7.4.

State Space Averaging

• A formal method for deriving the small-signal ac equations of a

switching converter
• Equivalent to the modeling method of the previous sections
• Uses the state-space matrix description of linear circuits
• Often cited in the literature
• A general approach: if the state equations of the converter can be
written for each subinterval, then the small-signal averaged model
can always be derived
• Computer programs exist which utilize the state-space averaging
method

Fundamentals of Power Electronics 63 Chapter 7: AC equivalent circuit modeling

 7.4.1. The state equations of a network

• A canonical form for writing the differential equations of a system

• If the system is linear, then the derivatives of the state variables are
expressed as linear combinations of the system independent inputs and
state variables themselves
• The physical state variables of a system are usually associated with the
storage of energy
• For a typical converter circuit, the physical state variables are the inductor
currents and capacitor voltages
• Other typical physical state variables: position and velocity of a motor shaft
• At a given point in time, the values of the state variables depend on the
previous history of the system, rather than the present values of the
system inputs
• To solve the differential equations of a system, the initial values of the
state variables must be specified
Fundamentals of Power Electronics 64 Chapter 7: AC equivalent circuit modeling
 State equations of a linear system, in matrix form

dx(t)
A canonical matrix form: K = A x(t) + B u(t)
dt
y(t) = C x(t) + E u(t)
d x1(t)
State vector x(t) contains
x1(t) dt
inductor currents, capacitor dx(t) d x2(t)
voltages, etc.: x(t) = x2(t) , =
dt dt

Input vector u(t) contains independent sources such as vg(t)

Output vector y(t) contains other dependent quantities to be computed, such
as ig(t)
Matrix K contains values of capacitance, inductance, and mutual
inductance, so that K dx/dt is a vector containing capacitor currents and
inductor winding voltages. These quantities are expressed as linear
combinations of the independent inputs and state variables. The matrices A,
B, C, and E contain the constants of proportionality.
Fundamentals of Power Electronics 65 Chapter 7: AC equivalent circuit modeling
 Example

i(t) L
State vector + + v (t) – +
iR1(t) iC1(t) L iC2(t)
R2
v1(t) iin(t) R1 C1 v1(t) C2 v2(t)
x(t) = v2(t) +
i(t) R3 vout(t)
– – –
Matrix K
Input vector Choose output vector as
C1 0 0
K = 0 C2 0 u(t) = iin(t) vout(t)
y(t) =
0 0 L iR1(t)

To write the state equations of this circuit, we must express the inductor
voltages and capacitor currents as linear combinations of the elements of
the x(t) and u( t) vectors.
Fundamentals of Power Electronics 66 Chapter 7: AC equivalent circuit modeling
 Circuit equations

i(t) L
+ + v (t) – +
iR1(t) iC1(t) L iC2(t)
R2
iin(t) R1 C1 v1(t) C2 v2(t)
+
R3 vout(t)
– – –

dv1(t) v (t)
Find iC1 via node equation: iC1(t) = C1 = iin(t) – 1 – i(t)
dt R

dv2(t) v (t)
Find iC2 via node equation: iC2(t) = C2 = i(t) – 2
dt R2 + R3

di(t)
Find vL via loop equation: vL(t) = L = v1(t) – v2(t)
dt

Fundamentals of Power Electronics 67 Chapter 7: AC equivalent circuit modeling

 Equations in matrix form

dv1(t) v (t)
The same equations: iC1(t) = C1 = iin(t) – 1 – i(t)
dt R
dv (t) v (t)
iC2(t) = C2 2 = i(t) – 2
dt R2 + R3
di(t)
vL(t) = L = v1(t) – v2(t)
dt
Express in matrix form:

dv1(t)
– 1 0 –1
C1 0 0 dt R1 v1(t)
dv2(t) 1
0 C2 0 = 0 – 1 1 v2(t) + 0 iin(t)
dt R2 + R3 0
0 0 L di(t) i(t)
1 –1 0
dt

dx(t)
K = A x(t) + B u(t)
dt

Fundamentals of Power Electronics 68 Chapter 7: AC equivalent circuit modeling

 Output (dependent signal) equations

i(t) L
vout(t) + + v (t) – +
y(t) = iR1(t) iC1(t) L iC2(t)
iR1(t) R2
iin(t) R1 C1 v1(t) C2 v2(t)
+
R3 vout(t)
– – –

Express elements of the vector y as linear combinations of elements

of x and u:
R3
vout(t) = v2(t)
R2 + R3

v1(t)
iR1(t) =
R1

Fundamentals of Power Electronics 69 Chapter 7: AC equivalent circuit modeling

 Express in matrix form

R3
The same equations: vout(t) = v2(t)
R2 + R3
v1(t)
iR1(t) =
R1

Express in matrix form:

R3 v1(t)
vout(t) 0 0
= R2 + R3 v2(t) + 0 iin(t)
iR1(t) 1 0
0 0 i(t)
R1

y(t) = C x(t) + E u(t)

Fundamentals of Power Electronics 70 Chapter 7: AC equivalent circuit modeling

 7.4.2. The basic state-space averaged model

Given: a PWM converter, operating in continuous conduction mode,

with two subintervals during each switching period.
During subinterval 1, when the switches are in position 1, the
converter reduces to a linear circuit that can be described by the
following state equations:
dx(t)
K = A 1 x(t) + B 1 u(t)
dt
y(t) = C 1 x(t) + E 1 u(t)

During subinterval 2, when the switches are in position 2, the

converter reduces to another linear circuit, that can be described by
the following state equations:
dx(t)
K = A 2 x(t) + B 2 u(t)
dt
y(t) = C 2 x(t) + E 2 u(t)

Fundamentals of Power Electronics 71 Chapter 7: AC equivalent circuit modeling

 Equilibrium (dc) state-space averaged model

Provided that the natural frequencies of the converter, as well as the

frequencies of variations of the converter inputs, are much slower than
the switching frequency, then the state-space averaged model that
describes the converter in equilibrium is
0=AX+BU
Y=CX+EU

where the averaged matrices and the equilibrium dc

are components are
A = D A 1 + D' A 2 X = equilibrium (dc) state vector
B = D B 1 + D' B 2 U = equilibrium (dc) input vector
C = D C 1 + D' C 2 Y = equilibrium (dc) output vector
E = D E 1 + D' E 2 D = equilibrium (dc) duty cycle

Fundamentals of Power Electronics 72 Chapter 7: AC equivalent circuit modeling

 Solution of equilibrium averaged model

Equilibrium state-space averaged model:

0=AX+BU
Y=CX+EU
Solution for X and Y:
X = – A– 1 B U
Y = – C A– 1 B + E U

Fundamentals of Power Electronics 73 Chapter 7: AC equivalent circuit modeling

 Small-signal ac state-space averaged model

dx(t)
K = A x(t) + B u(t) + A 1 – A 2 X + B 1 – B 2 U d(t)
dt

y(t) = C x(t) + E u(t) + C 1 – C 2 X + E 1 – E 2 U d(t)

where
x(t) = small – signal (ac) perturbation in state vector
u(t) = small – signal (ac) perturbation in input vector
y(t) = small – signal (ac) perturbation in output vector
d(t) = small – signal (ac) perturbation in duty cycle

So if we can write the converter state equations during subintervals 1

and 2, then we can always find the averaged dc and small-signal ac
models

Fundamentals of Power Electronics 74 Chapter 7: AC equivalent circuit modeling

 7.4.3. Discussion of the state-space averaging result

As in Sections 7.1 and 7.2, the low-frequency components of the

inductor currents and capacitor voltages are modeled by averaging
over an interval of length Ts. Hence, we define the average of the
state vector as:
t + Ts

x(t) T = 1 x(τ) dτ
s Ts t

The low-frequency components of the input and output vectors are

modeled in a similar manner.
By averaging the inductor voltages and capacitor currents, one
obtains:

d x(t) Ts
K = d(t) A 1 + d'(t) A 2 x(t) + d(t) B 1 + d'(t) B 2 u(t)
dt Ts Ts

Fundamentals of Power Electronics 75 Chapter 7: AC equivalent circuit modeling

 Change in state vector during first subinterval

During subinterval 1, we have

dx(t)
K = A 1 x(t) + B 1 u(t)
dt
y(t) = C 1 x(t) + E 1 u(t)

So the elements of x(t) change with the slope

dx(t)
= K – 1 A 1 x(t) + B 1 u(t)
dt

Small ripple assumption: the elements of x(t) and u(t) do not change
significantly during the subinterval. Hence the slopes are essentially
constant and are equal to
dx(t)
= K – 1 A 1 x(t) + B 1 u(t)
dt Ts Ts

Fundamentals of Power Electronics 76 Chapter 7: AC equivalent circuit modeling

 Change in state vector during first subinterval

dx(t) x(t) K –1 A 1 x Ts
+ B1 u Ts
= K – 1 A 1 x(t) + B 1 u(t)
dt Ts Ts

x(0)
K –1 dA 1 + d'A 2 x Ts
+ dB

Net change in state vector over first 0 dTs

subinterval:
x(dTs) = x(0) + dTs K – 1 A 1 x(t) Ts
+ B 1 u(t) Ts

final initial interval slope

value value length

Fundamentals of Power Electronics 77 Chapter 7: AC equivalent circuit modeling

 Change in state vector during second subinterval

Use similar arguments.

State vector now changes with the essentially constant slope

dx(t)
= K – 1 A 2 x(t) + B 2 u(t)
dt Ts Ts

The value of the state vector at the end of the second subinterval is
therefore

x(Ts) = x(dTs) + d'Ts K – 1 A 2 x(t) Ts

+ B 2 u(t) Ts

final initial interval slope

value value length

Fundamentals of Power Electronics 78 Chapter 7: AC equivalent circuit modeling

 Net change in state vector over one switching period

We have:

x(dTs) = x(0) + dTs K – 1 A 1 x(t) Ts

+ B 1 u(t) Ts

x(Ts) = x(dTs) + d'Ts K – 1 A 2 x(t) Ts

+ B 2 u(t) Ts

Eliminate x(dTs), to express x(Ts) directly in terms of x(0) :

x(Ts) = x(0) + dTsK – 1 A 1 x(t) Ts

+ B 1 u(t) Ts
+ d'TsK – 1 A 2 x(t) Ts
+ B 2 u(t) Ts

Collect terms:

x(Ts) = x(0) + TsK – 1 d(t)A 1 + d'(t)A 2 x(t) Ts

+ TsK – 1 d(t)B 1 + d'(t)B 2 u(t) Ts

Fundamentals of Power Electronics 79 Chapter 7: AC equivalent circuit modeling

 Approximate derivative of state vector

x(t) K –1 A 1 x Ts
+ B1 u Ts
K –1 A 2 x Ts
+ B2 u Ts

x(t) Ts
x(Ts)
x(0)
K –1 dA 1 + d'A 2 x + dB 1 + d'B 2 u Ts
Ts

0 dTs Ts t
Use Euler approximation:
d x(t) x(Ts) – x(0)
≈
Ts

dt Ts
We obtain:
d x(t) Ts
K = d(t) A 1 + d'(t) A 2 x(t) + d(t) B 1 + d'(t) B 2 u(t)
dt Ts Ts

Fundamentals of Power Electronics 80 Chapter 7: AC equivalent circuit modeling

 Low-frequency components of output vector

y(t) C 1 x(t) Ts
+ E 1 u(t) Ts

y(t) Ts

C 2 x(t) Ts
+ E 2 u(t) Ts
0
0 dTs Ts t

Remove switching harmonics by averaging over one switching period:

y(t) Ts
= d(t) C 1 x(t) Ts
+ E 1 u(t) Ts
+ d'(t) C 2 x(t) Ts
+ E 2 u(t) Ts

Collect terms:
y(t) Ts
= d(t) C 1 + d'(t) C 2 x(t) Ts
+ d(t) E 1 + d'(t) E 2 u(t) Ts

Fundamentals of Power Electronics 81 Chapter 7: AC equivalent circuit modeling

 Averaged state equations: quiescent operating point

The averaged (nonlinear) state equations:

d x(t) Ts
K = d(t) A 1 + d'(t) A 2 x(t) + d(t) B 1 + d'(t) B 2 u(t)
dt Ts Ts

y(t) Ts
= d(t) C 1 + d'(t) C 2 x(t) Ts
+ d(t) E 1 + d'(t) E 2 u(t) Ts

The converter operates in equilibrium when the derivatives of all

elements of < x(t) >Ts are zero. Hence, the converter quiescent
operating point is the solution of
0=AX+BU
Y=CX+EU

where A = D A 1 + D' A 2 and X = equilibrium (dc) state vector

B = D B 1 + D' B 2 U = equilibrium (dc) input vector
C = D C 1 + D' C 2 Y = equilibrium (dc) output vector
E = D E 1 + D' E 2 D = equilibrium (dc) duty cycle
Fundamentals of Power Electronics 82 Chapter 7: AC equivalent circuit modeling
 Averaged state equations: perturbation and linearization

Let x(t) Ts
= X + x(t) with U >> u(t)
u(t) Ts
= U + u(t) D >> d(t)
y(t) Ts
= Y + y(t) X >> x(t)
d(t) = D + d(t) ⇒ d'(t) = D' – d(t) Y >> y(t)

Substitute into averaged state equations:

d X+x(t)
K = D+d(t) A 1 + D'–d(t) A 2 X+x(t)
dt

+ D+d(t) B 1 + D'–d(t) B 2 U+u(t)

Y+y(t) = D+d(t) C 1 + D'–d(t) C 2 X+x(t)

+ D+d(t) E 1 + D'–d(t) E 2 U+u(t)

Fundamentals of Power Electronics 83 Chapter 7: AC equivalent circuit modeling

 Averaged state equations: perturbation and linearization

dx(t)
K = AX + BU + Ax(t) + Bu(t) + A 1 – A 2 X + B 1 – B 2 U d(t)
dt

first–order ac dc terms first–order ac terms

+ A 1 – A 2 x(t)d(t) + B 1 – B 2 u(t)d(t)

second–order nonlinear terms

Y+y(t) = CX + EU + Cx(t) + Eu(t) + C 1 – C 2 X + E 1 – E 2 U d(t)

dc + 1st order ac dc terms first–order ac terms

+ C 1 – C 2 x(t)d(t) + E 1 – E 2 u(t)d(t)

second–order nonlinear terms

Fundamentals of Power Electronics 84 Chapter 7: AC equivalent circuit modeling
 Linearized small-signal state equations

Dc terms drop out of equations. Second-order (nonlinear) terms are

small when the small-signal assumption is satisfied. We are left with:

dx(t)
K = A x(t) + B u(t) + A 1 – A 2 X + B 1 – B 2 U d(t)
dt

y(t) = C x(t) + E u(t) + C 1 – C 2 X + E 1 – E 2 U d(t)

This is the desired result.

Fundamentals of Power Electronics 85 Chapter 7: AC equivalent circuit modeling

 7.4.4. Example: State-space averaging of a
nonideal buck-boost converter

ig(t) Q1 D1
Model nonidealities:
+
i(t) • MOSFET on-
+ resistance Ron
vg(t) L C R v(t)
–
• Diode forward voltage
– drop VD

state vector input vector output vector

i(t) vg(t)
x(t) = u(t) = y(t) = ig(t)
v(t) VD

Fundamentals of Power Electronics 86 Chapter 7: AC equivalent circuit modeling

 Subinterval 1

ig(t) Ron
di(t)
L = vg(t) – i(t) Ron +
dt i(t)
dv(t) v(t)
C =– + L C R v(t)
dt R vg(t) –
ig(t) = i(t)
–

– Ron 0 vg(t)
L 0 d i(t) = i(t)
+ 10
0C dt v(t) 0 – 1 v(t) 00 VD
R

dx(t)
K A1 x(t) B1 u(t)
dt

i(t) vg(t)
ig(t) = 10 + 00
v(t) VD

y(t) C1 x(t) E1 u(t)

Fundamentals of Power Electronics 87 Chapter 7: AC equivalent circuit modeling
 Subinterval 2

VD
di(t)
L = v(t) – VD

+
–
dt ig(t) +
dv(t) v(t)
C =– – i(t)
dt R vg(t) + L C R v(t)
–
ig(t) = 0
i(t)
–

L 0 d i(t) = 0 1 i(t) 0 –1 vg(t)

+
0C dt v(t) –1 – 1 v(t) 0 0 VD
R
dx(t)
K A2 x(t) B2 u(t)
dt

i(t) vg(t)
ig(t) = 00 + 00
v(t) VD

y(t) C2 x(t) E2 u(t)

Fundamentals of Power Electronics 88 Chapter 7: AC equivalent circuit modeling
 Evaluate averaged matrices

– Ron 0 0 1 – DRon D'

A = DA 1 + D'A 2 = D + D' =
0 – 1 – 1 – 1 – D' – 1
R R R

In a similar manner,

B = DB 1 + D'B 2 = D – D'
0 0
C = DC 1 + D'C 2 = D 0
E = DE 1 + D'E 2 = 0 0

Fundamentals of Power Electronics 89 Chapter 7: AC equivalent circuit modeling

 DC state equations

– DRon D' Vg
0 = I + D – D'
0=AX+BU 0 – D' –1 V 0 0 VD
or, R
Y=CX+EU
I + Vg
Ig = D0 00
V VD

D 1
DC solution: D' R D' R
2 Vg
I = 1
V R – D 1 VD
1 + D2 on D'
D' R
1 D2 D Vg
Ig =
R D' 2R D'R VD
1 + D2 on
D' R

Fundamentals of Power Electronics 90 Chapter 7: AC equivalent circuit modeling

 Steady-state equivalent circuit

DC state equations: – DRon D' Vg

0 = I + D – D'
0 – D' –1 V 0 0 VD
R
I + Vg
Ig = D0 00
V VD

Corresponding equivalent circuit:

DRon D'VD
1:D D' : 1

+
–
Ig I +

Vg + V R
–

Fundamentals of Power Electronics 91 Chapter 7: AC equivalent circuit modeling

 Small-signal ac model

Evaluate matrices in small-signal model:

V – IRon + VD V – V – IRon + VD
A 1 – A 2 X + B1 – B2 U = – V + g = g
I 0 I

C1 – C2 X + E1 – E2 U = I

Small-signal ac state equations:

– DRon D' vg(t) Vg – V – IRon + VD

L 0 d i(t) = i(t)
+ D – D' + d(t)
0
0 C dt v(t) – D' – 1 v(t) 0 0 ➚
vD(t) I
R

i(t) vg(t)
i g(t) = D 0 + 00 0 + 0 d(t)
v(t) 00 v➚
D(t)
I

Fundamentals of Power Electronics 92 Chapter 7: AC equivalent circuit modeling

 Construction of ac equivalent circuit

Small-signal ac d i(t)
L = D' v(t) – DRon i(t) + D vg(t) + Vg – V – IRon + VD d(t)
equations, in dt
scalar form: dv(t) v(t)
C = –D' i(t) – + I d(t)
dt R
i g(t) = D i(t) + I d(t)
Corresponding equivalent circuits: i g(t)
input
eqn
inductor equation vg(t) +
d(t) Vg – V + V D – IRon – I d(t) D i(t)
L DRon
+
–

+ d i(t) –
L
dt –
D vg(t) +
–
+ D' v(t) capacitor
i(t) eqn dv(t) v(t)
C +
dt R
D' i(t) I d(t) C v(t) R

Fundamentals of Power Electronics 93 Chapter 7: AC equivalent circuit modeling

 Complete small-signal ac equivalent circuit

Combine individual circuits to obtain

d(t) Vg – V + V D – IRon
i g(t) L
1:D D' : 1

+
–
i(t) DRon +

vg(t) + I d(t) I d(t) C v(t)

– R

Fundamentals of Power Electronics 94 Chapter 7: AC equivalent circuit modeling

Converter System Modeling via MATLAB/Simulink

A powerful environment for system modeling and simulation

MATLAB: programming and scripting environment

Simulink: block diagram modeling environment that runs inside MATLAB

Things we can achieve, relative to Spice:

•  Higher level of abstraction, suitable for higher-level system models
•  More sophisticated controller models
•  Arbitrary system elements
But:
•  We have to derive our own mathematical models
•  Simulink signals are unidirectional as in conventional block
diagrams
At CoPEC, nearly all simulation is done within MATLAB/Simulink
 Open-loop buck converter
Time domain simulation including switching ripple
Closed-loop buck converter, digital control
Time domain simulation with switching ripple
 Open-loop buck-boost converter
Frequency domain simulation, averaged model

Control-to-output transfer function

 Closed-loop buck converter
Frequency domain simulation, averaged model

Loop gain: Bode plot

 MATLAB/Simulink discussion

•  A structured way to write the converter averaged equations,

suitable for implementation in Simulink:
State-space averaging

•  Some basic converter models, implemented in Simulink
•  How to plot small-signal transfer functions in Simulink
•  Modeling the discontinuous conduction mode
 Synchronous buck converter
Formulating state equations for Simulink model

Averaging the input current

tr Q
i g = di L + iL + r
Ts Ts
Averaging the inductor voltage

Averaging the capacitor current:
vL(t) For both intervals,
i C = i – i Load

dTs Ts t Resulting state equations:

tr Q
i g = di L + iL + r
Ts Ts
vL = d vg – i R on + R L – vout + d
– i R on + R L – vout
L di = vL = dvg – i R on + R L – v – i – i Load esr
dt
with vout = v + i – i Load esr C dv = i C = i – i Load
dt
so vL = dvg – i R on + R L – v – i – i Load esr
vout = v + i – i Load esr
 Basic buck converter model
Averaged model for Simulink

Integration of state
variables

Outputs

Independent
inputs

(used in current
Embedded MATLAB mode control)
code block:
•  Load inputs from u
vector
•  Set circuit parameters
•  Calculate state
equations and outputs
•  Place results in output
y vector
 Time-domain simulation
Synchronous buck example, Simulink

Simulink model
employing
synchronous buck
model, with voltage
mode control

Output voltage transient

response

Generating a Bode Plot from the Simulink file

1. Set transfer function input and output points

•  Right-click on the desired wire

•  Select “Linearization Points”, then “input point” or “output point”
 Generating a Bode Plot from Simulink, p. 2

%% Bode plotter using linearization tool

% requires simulink control design toolbox

mdl = 'buckCPM4Vmodetester'; % set to file name of simulink model. Must
have i/o points set within this model

io = getlinio(mdl) % get i/o signals of mdl

op = operspec(mdl)

op = findop(mdl,op) % calculate model operating point

lin = linearize(mdl,op,io) % compute state space model of linearized
system

ltiview(lin) % send linearized model to LTI Viewer tool

•  Save this as a script (“.m file”) and run it whenever you want to
generate a Bode plot
•  This script finds the steady-state operating point and linearizes the
model
•  The last line opens the LTI Viewer tool, which generates various
small-signal plots including Bode, step response, pole/zero, Nyquist,
etc.
 Control-to-output transfer function Gvd
Generated by Simulink

Synchronous buck example of previous slides

 Modeling DCM
Buck example, Simulink model

iL(t)
ipk

dTs d2Ts Ts t

In DCM, the average inductor current Treat the average inductor current
can be expressed as:
as an independent state, and solve
for d2:


   



 







 


     





Note that d2 = 1 – d in CCM, and
 
  
 d2 < 1 – d in DCM



 Combined CCM-DCM model
Buck converter

Simulink model

State equations:

    

 

  
 
    





    


 

Buck control-to-output transfer function
CCM vs. DCM

R = 3Ω: CCM

R = 24Ω: DCM
 9.5.4. Design example

L
50 µH
+
iload
vg(t) + C R
v(t)
28 V – 500 µF 3Ω
Sensor
– H(s) gain
fs = 100 kHz
Transistor Error
gate driver signal
δ Pulse-width vc G (s) ve –+ Hv
modulator c

VM = 4 V Compensator vref
5V

Fundamentals of Power Electronics 46 Chapter 9: Controller design

 Closed-loop buck converter
Simulink frequency domain simulation, averaged model

fc

ϕm

Loop gain: Bode plot

Injection point for
measurement of
loop gain T(s)
Transfer function blocks:

Implementing the PID compensator

 Closed-loop line-to-output transfer function
Simulink frequency domain simulation

Closed loop Gvg

Open loop

Closed loop

Open loop Gvg

Script that generates both plots