Code covered by the BSD License  

Highlights from
Configurable Simulink Model for DC-DC Converters with PWM PI Control

image thumbnail
from Configurable Simulink Model for DC-DC Converters with PWM PI Control by Yi Cao
A Simulink model configurable to buck, boost and buck-boost DC-DC converters with PWM PI control

Configurable Simulink Model for DC-DC Converters

Configurable Simulink Model for DC-DC Converters

Contents

DC-DC Converters

There are three kinds of switching mode DC-DC converters, buck, boost and buck-boost. The buck mode is used to reduce output voltage, whilst the boost mode can increase the output voltage. In the buck-boost mode, the output voltage can be maintained either higher or lower than the source but in the opposite polarity. The simplest forms of these converters are schematically represented in Figure 1.

Figure 1. Buck (left), Boost (middle) and Buck-Boost (right) converters.

These converters consist of the same components, an inductor, $L$, a capacitor, $C$ and a switch, which has two states $u=1$ and $u=0$. All converters connect to a DC power source with a voltage (unregulated), $V_\mathrm{in}$ and provide a regulated voltage, $v_\mathrm{o}$ to the load resistor, $R$ by controlling the state of the switch. In some situations, the load also could be inductive, for example a DC motor, or approximately, a current load, for example in a cascade configuration. For simplicity, here, only current and resistive loads are to be considered.

Principles

The working principles of the these DC-DC converters can be explained as follows. In the buck mode, when the switch is on position 1, the DC source supplies power to the circuit which results an output voltage across the resistor. When the switch changes its position to 0, the energy stored in the inductor and capacitor will discharge through the resistor. Appropriately controlling the switching position can maintain the output voltage at a desired level lower than the source.

In the boost mode, when the switch is on position 1, the circuit is separated into two parts: on the left, the source is charging the inductor, meanwhile the capacitor on the right maintains the output voltage using previously stored energy. When the switch changes its position to 0, both the DC source and energy stored in the inductor will supply power to the circuit on the right, hence boost the output voltage. Again, the output voltage can be maintain at desired level by controlling the switching sequence.

Finally, for the buck-boost mode, switch positions 1 and 0 represents charging and discharging modes of the inductor. Appropriately controlling the switching sequence can result in output voltage higher or lower than the DC source. Since the inductor cannot change the direction of current, the output voltage is opposite to the DC source.

Model under ideal assumptions

Under ideal assumptions: ideal switch, ideal capacitor and ideal inductor, these converters can be described using ordinary differentiation equations as follows:

Buck converter:

$C{dv_{c} \over dt} = i_{L} - v_{c}/R - i_{o}$

$L{di_{L} \over dt} = uv_{in} - v_{c}$

where, $i_\mathrm{o}$ is the load current.

Boost converter:

$C{dv_\mathrm{c}\over dt}=(1-u)i_\mathrm{L}-v_\mathrm{c}/R-i_\mathrm{o}$

$L{di_\mathrm{L}\over dt}=v_\mathrm{in} - (1-u) v_\mathrm{c}$

Buck-boost converter:

$C{dv_\mathrm{c}\over dt}=(1-u)i_\mathrm{L}-v_\mathrm{c}/R-i_\mathrm{o}$

$L{di_\mathrm{L}\over dt}=uv_\mathrm{in} - (1-u) v_\mathrm{c}$

Introduce the following state, time and load normalization:

$x_1={v_\mathrm{c}\over v_\mathrm{in}}$,

$x_2={i_\mathrm{L}\over v_\mathrm{in}}\sqrt{{L}\over{C}}$,

$\tau = {t \over \sqrt{LC}}$,

$\gamma = {\sqrt{LC}\over R}$,

$d = {i_\mathrm o \over v_\mathrm{in}}\sqrt{{L}\over {C}}$

Then the normalized state equations of three converters are as follows:

Normalized buck model

$\dot{x_1} = -\gamma x_1 + x_2 - d$

$\dot{x_2}= -x_1 + u $

where, with an abuse of notation, `.' represents the derivation with respect to the normalized time, $\tau$.

Normalized boost model

$\dot{x_1} = -\gamma x_1 + (1-u)x_2 - d$

$\dot{x_2} = -(1-u)x_1 + 1$

Normalized buck-boost model

$\dot{x_1} = -\gamma x_1 + (1-u)x_2 -d$

$\dot{x_2} = -(1-u)x_1 + u$

Model with body resistors

In more general cases, a body resistor of the inductor, $R_\mathrm{L}$ and an equivalent series resistor (ESR) of the capacitor, $R_\mathrm{c}$ can be added to the above models.

Buck model with $R_\mathrm{L}$ and $R_\mathrm{c}$

Since,

$C{dv_\mathrm{c}\over dt} = i_\mathrm{L} - v_\mathrm{o}/R - i_\mathrm{o}$

$v_\mathrm{o} = v_\mathrm{c} + R_\mathrm{c}C{dv_\mathrm{c}\over dt}$

$L{di_\mathrm{L}\over dt} = uv_\mathrm{in} - v_\mathrm{o} - R_\mathrm{L}i_\mathrm{L}$

Inserting the second equation into the first leads to:

$C{dv_\mathrm{c}\over dt} = i_\mathrm{L} - v_\mathrm{c}/R - {R_\mathrm{c}\over R}C{dv_\mathrm{c}\over dt}- i_\mathrm{o}$

$\left(1+{R_\mathrm{c}\over R}\right)C{dv_\mathrm{c}\over dt}=i_\mathrm{L} - v_\mathrm{c}/R - i_\mathrm{o}$

Hence,

$v_\mathrm{o}={Rv_\mathrm{c}\over R+R_\mathrm{c}}+{RR_\mathrm{c}\over R+R_\mathrm{c}}(i_\mathrm{L}-i_\mathrm{o})$

and the overall model is

$C{dv_\mathrm{c}\over dt}={R\over R+R_\mathrm{c}}\left(i_\mathrm{L} - {v_\mathrm{c}\over R} - i_\mathrm{o}\right)$

$L{di_\mathrm{L}\over dt} = uv_\mathrm{in} - {Rv_\mathrm{c}\over R+R_\mathrm{c}} - \left(R_\mathrm{L}+{RR_\mathrm{c}\over R+R_\mathrm{c}}\right) i_\mathrm{L} + {RR_\mathrm{c}i_\mathrm{o}\over R+R_\mathrm{c}}$

$v_\mathrm{o}={Rv_\mathrm{c}\over R+R_\mathrm{c}}+{RR_\mathrm{c}\over R+R_\mathrm{c}}(i_\mathrm{L}-i_\mathrm{o})$

Boost model with $R_\mathrm{L}$ and $R_\mathrm{c}$

$C{dv_\mathrm{c}\over dt} = (1-u)i_\mathrm{L} - v_\mathrm{o}/R - i_\mathrm{o}$

$L{di_\mathrm{L}\over dt} = v_\mathrm{in} - (1-u) v_\mathrm{o} - R_\mathrm{L}i_\mathrm{L}$

$v_\mathrm{o} = {Rv_\mathrm{c}\over R+R_\mathrm{c}}+{RR_\mathrm{c}\over R+R_\mathrm{c}}((1-u)i_\mathrm{L}-i_\mathrm{o})$

Buck-boost model with $R_\mathrm{L}$ and $R_\mathrm{c}$

$C{dv_\mathrm{c}\over dt} = (1-u)i_\mathrm{L} - v_\mathrm{o}/R-i_\mathrm{o}$

$L{di_\mathrm{L}\over dt}=uv_\mathrm{in}-(1-u) v_\mathrm{o}-R_\mathrm{L}i_\mathrm{L}$

$v_\mathrm{o}={Rv_\mathrm{c}\over R+R_\mathrm{c}}+{RR_\mathrm{c}\over R+R_\mathrm{c}}((1-u)i_\mathrm{L}-i_\mathrm{o})$

Simulink Model

These three modes of DC-DC converters have been uniformly implemented in the MATLAB/Simulink as show in Figure~\ref{fig:simulinkmodel}.

Figure 2. A uniform Simulink model of DC-DC converters.

The input-output connections of the model is shown in Figure 3.

Figure 3. Input and output connections of the DC-DC converter model.

The first input to the model is the switch signal eight 1 or 0. The second one defines the DC source voltage and internal resistance. The third input is used to define the output current. The model has two outputs, the output voltage and the inductor current, which are the states of the system.

The model can be configure with a number of parameters as shown in Figure~\ref{fig:simulinkmodelparameters}. These parameters are: the capacitance, $C$, inductance, $L$, the internal resistance of the capacitor and the inductor, $R_C$ and $R_L$ respectively. Three converter modes can be selected through the pull-down menu. One can also define either zero or non-zero value to the initial capacitor voltage by selecting or de-selecting the ``zero capacitor voltage'' option. Finally, the option ``Positive Inductor Current'' defines whether the condition $i_L\ge 0$ should be enforced or not.

Figure 4. Parameters of the DC-DC converter model.


Published with MATLAB® 7.7

Contact us