23

CHAPTER 2

MATHEMATICAL MODELING OF CONVERTERS

2.1 INTRODUCTION

Average –value models are commonly used in the design and

analysis of power electronic based systems. Numerous averaging
methodologies have been developed that are applicable to specific classes of
power electronic circuits, including state-space averaging, canonical cell
averaging, switching average, sampled-data modeling and multi-rate
averaging. Historically, the development of each averaging technique and the
extensions thereof, were motivated by a specific converter topology for which
previous averaging techniques were inaccurate or cumbersome. In general,
when a new converter topology is encountered, it is typically not clear as to
which averaging technique or extension should be used for that converter.
Consequently, substantial analytical effort is generally required to develop
and implement the corresponding average-value model.

State-Space Averaging (SSA) is one of the essential techniques for

analyzing switched mode power conversion circuits (Agarwal 2003). Whether
a computer is used to run simulations or do hand calculations, SSA allows the
user to extend standard DC and AC circuit analysis techniques to switching
circuits. Using SSA models can also run transient analyses much faster. The
first part of this chapter discusses the mathematical modeling of the buck
converter and the second part discusses the mathematical modeling of the
boost converter using state space averaging techniques.
 24

2.2 BUCK CONVERTER

The step down DC-DC converter, commonly known as a buck

converter is shown in Figure 2.1. As the switch is turned on to mode 1, the
diode is reverse biased and the current flows through the inductor, into the
voltage sink. After a time t, the switch is turned off in mode 2. The inductor
current then freewheels through the diode in mode 2. The second mode is
terminated at the inverse function of the switching period.

Figure 2.1 Circuit Diagram of the Buck Converter

Assuming state variables,

x1 = iL (current through the inductor)
x2 = Vc (voltage across capacitor)
D = duty ratio at the operating point (in steady state)
Vo = DC output voltage at the operating point
^
d = small perturbation in the duty ratio
^
v0 = small perturbation in the output voltage
Vs = input DC voltage source
 25

The state variables are,

* di L
x1 =i L  x1 = (2.1)
dt
* dVc
x 2 =Vc  x 2 = (2.2)
dt
i 
x L (2.3)
Vc 

2.2.1 Mode 1 Buck Converter when the Switch is Turned On

Figure 2.2 shows the mode 1 operation of the buck converter. In

mode 1, the switch is turned on for the duration DTs and the diode D is in the
reverse biased condition. The current flows from the supply to the load.

Figure 2.2 Circuit Diagram of the Buck Converter with Switch Turned On

At mode 1, by applying Kirchhoff’s law

di L
Vs  L  Vo (2.4)
dt
io  i c  i L Vo  Vc (2.5)
ic  CdVc / dt (2.6)
 26

*  x 2 Vs
x1   (2.7)
L L
dVc Vo
iL  C  (2.8)
dt R
* x1 x2
x2   (2.9)
C RC

The state equations are,

*
x  Ax  BVs (2.10)
Vo  Cx (2.11)
thereby,
 0 1/ L 
A1    (2.12)
1/ C 1/ RC 
1/ L 
B1    (2.13)
 0 
C1  0 1 (2.14)

2.2.2 Mode 2 Buck Converter when the Switch is Turned Off

Figure 2.3 shows the mode 2 operation of the buck converter. In

mode 2, the switch is turned off for the duration (1-D)Ts and the diode D is in
the forward biased condition, the load current continues to flow through the
freewheeling diode D, and the energy stored in the inductor is transferred to
the capacitor which appears across the load.
 27

Figure 2.3 Circuit Diagram of the Buck Converter with Switch Turned Off

At mode 2, by applying Kirchhoff’s laws,

di L
L  Vc  0 (2.15)
dt
io  i c  i L Vo  Vc (2.16)
io  Vo / R (2.17)
and
ic  CdVo / dt (2.18)
*
x2
x1   (2.19)
L
dVc Vo
i L  C.  (2.20)
dt R
* x2 x2
x2   (2.21)
C RC

The state equations are,

*
x  Ax  BVs (2.22)
Vo  Cx (2.23)
 28

There by,
 0 1/ L 
A2    (2.24)
1/ C 1/ RC 
0
B2    (2.25)
0
C 2   0 1 (2.26)

A1 and A2 are state matrices, B1 and B2 are excitation vectors.

The state-space averaging is an approximate technique, that can be

used to describe the input and output relations of a switching converter,
having different modes of operation. Although the original system is linear for
any given switching condition, the resulting system generally is non-linear.
Therefore, a small signal approximation has to be employed, to obtain the
linearised small signal behavior, before other techniques can be applied.

*
x  [A1 D  A 2 (1  D)]x  [B1D  B2 (1  D)]Vs (2.27)
V0  [C1D  C2 (1  D)]x (2.28)

The steady state values of the state variables, the sources and the
duty cycle ratio are assumed as,
x = X ; v0 = V0 ; d = D;
AX + BVs = 0 (2.29)
V0 = CX (2.30)
where,
A  A1D  A 2 (1  D) (2.31)

 0 1/ L 
A  (2.32)
1/ C 1/ RC 
B  B1D  B2 (1  D) (2.33)
 29

D / L 
B  (2.34)
 0 
C  C1D  C 2 (1  D) (2.35)

C  0 1 (2.36)

The steady state equation is,

AX  BVs  0 (2.37)

X  A 1 BVs (2.38)

The state equation during perturbation is,

*
* ^ ^ ^ ^ ^ ^
x  x  [A1 (D  d)  A 2 (1  D  d)])X  x)  B1 (D  d)  B2 (1  D  d)VS
(2.39)
^ ^ ^ ^
V0  v 0  [C1 (D  d)  C 2 (1  D  d)](X  x) (2.40)
simplifying,
*
x  Ax  [(A1  A 2 )X  (B1  B2 )VS ]d (2.41)
v o  Cx  [(C1  C2 )X]d (2.42)

Transforming to s-domain,
sx(s)  Ax(s)  [(A1  A 2 )X  (B1  B2 )Vs ]d  s  (2.43)

v0 (s)  Cx(s)  [(C1  C2 )X]d(s) (2.44)

Applying laplace transform with zero initial condition and using the
superposition theorem, the small signal duty cycle d to output v0, the transfer
function can be obtained considering zero line perturbation.
 30

To obtain the transfer function,

we have
v 0 (s)
 C[sI  A]1[(A1  A 2 )X  (B1  B2 )Vs ]  (C1  C 2 )X (2.45)
d(s)

0 0
A1  A 2    (2.46)
0 0
1/ L 
B1  B2    (2.47)
 0 
C1  C 2  0 0 (2.48)

The transfer function describing the output to input of the buck

converter is given by
v o (s) Vs / LC
 2 (2.49)
d(s) s  (1/ RC)s  1/ LC
where D = duty ratio at the operating point, Vs = supply voltage, L = filter
inductor, C = filter (output) capacitor, vo =output voltage and R = load resistor.

It can be seen from equation 2.49, that for the buck converter the
control-to-output transfer function is dependent on the operating point and its
validity is limited to in and around the operating point. As the operating
region of the buck converter is wide, the conventional way of designing the
controllers involves selecting the worst case operating point, i.e. under the
minimum line and maximum load conditions. The transfer function of the
buck converter under the worst case conditions is taken as the base in the
design of the controller.

2.3 BOOST CONVERTER

The step up DC-DC converter, commonly known as a boost

converter is shown in Figure 2.4. In mode 1, the switch is on, the current
 31

through the inductor increases and the energy stored in the inductor builds up.
In mode 2 the switch is off, current through the inductor continues to flow via
the diode D, the RC network and back to the source. The inductor releases its
energy and the polarity of inductor voltage is such that its terminal connected
to the diode is positive with respect to its other terminal connected to the
source.

Figure 2.4 Circuit Diagram of the Boost Converter

Assuming state variables,

x1 = iL (current through the inductor)
x2 = Vc (voltage across capacitor)
D = duty ratio at the operating point (in steady state)
Vo = dc output voltage at the operating point
^
d = small perturbation in the duty ratio
^
v0 = small perturbation in the output voltage
Vs = input dc voltage source

The state variables are,

* Vs * di
x1  x1  i L  x1  L (2.50)
L dt
* dVc
x 2  Vc  x 2  (2.51)
dt
 32

i 
x L (2.52)
Vc 

2.3.1 Mode 1 Boost Converter when the Switch is Turned On

Figure 2.5 shows the mode 1 operation of the boost converter. In

this mode, the switch is on for the duration DTs and the diode D is reverse
biased and the current through the inductor increases, so that the energy
stored in the inductor builds up.

Figure 2.5 Boost Converter with the Switch Turned On

At mode 1, by applying Kirchhoff’s law,

di L
Vs  L (2.53)
dt
Vo  Vc
io  Vo / R , io  i c  0 (2.54)
ic  CdVc / dt (2.55)
* Vs
x1  (2.56)
L
 33

dVc Vo
C  0 (2.57)
dt R
*
x2
x2   (2.58)
RC

The state equations are,

*
x  Ax  BVs (2.59)
Vo  Cx (2.60)

Thereby,
0 0 
A1    (2.61)
0 1/ RC 
1/ L 
B1    (2.62)
 0 
C1  0 1 (2.63)

2.3.2 Mode 2 Boost Converter when the Switch is Turned Off

Figure 2.6 shows the mode 2 operation of the boost converter. In

this mode, the switch is off for the duration (1-D)Ts and the diode D is
forward biased. Current through the inductor continues to flow via the diode
D, the RC network and back to the source. The inductor releases its energy
and the polarity of inductor voltage is such that its terminal connected to the
diode is positive with respect to its other terminal connected to the source.
 34

Figure 2.6 Boost Converter with the Switch Turned Off

At mode 2, by applying Kirchhoffs law,

di L
Vs  L  Vo (2.64)
dt
io  i c  0 , V0  Vc (2.65)
ic  CdVc / dt (2.66)
*  x 2 Vs
x1   (2.67)
L L
dVc V0
iL  C  (2.68)
dt R
*
x1 x 2
x2   (2.69)
C RC
 0 1/ L 
A 2   (2.70)
1/ C 1/ RC 
1/ L 
B 2   (2.71)
 0 
C 2  0 1 (2.72)

Averaging the state equations of two circuit modes

*
x  [A1 D  A 2 (1  D)]x  [B1D  B2 (1  D)]Vs (2.73)
Vo  [C1D  C 2 (1  D)]x (2.74)
 35

A  A1D  A 2 (1  D) (2.75)

 0 (1  D) / L 
A (2.76)
(1  D) / C 1/ RC 
B  B1D  B2 (1  D) (2.77)

1/ L 
B  (2.78)
 0 
C  C1D  C 2 (1  D) (2.79)

C  0 1 (2.80)

The steady state equation is

AX  BVs  0 (2.81)

X  A 1 BVs (2.82)

Vs LC  1/ RLC 
X (1  D) / LC  (2.83)
(1  D)2  

The state equation during perturbation is,

*
* ^ ^ ^ ^ ^ ^
x  x  [A1 (D  d)  A 2 (1  D  d)]  (X  x)  [B1 (D  d)  B2 (1  D  d)]Vs
(2.84)
^ ^ ^ ^
V0  v 0  [C1 (D  d)  C 2 (1  D  d)]  (X  x) (2.85)

Simplifying,
*
x  Ax  [(A1  A 2 )X  (B1  B2 )Vs ]d (2.86)
v 0  Cx  [(C1  C 2 )X]d (2.87)

Transforming to s-domain,
sx(s)  Ax(s)  [(A1  A 2 )X  (B1  B2 )Vs ]d(s) (2.88)
v o (s)  Cx(s)  [(C1  C 2 )X]d(s) (2.89)
 36

Applying laplace transform with zero initial condition and using the
superposition theorem, the small signal duty cycle d to outputV0 the transfer
function can be obtained considering zero line perturbation.
v o (s)
 C[sI  A]1[(A1  A 2 )X  (B1  B2 )Vs ]  C1  C2 )X (2.90)
d(s)

 0 1/ L 
A1  A 2    (2.91)
 1/ C 0 
0
B1  B2    (2.92)
0
C1  C 2  0 0 (2.93)

The transfer function describing the output to input of the boost

converter is given by
 L 
1 s 2
v 0 (s) Vs  R(1  D) 
 (2.94)
d(s) (1  D) 2  L 2 LC 
1  s R(1  D) 2  s (1  D)2 
 
where D = duty ratio at the operating point, Vs = supply voltage, L = filter
inductor, C = filter (output) capacitor, vo =output voltage and R = load
resistor.

It can be seen from equation 2.94, that for the boost converter, the
control-to-output transfer function is dependent on the operating point and its
validity is limited to in and around the operating point. As the operating
region of the boost converter is wide, the conventional way of designing the
controllers involves selecting the worst case operating point, i.e. under the
minimum line and maximum load conditions. The transfer function of the
boost converter under the worse case conditions is taken as the base in the
design of the controller.
 37

2.4 CONCLUSION

In this chapter the mathematical modeling of buck and boost

converters is derived and it is used in simulation studies in the coming
chapters. The state space averaging technique, is a very powerful technique
for the modern day design and analysis of power electronics-based systems.