Three Phase Motor Flux Optimization
Three Phase Motor Flux Optimization
ENVIRONMENT
AZOJETE June 2024. Vol. 20(2):435-452
Published by the Faculty of Engineering, University of Maiduguri, Maiduguri, Nigeria.
Print ISSN: 1596-2490, Electronic ISSN: 2545-5818
www.azojete.com.ng
1.0 Introduction
Induction motors are the most widely used electric machine in the industry because of their
ruggedness, cheapness and very low maintenance cost. Studies have shown that bulk of the
energy used in the industry is consumed, largely, by induction motors (Engel et al.,2015).
Therefore, efforts have been made over the years by various manufacturers of induction
motors in the technique of construction of Energy Efficient Motors (EEM). This has resulted in
significant improvements in efficiency and increase in energy savings. Induction motor have high
efficiency at rated speed and torque, but the efficiency of induction motor decreases greatly at
light loads because of an imbalance in copper and core losses. Energy saving can however, be
achieved by proper selection of the flux level in the motor (Eseosa and Christian, 2018). Losses
in the induction motor are classified into stator copper losses, rotor copper losses, core losses,
stray losses, friction and windage losses. The copper and core losses account for the largest
amount of losses in induction motor and therefore explain the interest to use them in the
analysis proposed in this paper. In order to increase the motor efficiency, the flux must be
reduced and balance obtained between copper and core losses. Basically, there are two
approaches to improving the efficiency of induction motor (Faduyile, 2009). These are the Loss-
based model approach and the Power measure-based approach. In the loss-based model
approach, the model-based controller computes losses by using the machine model and thereby
determines a flux level that minimizes the losses. This approach is fast and does not produce
torque ripple. In the power measure-based approach, the controller searches for the operating
points where the output power is constant (Rashtchi and Bizhani, 2015).
In this paper, Loss-based model approach or Loss model controller of induction motor drive
system that determines the optimal flux for loss minimization was analysed and implemented
in MATLAB/Simulink. The significance of the method applied in this paper is the reduction in
energy consumed by the induction motor for the same load torque over the period of motor
operation. The benefit is that it will go a long way to reduce operating cost and to make spare
energy available to meet other energy demands. In the long term, this benefit should also slow
the world’s overall energy consumption and protect the environment from pollution and
preserve the world natural resources from further depletion.
Induction motors are large consumers of the electric energy, and are designed to operate
efficiently at rated loads. Unfortunately, these motors operate on partial or low loads most of
the time and this situation leads to inefficient motor operation with attendant losses in energy
and increased operating cost. The proposed work seeks to reduce the inefficiency of Induction
motor operation under the conditions of partial or low load operation by developing a method
to maximize the efficiency of the motor under these conditions.
Several papers relating to loss minimization in Induction motors have been published in
recent times. Rashtchi and Bizhani (2015) presented a new online loss minimization algorithm
using particle swam optimization (PSO) to determine the optimum flux level for the efficiency
optimization of vector-controlled induction motor. Khan et al. (2015) developed a high
performance and robust control structure called the flux optimized direct torque control for
induction motor. Lee et al. (2013) presented a loss based model controller using a digital signal
processor (TM320LF2812), for Induction motor efficiency maximization and energy saving.
Sruthi et al. (2017) formulated a loss minimization algorithm for energy saving and efficiency
improvement in induction motor drive. The input voltage and frequency were optimized to
reduce total losses. Eseosa & Christian (2018), Afify et al. (2018), Chelliah (2009), Nam and
Uddin (2006) and Laroui, et al. (2021) presented various approaches in optimization control
techniques to improve the efficiency operation of induction motor. Aygun and Aktas (2018)
implemented a DTC technique combined with an online loss minimization model-based
controller to minimize the losses of an induction motor applied in Electric Vehicle (EV).
The aim of this paper is to develop a rotor flux optimization control model for an existing
induction motor and simulate its performances in MATLAB/Simulink for loss minimization and
efficiency improvement.
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
Where and , and are the direct axes and quadrature axes stator and rotor
voltages, and , , and are stator and rotor d-q currents, , , and are the
d-q stator and rotor flux respectively. and are stator and rotor resistances respectively.
is stator frequency and is the rotor electrical speed.
The equivalent circuit obtained from the mathematical model equations (1) - (8) of the
Induction motor is shown in Figure 1
2.2 Rotor flux-oriented control scheme for the three-phase induction motor
The scheme is derived from the rotor dynamic equations of the three-phase induction machine
in the synchronously rotating reference frame represented by equations (3)-(4) and (7)-(8).
The q-axis component of the rotor flux space vector is equated to zero because of the
alignment of the d-axis with the rotor flux vector (Olarinoye et al., 2022). It is mathematically
represented as follows;
(11)
The d-axis component of the rotor flux space vector is now equal to the rotor flux
magnitude . The rotor flux magnitude is expressed as follows;
(12)
Mathematically, applying equations (11) to equations (12), (3)-(4), (7) and (8) yields the
following results;
(13)
(14)
(15)
(16)
(17)
Equations (13) and (14) have been obtained considering that the short-circuited squirrel cage
rotor of the three-phase induction motor have zero voltages in the and axis of the rotor
windings
i.e. and .
The rotor flux dynamics is developed from equations (15) and (16) to give the following;
(18)
where (19)
Equation (18) is a first order system governed by rotor time constant . The Slip frequency
is obtained from equations (14) and (17) as follows;
(20)
The rotor flux angle , which is also the angle of transformation because of alignment of the
-axis to the rotor flux vector, is estimated by integrating the rotor flux angular frequency as
follows
(21)
The torque produced in the three-phase induction motor given by equation (9) is modified,
upon application of equations (11) and (13), to the following;
(22)
As a result of the alignment of the rotor flux magnitude with the d-axis, equations (18)-(21)
enables the estimation of rotor flux magnitude and the phase angle or the angle of
transformation.
Figure 2: Three Phase Induction Motor q-d axis Equivalent Circuits accounting for Core
Loss Resistance
The following voltage, flux linkage and current equations are derived from the equivalent
circuits of Figures 2 (a) and (b) and equations (1)-(4) to give the following equations;
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
Where (31)
and are the currents flowing through the core loss branches of the q and d axes
respectively. and are the torque currents responsible for producing electromagnetic
torque in the motor as seen from the equivalent circuit of Figure 1.
By applying equations (11) and (13) to the voltage equations (23) – (26), relationships between
the applied stator currents and torque currents were obtained as follows;
(32)
(33)
The power loss model was formulated considering that the total electrical power loss is the
sum of the copper and core losses in the motor. The expression of electrical power loss is
obtained from the equivalent circuits as follows;
(34)
A power loss model was derived by substituting current equations (27) – (33) into equation
(34). It is given as follows;
(35)
The criterion for minimum loss is obtained by differentiating equation (35) with respect to the
rotor flux magnitude as follows;
(36)
Equation (36) was solved to obtain the optimum value of rotor flux for minimum loss and hence
maximum efficiency. The optimum value of the rotor flux for a given load torque and
speed was derived to yield;
(37)
The efficiency of the induction motor was determined using the following equation;
Efficiency (38)
Equation (38) gives the efficiency as a function of electrical power loss in the machine.
For the purposes of controller design, the stator voltages are expressed as a function of the d
and q axis currents and the rotor flux by applying equations (5) - (8), (11) and (13) to equations
(23) and (24). The following expressions were obtained;
(39)
(40)
Where:
(41)
(42)
Equations (39) and (40) are first order non-linear systems for which and are inputs and
and are outputs. The voltage commands needed to give the desired currents, and
, are obtained with a feedback-based solution using current controllers. The formulation
for the controller gains was performed using equations (39) and (40) and illustrated in Figures
3 (a) and (b).
(46)
The expressions for the PI controller gains and were obtained by comparing the
normalized denominator of equation (46) with the standard second order denominator, as
follows;
(47)
(50)
(51)
(52)
The control block diagram of the flux control system is obtained from equation (18) and given
in Figure 5.
The flux controller gains are calculated from equation (53) as follows;
(54)
(55)
The supply voltages in the synchronous reference frame are then calculated based on the
following equations;
(56)
(57)
The control objective is to regulate the rotor flux and the torque to their commanded values.
Figure 6 shows the block diagram of the rotor flux-oriented optimizing drive scheme for the
three-phase induction machine. The inverter supplies the controlled three phase voltages to
the three-phase induction motor in such a way that the flux is regulated at its optimum value
for each load torque. The torque command is the output of the speed control loop. When the
motor drive is in operation, flux is produced in the machine. The flux is controlled to its
optimum value by regulating the d-axis stator current. This regulation ensures linear control of
torque by the q-axis stator current. The flux model receives stator currents , and speed
and delivers the flux supply frequency and the corresponding angular position
to ensure proper vector alignment. The block diagram of Figure 6 was implemented in
MATLAB/ Simulink.
+
Eqns.( +
56) & Transfo
rm- Inverte
(57) + r
ation
+ blk.
Eqn. Transfo
(18) rm-
ation
blk.
I
M
Eqn. + Eqn.
(21) (20)
+
The computed controller gains are as provided in Table 2. The loss minimization algorithm was
set to trigger at 2.5s. With the reference speed and load torque set to 250 rad/s and 3Nm
respectively, the rated flux was computed as 1Wb. The result of total electrical power loss in
the induction motor is presented in Figure 8.
300
250
200
Total Power Loss (W)
150
50
Proposed scheme
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
It can be seen that the conventional rotor flux control scheme produces a steady state power
loss of 200W in the motor. The proposed scheme on the other hand, upon activation at a time
of 2.5 s, produces a power loss of 108.04W in the same motor. This gives a significant reduction
of 54.02%, the consequence of which is a rise in motor efficiency from 78.9% to 87.41% as
shown in Figure 9.
90
80
70
Proposed
60 Conventional rotor flux scheme
control scheme
Efficiency (%)
50
40
30
20
10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (S)
Figure 10 shows the plot of the rotor flux magnitude. It can be seen that the flux value is 1Wb
until the proposed scheme is activated and the flux reduces significantly to its optimal value of
0.6357Wb for the given load torque. This verifies the analysis set forth, that there is an optimal
flux in the motor for which electrical losses are minimized and hence efficiency is maximized
for a given load torque.
1.4
1.2
1
rotor flux (Wb)
0.8
Conventional rotor
0.6
flux control scheme
0.4
Proposed scheme
0.2
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Figure 11 shows the speed profile of the induction motor and it can be seen that the motor
was set to accelerate and reach a steady state speed of 250 rad/s.
300
250
200
Speed (rad/s)
150
100
50
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Figure 12 shows the torque produced in the motor during the simulation time. The steady
state torque can be seen to equal 3Nm.
6
4
Torque (Nm)
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Table 3: Comparison of result between conventional, proposed rotor flux control scheme
and the conventional and optimized DTC control schemes
Variable Conventional Proposed method Conventional (Optimized
rotor flux DTC DTC)
control scheme
Total Power loss 200 108.04 231.1961 166.2806
(W)
Efficiency (%) 78.9 87.41 77.1 82.4
Flux (Wb) 1 0.6357 1 0.7539
The rotor flux-oriented control scheme enables the variation of motor speed. To this end, the
motor speed was ramped down from 250 rad/s to 150 rad/s and allowed to stay at that speed
between 4 and 5s of simulation time. Figures 13-15 show the performances of the motor under
the proposed scheme at variable speeds.
300
250
200
Speed (rad/s)
150
100
50
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
300
250
200
Total Power Loss (W)
150
Proposed scheme
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Figure 14: Total electric power loss under under variable speed.
The efficiency of the motor is plotted in Figure 15. It can be seen to decrease to a value of
84.8% after the motor speed is dropped to 150 rad/s because the power output drops with
speed. A table of comparison between these results and those obtained in Aygun & Aktas
(2018) is provided in Table 4.
90
80
70
Proposed scheme
60
Conventional rotor
Efficiency (%)
50
flux control scheme
40
30
20
10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Table 4: Comparison of results of the proposed rotor flux-oriented control scheme with
those of the optimized DTC at variable speed
Scheme Total power loss (W) Efficiency (%)
Proposed scheme at 250 rad/s 108.04 87.41
Proposed scheme at 150 rad/s 83.645 84.8
Optimized DTC at 250 rad/s 166.28 82.4
Optimized DTC at 150 rad/s 101 79.1
Table 4 shows the superior performance of the proposed method at variable speed when
compared with the performance of the optimized DTC method under the same operating
conditions. At a lower speed of 150 rad/s, power loss reduces by 17.2% while efficiency is
higher by 5.7% for a load torque of 3Nm under the proposed scheme as compared to the
optimized DTC scheme developed in Aygun & Aktas (2018).
Figures 16-20 show the effectiveness of the proposed optimization scheme at varying loads.
The load torque on the motor is reduced from 3Nm to 2Nm as shown in Figure 16.
Figure 16: Torque Produced in the motor with load torque varying from 3Nm to 2Nm
In Figure 17, the total power loss is observed to decrease to 72.03 W after the initiation of the
loss minimization algorithm because the flux in the motor is optimized for each load torque.
The optimal flux for the 2Nm load is 0.5188 Wb as seen in Figure 18. The total loss reduces
after the reduction in load from 3Nm to 2Nm because of the reduced output power.
300
250
200
Total Power Loss (W)
150
Conventional rotor
100
flux control scheme
50
Proposed scheme
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Figure 17: Total power loss with load torques varying from 3Nm to 2Nm
Conventional rotor
flux control scheme
Proposed scheme
Figure 18: Optimal flux of induction motor with load torque varying from 3Nm to 2Nm
Although the motor operates at a lower load torque, the operating efficiency is observed to
be the same as that for the 3Nm load torque. The value is still 87.41% as seen from Figure 19.
90
80
70
60
Proposed scheme
Efficiency (%)
50
20
10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Figure 19: Efficiency of the motor with load torque varying from 3Nm to 2Nm
This implies that the efficiency is maximized for all loads under the scheme. These results follow
the same trend as the results provided in Aygun and Aktas (2018). The advantage of the scheme
proposed in this paper is that the point of maximum efficiency is higher by 5.01%. Table 5 gives
a comparison between the results obtained in this work and those obtained in Aygun and Aktas
(2018) when the load torque varied from 3Nm to 2Nm.
The results suggest that the point of maximum efficiency is independent of the load torque but
the optimum flux at which it occurs depends on the load torque. The voltages supplied to the
stator windings of the three-phase induction motor are shown in Figure 20.
300
200
100
Phase Voltages (V)
-100
-200
-300
-400
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)
Figure 20: Supply voltages with load torque varying from 3Nm to 2Nm
It can be seen that the magnitude of the stator voltages reduced with load on the machine. This
is an energy saving ability of the proposed rotor flux optimization control strategy. The
reduction in voltage is a direct result of the reduction in flux to its optimal levels for each load.
The magnitude of the stator voltages rises with speed during the time of acceleration of the
induction motor. It reaches a value of 311V at steady state while the load torque of 3Nm is
applied. The value reduces to 180V after the optimizing control algorithm is activated at 2.5s.
The value further reduces to 150V after the load torque changes from 3Nm to 2Nm.
4. Conclusion
A rotor flux optimization control model for a three-phase induction motor was analysed and
simulated. A comparison between the conventional and proposed rotor flux-oriented control
drive shows that, for a reference speed of 250rad/s and a load torque of 3Nm, the rotor flux
reduced from a rated value of 1Wb to an optimal value of 0.6357Wb which invariably led to a
significant reduction of the total power loss, in the motor, from 200W to 108.04W and a
consequential increase in efficiency from 78.9% to 87.41%. The performance of the scheme at
varying load also shows a change in optimum flux from 0.6357Wb to 0.5188Wb after the load
torque was reduced from 3Nm to 2Nm. Although the motor operated at a lower load torque,
the flux optimization algorithm implemented in MATLAB\Simulink, for the proposed scheme,
maintained a maximum efficiency of 87.41%. It was observed that while the point of maximum
efficiency is independent of the load torque, the optimum flux at which it occurs depends on
the load torque. With the proposed drive scheme operating at variable speed mode in which
motor speed was ramped down from 250rad/s to 150rad/s at a load torque of 3Nm, power
loss was observed to further reduce significantly by 22.6%. These results suggest that the
proposed rotor flux optimization scheme is superior to the optimized DTC scheme proposed
in Aygun & Aktas (2018). The proposed scheme also presents an opportunity to save power
and therefore energy as the command voltages supplied to the induction motor was observed
to decrease with decreasing load applied to the motor.
References
Aygun, H. and Aktas, M. 2018. A novel DTC method with efficiency improvement of IM for EV
applications. Engineering, Technology & Applied Science Research, 8(5): 3456-3462.
Chelliah, TR., Yadav, JG., Srivastava, SP. and Agarwal, P. 2009. Optimal energy control of
induction motor by hybridization of loss model controller based on Particle Swarm
Optimization and search controller. World Congress on Nature & Biologically Inspired
Computing, Coimbatore, India, pp. 1178-1183, Doi: 10.1109/NABIC.2009.5393784.
Engel, E., Kovalev, IV. and Karandeev, D. 2015. Energy-saving IOP Conference Series: Materials
Science and Engineering, 94: 10. 1088/1757-899X/94/1/012008.
Eseosa, O. and Christian, A. 2018. Energy efficiency optimization of three phase induction
motor drives for industrial applications. International Journal of Engineering and Applied
Sciences, 5(8): 42-49.
Faduyile, OE. 2009. Effect of harmonics on the efficiency of a three-phase energy efficient and
standard motors. MSc. Thesis. University of Tennessee at Chattanooga. USA.
Kalhoodashti, HE. and Shahbazian, M. 2011. Hybrid speed control of induction motor using PI
and fuzzy controller. International Journal of Computer Applications, 30(11): 44-50.
Khan, H., Hussain, S. and Bazaz, MA. 2015. Direct torque control of induction motor drive
with flux optimization. In: International Conference on Advances in Computing,
Communications and Informatics (ICACCI), IEEE, 618-623.
Laroui, M., Nour, B., Moungla, H., Cherif, MA., Afifi, H. and Guizani, M. 2021. Edge and fog
computing for IoT: A survey on current research activities & future directions. Computer
Communications, 180(1): 210-231.
Lee, PM., Lee, DD., Nguyen, TV. and Nguyen, PH. 2013. Real-time loss minimization control in
induction machines based on DSP TMS320LF2812. Science and Technology Development
Journal, 16(4): 5-18.
Nam, SW. and Uddin, MN. 2006. Model-based loss minimization control of an induction motor
drive. In: IEEE International Symposium on Industrial Electronics, IEEE, 2367-2372.
Olarinoye, GA., Akorede, MF. and Akinropo, CD. 2022. Design and Simulation of a Three-
Phase Induction Motor Speed Control System. FUOYE Journal of Engineering and Technology,
7(1): 39 – 46.
Rashtchi, V. and Bizhani, H. 2015. Using of particle swarm optimization for loss minimization
of vector-controlled induction motors. Universal Journal of Electrical and Electronic
Engineering, 3(3): 71-80.
Sruthi, MP., Nagamani, C. and Ilango, GS. 2017. An improved algorithm for direct computation
of optimal voltage and frequency for induction motors. Engineering Science and Technology,
an International Journal, 20(5): 1439-1449.