parametric approaches, such as Auto Regressive with
external input (ARX), Auto Regressive Moving Avarage
with external input (ARMAX), Output Error (OE), and
Box-Jenkins (BJ) .
System identification is a process done to get a
mathematical model of a system based on input signal
data and output. The relation of input and output data is
then analyzed by Auto Regressive with external input
(ARX) algorithm .
The parametric approach with the ARX model
structure has the advantage that the physical system is
regarded as a black box, so that any kind of component
present in the physical system and any kind of material is
not at issue and need not be noticed. The modeling
process is done through the process of generating test
signals and input data input and output using
microcontroller atmega 8535. Data sent by
microcontroller will be accepted by the software to be
identified, so obtained a mathematical model from the
plant. The plant identified in this research study is a
prototype of BLDC motor rotation speed setting.
BLDC motor is one type of permanent magnet
synchronous motor (PMSM) which has many advantages
compared to direct current (DC) and induction motor,
such as better torque and speed characteristics, high
dynamic response, high efficiency and reliability, low
maintenance costs, small motor size, no friction in the
commutator and the most important of the advantages of
BLDC motor is the need for speed and position
adjustment , the ability to control speed with the
change of input voltage .
The purpose of this research is to identify the
parameters of BLDC motor, by placing some sensors on
the BLDC motor system, to get current, voltage, and
speed data. Data measurement results are then analyzed
using Matlab's System Identification Toolbox (SIT), so
that the output obtained in the form of mathematical
models in the form of transfer functions. The
identification process is done by several stages: design
hardware design, input-output data entry, model
structure selection, and identification method. The result
of SIT analysis obtained in the form of transfer function
is used to perform optimization with artificial intelligent
2 Identification method
2.1 Hardware design
The design configuration to identify the BLDC motor
parameters is done by arranging the interface circuit
according to Fig. 2.
3 phase Inverter
Fig. 2. Block Diagram of BLDC Parameter Identification.
Block diagram of BLDC motor system work is speed
is regulated through the amount of PWM value, PWM
value is set by programming through arduino
microcontroller. To determine the switching order of the
inverter driver, arranged through the hall sensor placed
on the BLDC motor, after the hall sensor has the rotor
position information, it will then be used to switch to six
mosfets alternately which will generate AC voltage, so
as to drive the BLDC motor.
2.2 Design testing to get transfer function
The test result data is identified to know the parameters
by using the System Identification toolbox (SIT) tool, as
shown in Fig. 3. SIT is a tool for creating mathematical
models, and is based on the collection of input / output
data from the system.
The mathematical model can be adjusted to the
assumed process model setting, until an accurate
mathematical model is obtained and matches the input
data in the SIT package. Fig .3 shows interface design on
Matlab program. On the left side of the interface there is
space to enter import data that needs to be inserted in the
3 Result and discussion
3.1 Hardware experiment
The test procedure is performed to obtain input and
output data on BLDC motor speed control system.
3.2 System identification toolbox
Test results data that is the current, voltage and motor
rotation speed is incorporated into the application system
identification toolbox (SIT) contained in the Matlab
program. SIT application is used to obtain transfer
function based on input and output data as in Fig. 4.
The model that describes the measured relationship
between input and output signals. The output results are
partially determined by the input. In most cases, the
output is also affected by the interference signal. Shown
the relationship between each input and output.
The next step is to perform the identification transfer
function by using data estimation based on time domain
motor data. Data have an explanation is output data 1, 1
and 207 samples. The number of poles is 2, and the
number of zeros is 0. Results of processing with System
Identification Toolbox can be seen in Fig. 7.
All these signals are a function of time, and the input
value at time t. Often, in the context of identification,
only discrete time points are considered, since the
measuring equipment typically records the signal only at
discrete-time time, spreading in time with the sampling
time unit interval T. The modeling problem then
illustrates how these three signals relate to each other.
Result of research to identify parameters system BLDC
motor, it is necessary to approach by designing
hardware, testing and analysis with System Identification
Toolbox. The hardware design is used to obtain accurate
data about inputs and outputs. Compared with
conventional methods, the identification process can be
done more quickly and accurately. To carry out
identification with the SIT tool, the correct transfer
function type must be selected according to the required
response. The accuracy of the calculation of the transfer
function needs to be compared with the input data
provided to the Matlab application. In addition to the
transfer function, Matlab can provide graphical results
for input, output and impulse response.
S. Mirjalili , Knowledge-Based Systems Mothflame optimization algorithm : A novel natureinspired heuristic paradigm, Knowledge-Based Syst ., vol. 89 , pp. 228 - 249 , ( 2015 ).
A. A. A. El-Gammal and A. A. El-Samahy , A modified design of PID controller for DC motor drives using Particle Swarm Optimization PSO , 2009 Int. Conf. Power Eng. Energy Electr. Drives , pp. 419 - 424 , ( 2009 ).
Jamaaluddin;Imam Robandi, Short Term Load Forecasting New Year Celebration Using Interval Type-2 Fuzzy Inference System (Case Study: Java - Bali Electrical System) , in GCEAS 2016 , pp. 1 - 13 . ( 2016 ).
W. Araydah , T. A. Tutunji , and I. Al-naimi, System Identification for a Liquid Flow Process, no . Imc , ( 2017 ).
M. Fruk , G. Vujisić, and T. Špoljarić , Parameter Identification of Transfer Functions Using MATLAB , pp. 571 - 576 .
Electr. …, vol. 3 , no. 9 , pp. 1209 - 1220 , ( 2003 ).
I. Anshory , I. Robandi , and Wirawan, Monitoring and optimization of speed settings for Brushless Direct Current (BLDC) using Particle Swarm Optimization (PSO) , in Proceedings - 2016 IEEE Region 10 Symposium , TENSYMP 2016 , ( 2016 ).
A. M. Ahmed and R. Base , Brushless DC Motor Speed Control using both PI Controller and Fuzzy PI Controller , vol. 109 , no. 10 , pp. 29 - 35 , ( 2015 ).
Chakraborti , Novel reverse regeneration technique of BLDC motor for capacitor charging , in 2014 International Conference on Control, Instrumentation, Energy and Communication (CIEC) , pp. 246 - 253 . ( 2014 )