International Journal of Power Electronics and Drive Systems (IJPEDS)

Vol. 15, No. 3, September 2024, pp. 1319~1330

ISSN: 2088-8694, DOI: 10.11591/ijpeds.v15.i3.pp1319-1330  1319

Bidirectional power flow in an electric vehicle using predictive

control algorithm including sneak circuit analysis

Sri Latha Eti1, Kasibhatla Rama Sudha1, Molleti Venkata Pankaj Lahari2, Akkapeddi Chandra Sekhar3
1
Department of Electrical Engineering, Andhra University, Visakhapatnam, India
2
Department of Electrical and Electronics Engineering, GVP College of Engineering, Visakhapatnam, India
3
Department of Mathematics, GITAM University, Visakhapatnam, India

Article Info ABSTRACT

Article history: The use of electric vehicle battery chargers is becoming more common for
transferring power between grid and energy storage system. This paper
Received Sep 14, 2023 focuses on providing a comprehensive understanding of various modes of
Revised Jan 27, 2024 operation of these battery chargers. Two different controllers are used; one is
Accepted Feb 17, 2024 the predictive power controller at the grid end to generate active and reactive
power references to operate a 3-phase AC/DC converter connected between
the power grid and the DC link and the other is a hysteresis current controller
Keywords: used to operate a reversible DC/DC converter connected between the DC link
and electric vehicle battery and is considered as the main component of the
3-Φ bidirectional AC/DC energy storage system. Also, sneak circuit analysis is carried out on the
converter DC/DC converter and the controller is designed accordingly. Results from
Hysteresis current controller simulations validate the suggested control scheme's viability and efficacy. The
Predictive control algorithm adopted topology is validated in real time with dSpace 1104 hardware in the
Reversible DC/DC converter loop prototype operating in different scenarios, both in steady-state and during
Sneak path analysis transients.
Space vector modulated This is an open access article under the CC BY-SA license.
converter

Corresponding Author:
Kasibhatla Rama Sudha
Department of Electrical Engineering, Andhra University
Visakhapatnam, Andhra Pradesh, India
Email: prof.kramasudha@andhrauniversity.edu.in

1. INTRODUCTION
The term "bidirectional power transfer" refers to the mutual exchange of active power between the
electric vehicle charger and the power grid. Bidirectional AC-DC converter acts as a crucial component in the
energy storage system as it is able to provide a bidirectional power exchange operation and synchronization
with the grid. It tackles with power quality issues caused by high harmonic distortions (THD), low power factor
issues, voltage alterations, DC voltage and current ripples [1]. Therefore, the AC-DC converter control
algorithm must be effective and stable.
To obtain an improved efficiency and usefulness of the bidirectional AC-DC converter, several studies
have been carried out. The mostly preferred technique is a linear voltage-control technique (VCT) which
employs linear proportional integral derivative (PID) controllers and a space vector modulation scheme (SVM)
to control the power flow indirectly using the active component and reactive component of grid currents [2].
Using coordinate transformation, the observed grid AC currents are converted into active currents and reactive
currents, which are described in the α-β coordinates [3], [4]. The SVM technique generates a switching
sequence for insulated gate bipolar transistor (IGBTs) in every sampling period so that the average output
voltage of the 3-Φ AC-DC converter is proportional to the signals obtained from the controllers making it
operate with a fixed switching frequency at a pre-defined sampling time [5].

Journal homepage: http://ijpeds.iaescore.com

1320  ISSN: 2088-8694

The bidirectional battery charger's entire electrical diagram is displayed in Figure 1. The
charging/discharging currents and the active/reactive power supplied by the power grid and electric vehicles,
respectively, should be tracked by the bidirectional chargers. Taking the nominal battery voltage of 300 V into
consideration, a 3-Φ AC/DC bidirectional converter circuit, is chosen at the rectification stage with a 100 V
three phase supply voltage [6]. It is connected to the AC supply mains in series with ‘line resistance (R a)’ and
‘line inductance (La)’. A capacitor (Cdc) is connected at the output terminals of the converter to provide a
smooth DC output voltage. Reversible DC-DC converter is connected in the second stage, to which an output
inductance (Ldc) and the electric vehicle battery are connected [7]–[10].

Figure 1. Grid connected bi-directional converter and reversible DC - DC converter circuit

In this paper, the description of the circuits and system operation of the 3-Φ bidirectional AC-DC
rectifier and reversible DC-DC converter are explained in section 2. The sneak circuit analysis is carried out
and the proper switching control sequence for the DC-DC converter is obtained in section 3, followed by
section 4, where the controllers for the AC-DC rectifier and the DC-DC converter is explained. Simulation
results were presented in section 5 and a detailed explanation is presented for the results obtained. Conclusion
of the paper is provided in section 6 followed by bibliography.

2. DESCRIPTION OF CIRCUIT AND SYSTEM OPERATION

2.1. 3-Φ AC–DC converter
The 3-Φ bidirectional AC-DC converter topology shown in Figure 2 illustrates the power transfer
between the grid and the DC bus. Six IGBT switches (S1–S6) form the bidirectional AC–DC converter is
connected to the power grid using line inductance (L a) in series with line resistance (Ra). In order to stabilize
the DC bus voltage (Vdc), a capacitor (Cdc) is placed at the DC link to filter ripples in the output DC
voltage [11]–[13]. The bidirectional AC-DC converter has two operating modes. In the first operating mode,
which is also called as the rectifier mode, it enables power transfer from 3-Φ source to the DC link. In the
second operating mode, also called as the inverter mode, the power flows from the DC link to the 3-Φ voltage
terminals.

Figure 2. 3-Φ bidirectional AC–DC converter circuit

Int J Pow Elec & Dri Syst, Vol. 15, No. 3, September 2024: 1319-1330
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1321

Using the electrical system depicted in Figure 2, the power circuit for the 3-Φ bidirectional AC-DC
converter transforms electrical energy from AC and DC form. Two switches are connected in each leg of the
converter, which must operate in complimentary mode to prevent short circuits. As a result, the switching states
of the 3-Φ bidirectional AC-DC converter can be obtained from the three gating signals namely, S a, Sb, and Sc
and can be defined as (1)-(4) [14]. The switching signals are defined as (1).
1; S1 /S3 /S5 − ON
SK = { (1)
0; S2 /S4 /S6 − OFF

Thus, the switching vector (SK) can be expressed as (2) [15].

1 1
1 − − Sa
Sα 2 2 2
[S ] = [ ] [Sb ] (2)
β 3 √3 √3
0 − Sc
2 2

Here, α-β coordinates are used for simplification of the analysis of 3-Φ circuits. The 3-Φ voltages (Va, Vb, Vc)
are expressed in the α-β coordinates similar to (2) as (3).
1 1
1 − − Va
Vα 2 2 2
[V ] = [ ] [Vb ] (3)
β 3 √3 √3
0 − Vc
2 2

In terms of switching states, the voltage vector can be expressed as (4).

Vα Sα
[V ] = [S ] × Vdc (4)
β β

Where, Vdc - the DC output voltage obtained from the converter.

The six valid switching states are presented in the Table 1. The switching states 0/0/0 and 1/1/1 are
avoided for safety as the three phases are never turned on at the same time. The average power output in all the
six switching states is 2 Vdc/3. The power flow equations from grid side to output in α-β coordinates are as (5).
digαβ
L = Vgαβ − Rigαβ − Vαβ (5)
dt

Where, Vgαβ and igαβ indicate the grid side voltage and current. The derivative of current obtained from the
Euler's approximation in expanded form is written as (6).
digαβ igαβ (k+1)−igαβ (k)
= (6)
dt Ts

On substitution into (5) and simplifying as (7).

igαβ (k+1)−igαβ (k) 1
= [Vgαβ (k) − Rigαβ (k) − Vαβ (k)] (7)
Ts L

Ts
igαβ (k + 1) = [Vgαβ (k) − Rigαβ (k) − Vαβ (k)] + igαβ (8)
L

The 3-Φ grid currents in the standard α-β coordinates for the next time instant shown in (8). Thus, the
expressions for the active power and the reactive power for the time instant (k+1) are calculated as (9).
3 ∗
P(k + 1) = Re[Vgαβ (k + 1) ∗ igαβ (k + 1)] (9)
2

3
P(k + 1) = [Vgα (k + 1) ∗ igα (k + 1) + Vgβ (k + 1) ∗ igβ (k + 1)] (10)
2

Here in (11) and (12), Vgαβ (k + 1) represents the predicted values of grid voltages for the time instant (k+1)
and are approximated to the measured values Vgαβ (k).

Vgα (k + 1) ≅ Vgα (k) ; Vgβ (k + 1) ≅ Vgβ (k) (11)

3
P(k + 1) = [Vgα (k) igα (k + 1) + Vgβ (k) igβ (k + 1)] (12)
2
Bidirectional power flow in an electric vehicle using predictive control algorithm including … (Sri Latha Eti)
1322  ISSN: 2088-8694

In the same way, the reactive power equation can be expressed as (13) and (14).
3 ∗
Q(k + 1) = Img[Vgαβ (k + 1) igαβ (k + 1)] (13)
2

3
Q(k + 1) = [Vgβ (k) igα (k + 1) − Vgα (k) igβ (k + 1)] (14)
2

According to the needs of the grid and the consumers, switching pulses to the converter can be obtained using
the predicted values of the active and reactive power.

Table 1. Switching states of bidirectional AC–DC converter in α - β coordinates

Switching states in 3-Φ coordinates Switching states in α-β coordinates Voltages in α-β coordinates
Sa Sb Sc Sα Sβ Vα Vβ
0 0 1 -1/3 -1/√3 -𝑉𝑑𝑐 /3 -𝑉𝑑𝑐 /√3
0 1 0 -1/3 1/√3 -𝑉𝑑𝑐 /3 𝑉𝑑𝑐 /√3
0 1 1 -2/3 0 -2𝑉𝑑𝑐 /3 0
1 0 0 2/3 0 2𝑉𝑑𝑐 /3 0
1 0 1 1/3 -1/√3 𝑉𝑑𝑐 /3 -𝑉𝑑𝑐 /√3
1 1 0 1/3 1/√3 𝑉𝑑𝑐 /3 𝑉𝑑𝑐 /√3

2.2. Reversible DC–DC converter

Basically, DC-DC converters are useful for converting voltages to higher or lower levels depending
on the circuit operation and requirement. Reversible DC-DC converter in this case are particularly used to
charge the electric vehicle batteries [16]. The electric vehicle battery charger is assumed to be charging lithium
ion batteries with a nominal voltage of 300 V. The DC-DC converter regulates the battery current (ibat) and
voltage of the battery cells at their nominal ranges.
The voltage and current relations of the DC-DC converter and battery cells can be determined from
Figure 3 as (15).
dibat
Vdc = Ldc + Vbat (15)
dt

Where, Vdc is the DC link voltage; Vbat, and ibat are the voltage of the battery and the current passing through
the inductor Ldc, respectively. On simplification, (15) can be rewritten as (16).
dibat 1
= [Vdc − Vbat ] (16)
dt Ldc

From Euler’s approximation, the battery current can be represented for the time instant (k+1) as (17).
ibat (k+1)−ibat (k) 1
= [Vdc − Vbat ] (17)
Ts Ldc

The (18) will be useful in controlling the operation of reversible DC-DC converter by comparison of the
upcoming current value with the reference current value. It also regulates the voltage (V bat) required for
charging the batteries.
Ts
ibat (k + 1) = [Vdc − Vbat ] + ibat (k) (18)
Ldc

Figure 3. Reversible DC–DC converter circuit

Int J Pow Elec & Dri Syst, Vol. 15, No. 3, September 2024: 1319-1330
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1323

3. SNEAK CIRCUIT ANALYSIS OF DC-DC CONVERTER

In general, a power converter has several operating stages and can be switched using the control
technique. Because every operating stage corresponds to a sub circuit, a current flow through an unexpected
path causes undesired results in certain conditions called sneak current paths [17]. Therefore, sneak circuit
analysis is a must in power converters for eliminating these unexpected paths by proper control strategy or
changes in topology. To find the sneak current paths in the DC–DC converter shown in Figure 4(a), a directed
graph is established using graph theory. All the branches of the converter are called as edges;
the node representing the intersection of two or more branches is called as vertex; the component connected in
the branch is the name of edge; the current flow direction through the component in a branch is the edge
direction [18], [19]. Thus, the directed graph of the DC-DC converter obtained from the above method is shown
in Figure 4(b).
From the definition of the adjacency matrix [17], for the DC–DC converter of the network graph in
Figure 4(b) can be obtained as (19).

0 1 0 1
𝐴 = [1 0 1 1] (19)
0 1 0 1
1 1 1 0

The connection matrix elements can be defined as (20).

1; for i = j
cij = {0; for i ≠ j and no cuurent path from i to j (20)
Symbol of the element; for i ≠ j

As a result, the generalized DC-DC converter connection matrix can be written as (21).

1 𝐺1 0 𝑉𝑑𝑐
𝐺1 1 𝐿𝑑𝑐 𝐺2
𝑐= (21)
0 𝐿𝑑𝑐 1 𝐶𝑏 + 𝑅
𝑉𝑑𝑐 𝐺2 𝐶𝑏 + 𝑅 1

Since, the current always flows in a loop in the circuit, it is possible to identify every potential current
loop in the converter by computing the determinant of matrix C [18], [19].

|DetC| = 1 − Cb2 − R2 − 2RCb − L2dc + Ldc Cb G2 + Ldc G2 R − G22 + G12 − G12 Cb2 − G12 R2
−2G1 Cb R + G1 Vdc Ldc Cb + G1 Vdc Ldc R − G1 G2 Vdc − G1 Vdc Ldc Cb − Vdc G1 Ldc R
2 2 2
+Cdc G1 G2 − Cdc + Cdc Ldc

From the above equation, loops having both switches G1 and G2 can be avoided as both switches do not urn
on simultaneously. Quadratic terms are also avoided as they don’t make any sense [20]. Thus, from the equation
(22), it is clear that the circuit consists of 5 valid current paths. Out of these five paths, the desired current paths
are as in (23).

|DetC| = Cb R + Ldc G2 Cb + Ldc G2 R + Vdc G1 Ldc Cb + Vdc G1 Ldc R (22)

Vdc − G1 − Ldc – R
R − Ldc − G2 (23)

To find the operating stages of the circuit, the possible sub - circuits for each current path are identified
using Mesh Analysis. From the circuit shown in Figure 4(a), the three meshes are represented as (24).

𝑀1 = {𝑉𝑑𝑐 , 𝐺1 , 𝐺2 }
𝑀2 = {𝐺2 , 𝐿𝑑𝑐 , 𝐶𝑏 }
𝑀3 = {𝐶𝑏 , 𝑅} (24)

On application of and (∪) operation and ring sum (⊕) operation of the graph theory, the three possible mesh
circuits can be identified as (25).

Bidirectional power flow in an electric vehicle using predictive control algorithm including … (Sri Latha Eti)
1324  ISSN: 2088-8694

𝑀2 ∪ 𝑀3 = {𝐺2 , 𝐿𝑑𝑐 , 𝑅}
𝑀2 ⊕ 𝑀3 = {𝐺2 , 𝐿𝑑𝑐 , 𝐶𝑏 , 𝑅}
𝑀1 ∪ 𝑀2 ⊕ 𝑀3 = {𝑉𝑑𝑐 , 𝐺1 , 𝐺2 , 𝐶𝑏 , 𝑅}
𝑀3 = {𝐶𝑏 , 𝑅} (25)

It can be observed from the above (25) that 𝑀2 ⊕ 𝑀3 , 𝑀1 ∪ 𝑀2 ,⊕ 𝑀3 , 𝑎𝑛𝑑 𝑀3 represent the three
operating stages. Therefore, the possible paths for current flow in the converter, meshes and the operating
stages are obtained. Now, the behavior of the converter during the presence of sneak circuit paths while
operating need to be determined by carrying out sneak circuit analysis. In general, the switching components
G1 and G2 are turned ‘ON’ and turned ‘OFF’ in a pre-determined order according to the control strategy.
Therefore, the operating modes of the DC-DC converter will consist of the sub-circuits corresponding to
various switching states [21].
Let ‘1’ be ON state of the switches G1 and G2 and ‘0’ indicate their OFF state. Then, the possible
switching states that are valid can be expressed as (26).

𝐺1 𝐺2 = 1 0
𝐺1 𝐺2 = 0 1
𝐺1 𝐺2 = 0 0 (26)

The switching state 𝐺1 𝐺2 =1 1 is invalid as both the switches are always operated as conjugates. Thus, the
three valid switching states constitute two switching sequences:
- Control sequence I comprise of switching sequence, as in (27).

𝐺1 𝐺2 =1 0 ; 0 1 ; 1 0 ; 0 1 ; 1 0 ; 0 1; (27)

- Control sequence II comprises of switching sequence, as in (28).

𝐺1 𝐺2 =1 0 ; 0 1 ; 0 0 ; 1 0 ; 0 1 ; 0 0 ; 1 0 ; 0 1; 0 0 ; (28)

(a) (b)

Figure 4. Representing a DC-DC converter circuit (a) with nodes and (b) as a directed network graph

The equivalent circuits corresponding to each switching state as indicated by (26) are depicted by the
Figures 5(a)-5(d), respectively. Thus, the operating modes under control sequence I represented by (27) are
shown in above Figures 5(a) and 5(b) and are considered as normal operating modes. The operating modes
under control sequence II represented by (28) have two different paths for current and are assumed to be the
sneak operating modes and can be observed in Figure 6 and Figure 7, respectively.
Due to these sneak operating paths, the number of operating stages increases which leads to energy
imbalance and the desired performance characteristics of the converter gets disturbed. To eliminate these sneak
operating modes, any of the following mentioned ways can be implemented: i) Proper selection of control
sequence; ii) Proper parameter design; and iii) Topology modification. In this paper, proper selection of control
sequence method is followed. Thus, the control sequence I is chosen to avoid sneak circuit paths occurring
during the operation of DC-DC converter.

Int J Pow Elec & Dri Syst, Vol. 15, No. 3, September 2024: 1319-1330
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1325

(a) (b)

(c) (d)

Figure 5. Equivalent circuit of the DC- DC converter for: (a) switching state G1G2 = 1 0, (b) switching
state G1G2 = 0 1, (c) switching state G1G2 = 0 0, and (d) switching state G1G2 = 0 0

Figure 6. Sneak operating path I

Figure 7. Sneak operating path II

4. CONTROLLER STRUCTURE AND DESIGN CONSIDERATIONS

4.1. Predictive control technique for bidirectional AC-DC converter
In general, the conventional predictive control technique is a current control technique, in which, the
output current continuously tracks its reference value by using the discrete behavior of the converter. Figure 8
shows the predictive control technique for the 3-Φ bidirectional AC–DC converter. In this circuit, the grid
current and grid voltage are transformed into the standard α-β coordinates using (3). The six switching pulses,
obtained from Table 1 are used to calculate the reference voltage values [22]. The predictive grid current in α-
β coordinates for (k+1) instant is obtained from (8) and the instantaneously changing values of active and
reactive power for time instant (k+1) are calculated using (12) and (14).
The command of the reference active power (P ref) is determined from the reference battery current
signal using (29).

Pref = Vbat × i∗batref (29)

The reference reactive power command Qref is a presumed signal which implies the different stages of power
flow in the system. Depending on changes in magnitude and sign convention of instantaneous active power
Bidirectional power flow in an electric vehicle using predictive control algorithm including … (Sri Latha Eti)
1326  ISSN: 2088-8694

and reactive power values, the most suitable switching state of the converter for complete control of active
power and reactive power during each and every sector is chosen [23], [24]. The error signals obtained from
(30) which gives the minimal error and the instant which provides the minimal error is chosen to be the next
switching instant to the bidirectional rectifier.

∆𝑆 = √(𝑃(𝑘 + 1) − 𝑃𝑟𝑒𝑓 )2 + (𝑄(𝑘 + 1) − 𝑄𝑟𝑒𝑓 )2 (30)

4.2. Hysteresis current control

The control circuitry for reversible DC–DC converter is represented in Figure 9. in which the
predictive battery current is calculated using (18). The gating pulses to IGBT 1 and IGBT 2 are generated from
the hysteresis current controller taking the battery reference current and comparing with the current obtained
for the (k+1) instant. When the battery current for the (k+1) instant is more than the battery reference current,
the error will be more than the upper hysteresis limit and IGBT G1 is triggered. In that case, the DC–DC
Converter works as a buck converter and the battery of the electric vehicle is in charging mode. When the
battery current for the (k+1) instant is less than the battery reference current, the error will be less than the
lower hysteresis limit and IGBT G2 is triggered. In this case, the DC–DC converter works as a boost converter
and the battery of the electric vehicle is in discharging mode and supplies the grid [25].

Figure 8. Predictive control technique for 3-Φ bidirectional AC–DC converter

Figure 9. Control circuit for reversible DC-DC waveform

5. SIMULATION ANALYSIS
Simulation results are obtained from MATLAB/Simulink R2019a and are presented in this section.
Also, the results obtained from dSpace DS1104 control board connected to host computer with
MATLAB/Simulink installed in it and hardware in the loop interface were presented in the below section which
provides the real time validation of the system under consideration. The specifications of the system
represented in Figure 1 are provided in the Table 2.

Table 2. Model parameters chosen in simulation

Parameter Value
Per phase line resistance (Ra) in Ohms 0.25
Per phase line inductance (La) in Henry 10×10-3
DC capacitor (Cdc) in Farads 550×10-6
DC inductor at output terminals (Ldc) in Henry 100×10-3
Battery capacitor (Cbat) in Farads 750×10-6
Line to line voltage at grid (Vg) in volts 100
Supply frequency in Hertz 50
Battery voltage (Vbat) in Volts 300
Sampling frequency (fs) in Hertz 20×103

Int J Pow Elec & Dri Syst, Vol. 15, No. 3, September 2024: 1319-1330
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1327

The reference battery charging current (ibatref) is initially maintained at 6.675 A, the battery voltage is
maintained at 299.2 V which can be observed in Figures 10(a) and 10(b) respectively. During this period, the
active power reference (Pref) is maintained at 2 KW and the reactive power reference (Qref) is maintained at
0 KVar which can be observed in Figures 11(a) and 11(b) respectively. At t= 1 sec, the battery reference current
is stepped down to -6.675 A, which declines the active power reference value to -2 KW. The battery reference
current is made 0 at t=2 sec, which makes the active power reference to trace 0 KW at t=2 sec. The battery
current reference is again increased to 6.675 A at t=4 sec, which raises the active power reference command to
2 KW.

(a)

(b)

Figure 10. Output waveforms of (a) battery voltage and (b) battery current

The reactive power reference command also is raised to +1 KVar at this instant. The battery reference
current is reduced to -3.33 A at t=5 sec, there is a voltage drop by a small value at this instant. The active power
command falls to -1KW and reactive power command is maintained at +1 KVar during this period from t= 5
to 6 sec. The battery current reference is raised to 6.675 A at t=6 sec, when the reference reactive power
command is reduced to -1KVar. Again, at t=7 sec, the battery current reference command is reduced to -3.33
A, which reduces the active power command to -1 KW and battery voltage drops by a small value.
For simplicity, the simulation results of the per phase voltage in ‘a’ phase of the grid and the three-
phase currents are presented at different intervals separately. The observations from the simulation results
shown in Figures 12(a)-12(g) can be tabulated during different intervals of time as in Table 3, and inferences
can be drawn according to the observations.
With MATLAB/Simulink installed in a personal computer serving as a host and a dSpace 1104 control
board, the below results have been obtained as in Figure 13. This technique is applied over sample duration of
5ms. The results are obtained for two modes of operation of the system model and the three-phase voltage and
current waveform are observed. From the Figure 13, it is observed that the voltage and current waveform are
in-phase until t=2 sec, which implies that the battery is charging from supply. After t=2 sec, the three-phase
voltages are lagging behind the currents indicating that the battery is supplying power to the supply mains.

(a) (b)

Figure 11. Reference command signals: (a) active power and (b) reactive power

Bidirectional power flow in an electric vehicle using predictive control algorithm including … (Sri Latha Eti)
1328  ISSN: 2088-8694

(a) (b)

(c) (d)

(e) (f)

(g)

Figure 12. Per phase Voltage Vga and 3-Φ Currents Igabc*10 of power grid: (a) during time interval 0<t<1,
(b) during time interval 1<t<2, (c) during time interval 2<t<3, (d) during time interval 3<t<4, (e) during
time interval 4<t<5, (f) during time interval 5<t<6, and (g) during time interval 6<t<7

Table 3. Observations from grid voltage and current waveforms

Time interval Grid voltage Vga (V) & grid Observations
in sec currents Igabc (A)*10
0<t<1 Vga & Iga are in phase The EV battery charges from grid supply
1<t<2 Vga & Iga are in phase opposition The EV supplies active power to the grid
2<t<3 Iga is leading Vga by 900 The converter provides reactive power compensation to the grid.
3<t<4 Iga is lagging Vga by 900 The grid supplies reactive power to load
4<t<5 Iga is leading Vga The power grid supplies load with both positive active and
reactive power.
5<t<6 Iga is lagging Vga The power grid receives active power from the battery.
6<t<7 Iga is lagging Vga The system works as Static VAR compensator and battery is
charged from grid
t>7 Iga is lagging Vga The system works as Static VAR compensator and battery
delivers active power to grid

Int J Pow Elec & Dri Syst, Vol. 15, No. 3, September 2024: 1319-1330
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1329

(a) (b)

Figure 13. Results obtained from dSpace DS 1104 control board using hardware in the loop mode:
(a) three phase voltages and (b) three phase currents

6. CONCLUSION
The proposed predictive control algorithm in this paper aims to transfer power between the grid and
the DC link in energy storage systems. The controller receives active power commands from the customer and
reactive power commands from the utility, adjusting the line current and battery charging or discharging current
accordingly to meet the demands. Also, the battery of the electric vehicle remains unaffected by the reactive
power operation. The simulation results demonstrate fast dynamic response and excellent steady-state
performance. To validate the system in real-time, the dSpace 1104 control board is connected to a standalone
personal computer with MATLAB installed, operating in hardware in the loop mode.

REFERENCES
[1] B. Devi and V. S. Kumar, “Lithium-ion battery management system: a review,” in 2022 International Conference on Power, Energy,
Control and Transmission Systems (ICPECTS), Dec. 2022, pp. 1–6. doi: 10.1109/ICPECTS56089.2022.10047494.
[2] S. Shete, P. Jog, R. K. Kumawat, and D. K. Palwalia, “Battery management system for SOC estimation of lithium-ion battery in
electric vehicle: a review,” in 2021 6th IEEE International Conference on Recent Advances and Innovations in Engineering
(ICRAIE), Dec. 2021, pp. 1–4. doi: 10.1109/ICRAIE52900.2021.9703752.
[3] T. He, D. D.-C. Lu, M. Wu, Q. Yang, T. Li, and Q. Liu, “Four-quadrant operations of bidirectional chargers for electric vehicles in
smart car parks: G2V, V2G, and V4G,” Energies, vol. 14, no. 1, p. 181, Dec. 2020, doi: 10.3390/en14010181.
[4] S. Kouro, P. Cortes, R. Vargas, U. Ammann, and J. Rodriguez, “Model predictive control—a simple and powerful method to control power
converters,” IEEE Transactions on Industrial Electronics, vol. 56, no. 6, pp. 1826–1838, Jun. 2009, doi: 10.1109/TIE.2008.2008349.
[5] V. Monteiro, J. G. Pinto, and J. L. Afonso, “Operation modes for the electric vehicle in smart grids and smart homes: present and proposed
modes,” IEEE Transactions on Vehicular Technology, vol. 65, no. 3, pp. 1007–1020, Mar. 2016, doi: 10.1109/TVT.2015.2481005.
[6] Y. Zhang, W. Xie, Z. Li, and Y. Zhang, “Model predictive direct power control of a PWM rectifier with duty cycle optimization,”
IEEE Transactions on Power Electronics, vol. 28, no. 11, pp. 5343–5351, Nov. 2013, doi: 10.1109/TPEL.2013.2243846.
[7] A. Bouafia, J.-P. Gaubert, and F. Krim, “Analysis and design of new switching table for direct power control of three-phase PWM
rectifier,” in 2008 13th International Power Electronics and Motion Control Conference, Sep. 2008, pp. 703–709. doi:
10.1109/EPEPEMC.2008.4635347.
[8] A. Calle-Prado, S. Alepuz, J. Bordonau, J. Nicolas-Apruzzese, P. Cortes, and J. Rodriguez, “Model predictive current control of
grid-connected neutral-point-clamped converters to meet low-voltage ride-through requirements,” IEEE Transactions on Industrial
Electronics, vol. 62, no. 3, pp. 1503–1514, Mar. 2015, doi: 10.1109/TIE.2014.2364459.
[9] A. K. Ali and R. G. Omar, “Finite control set model predictive direct current control strategy with constraints applying to drive
three-phase induction motor,” International Journal of Electrical and Computer Engineering (IJECE), vol. 11, no. 4, pp. 2916–
2924, Aug. 2021, doi: 10.11591/ijece.v11i4.pp2916-2924.
[10] L. E. Jumaa Alkurawy and K. G. Mohammed, “Model predictive control of magnetic levitation system,” International Journal of
Electrical and Computer Engineering (IJECE), vol. 10, no. 6, pp. 5802–5812, Dec. 2020, doi: 10.11591/ijece.v10i6.pp5802-5812.
[11] S. K. Dube, R. Nair, and P. Das, “Analysis and design of an integrated bidirectional three-phase AC–DC resonant converter,” IEEE
Transactions on Industrial Electronics, vol. 70, no. 5, pp. 4369–4379, May 2023, doi: 10.1109/TIE.2022.3183274.
[12] T. Noguchi, H. Tomiki, S. Kondo, and I. Takahashi, “Direct power control of PWM converter without power-source voltage
sensors,” IEEE Transactions on Industry Applications, vol. 34, no. 3, pp. 473–479, 1998, doi: 10.1109/28.673716.
[13] M. Kesler, M. C. Kisacikoglu, and L. M. Tolbert, “Vehicle-to-grid reactive power operation using plug-in electric vehicle
bidirectional offboard charger,” IEEE Transactions on Industrial Electronics, vol. 61, no. 12, pp. 6778–6784, Dec. 2014, doi:
10.1109/TIE.2014.2314065.
[14] L. Pan and C. Zhang, “Model predictive control of a single-phase PWM rectifier for electric vehicle charger,” Energy Procedia,
vol. 105, pp. 4027–4033, May 2017, doi: 10.1016/j.egypro.2017.03.850.
[15] J. G. Pinto et al., “Bidirectional battery charger with grid-to-vehicle, vehicle-to-grid and vehicle-to-home technologies,” in IECON 2013
- 39th Annual Conference of the IEEE Industrial Electronics Society, Nov. 2013, pp. 5934–5939. doi: 10.1109/IECON.2013.6700108.
[16] H. Tao, G. Zhang, and Z. Zheng, “Onboard charging DC/DC converter of electric vehicle based on synchronous rectification and
characteristic analysis,” Journal of Advanced Transportation, vol. 2019, pp. 1–10, Jun. 2019, doi: 10.1155/2019/2613893.

Bidirectional power flow in an electric vehicle using predictive control algorithm including … (Sri Latha Eti)
1330  ISSN: 2088-8694

[17] M. P. Kazmierkowski, M. Jasinski, and G. Wrona, “DSP-based control of grid-connected power converters operating under grid
distortions,” IEEE Transactions on Industrial Informatics, vol. 7, no. 2, pp. 204–211, May 2011, doi: 10.1109/TII.2011.2134856.
[18] B. Zhang and D. Qiu, “Sneak circuits in power converters: concept, principle and application,” CPSS Transactions on Power
Electronics and Applications, vol. 2, no. 1, pp. 68–75, Apr. 2017, doi: 10.24295/CPSSTPEA.2017.00008.
[19] W. K. Chen and C. Satyanarayana, “Applied graph theory: graphs and electrical networks,” IEEE Transactions on Systems, Man,
and Cybernetics, vol. 8, no. 5, pp. 418–418, 1978, doi: 10.1109/TSMC.1978.4309987.
[20] D. C. Osborn, “Sneak paths in X-Y matrix arrays,” IEEE Journal of Solid-State Circuits, vol. 4, no. 6, pp. 312–317, Dec. 1969, doi:
10.1109/JSSC.1969.1050029.
[21] S.-Z. Chen, Y. Wang, G. Zhang, L. Chang, and Y. Zhang, “Sneak circuit theory based approach to avoiding short-circuit paths in
reconfigurable battery systems,” IEEE Transactions on Industrial Electronics, vol. 68, no. 12, pp. 12353–12363, Dec. 2021, doi:
10.1109/TIE.2020.3044805.
[22] J. Li, D. Qiu, and B. Zhang, “Sneak circuit analysis for n-stage resonant switched capacitor converters based on graph theory,” in IECON
2007 - 33rd Annual Conference of the IEEE Industrial Electronics Society, 2007, pp. 1581–1585. doi: 10.1109/IECON.2007.4460006.
[23] F. Senani, A. Rahab, F. Louar, F. Bourourou, and H. Benalla, “Active and reactive power control of DFIG using PI and DPC controllers,”
in 2015 4th International Conference on Electrical Engineering (ICEE), Dec. 2015, pp. 1–6. doi: 10.1109/INTEE.2015.7416841.
[24] Z. Tao and L. Li, “Control loop design and bidirectional control strategy of a bidirectional DC/DC converter,” in IECON 2017 -
43rd Annual Conference of the IEEE Industrial Electronics Society, Oct. 2017, pp. 5720–5725. doi: 10.1109/IECON.2017.8216992.
[25] W. Yu, H. Qian, and J.-S. Lai, “Design of high-efficiency bidirectional DC–DC converter and high-precision efficiency measurement,”
IEEE Transactions on Power Electronics, vol. 25, no. 3, pp. 650–658, Mar. 2010, doi: 10.1109/TPEL.2009.2034265.

BIOGRAPHIES OF AUTHORS

Sri Latha Eti received B.Tech degree from JNTU Hyderabad, India in 2008 and
M.E. degree in power electronics and drives from Andhra University, India in 2015. She is
currently pursuing Ph.D. from Andhra University, Visakhapatnam, India. She works as an
assistant professor in WISTM Engineering College, Visakhapatnam, India. She has 11 years
teaching experience and has published eight papers in international journals. Her areas of
interest include power electronic drives, electric vehicle battery charger design, and sneak
circuits analysis of DC-DC converters. She can be contacted at email: etisrilatha@gmail.com.

Kasibhatla Rama Sudha received B.E degree from GITAM University, India in
1991 and M.E degree in power systems from Andhra University, India in 1994. She received
her Ph.D. from Andhra University in 2004. She is currently working as professor in Andhra
University. She has over 30 years of teaching experience and has published over 86 articles in
prestigious international journals. Her areas of interest include artificial intelligence
applications to power systems, digital circuits & devices, embedded systems applications,
automation, and renewable energy sources. She can be contacted at email:
prof.kramasudha@andhrauniversity.edu.in.

Molleti Venkata Pankaj Lahari received B.Tech. in 2013 and M.Tech. in 2015
in electrical & electronics engineering and power electronics & drives respectively, from GVP
College of Engineering (Autonomous), Visakhapatnam, India. She received her Ph.D. from
Andhra University, India in 2022. She has a teaching experience of six years and research
experience of seven years. Her areas of interest are power electronics, electrical drives, real-
time modelling & simulation, and embedded systems applications in electrical engineering. She
can be contacted at email: laharimolleti14@gmail.com.

Akkapeddi Chandra Sekhar has completed his Ph.D. from JNTU Hyderabad,
India in 2008. He is currently working as a professor at Department of Mathematics in GITAM
University, Visakhapatnam, India and has over 30 years of teaching experience. He has two
patents registered by his name and has published 18 papers in international journals. His areas
of interest include algebraic structures in cryptography, mathematical analysis of dynamic
systems, and analysis of differential equations. He can be contacted at email:
cakkaped@gitam.edu.

Int J Pow Elec & Dri Syst, Vol. 15, No. 3, September 2024: 1319-1330