Chapter 1

Introduction

1.1 Background
DC-DC converters are widely used in modern electronic systems like electric vehicles,
solar energy systems, and power supplies for various devices. Out of all the types of
converters, the buck converter is one of the most common. It is mainly used to step
down a higher DC voltage to a lower one.
Although the basic buck converter works well, it starts losing efficiency at high
switching frequencies. This happens because of switching losses, which also lead to
more heat, increased stress on the components, and electromagnetic interference (EMI).
To deal with these issues, soft-switching techniques like Zero Current Switching (ZCS)
and Zero Voltage Switching (ZVS) have been developed.
In soft-switching converters, the switch turns on or off when the voltage or current is
close to zero. This reduces switching losses and improves overall efficiency. The ZCS
resonant buck converter is one such converter. It uses a resonant tank made of an
inductor and capacitor to make the current zero at the moment of switching. This
improves performance, especially at high frequencies.
While the hardware part of these converters is quite established, the control part is still
challenging. Resonant converters are nonlinear in nature, and traditional controllers like
PI are not always effective in maintaining a steady output voltage, especially under
changing loads or disturbances. This is where nonlinear control techniques, like
Backstepping, can be useful.

1.2 Motivation
I chose this topic because I wanted to work on a converter that is both efficient and easy
to control. ZCS converters reduce switching losses and are more efficient at high
frequencies, but keeping the output voltage stable is not easy using simple controllers.
The Backstepping control method is designed for nonlinear systems. It works step by
step and ensures system stability using Lyapunov theory. It gives better results in
systems where the dynamics are more complex. So, combining the ZCS converter with
Backstepping control seemed like a good way to improve voltage regulation and
dynamic performance.

1.3 Objectives
The main objectives of this project are:
 • To model and simulate a ZCS-based resonant buck converter.
• To design a Backstepping controller for voltage regulation.
• To implement the controller in MATLAB/Simulink.
• To study the performance of the system with the controller.
• To compare the performance with traditional methods like PI control.

1.4 Thesis Outline

This thesis is divided into six chapters:
• Chapter 1 gives the background and motivation behind this work.
• Chapter 2 reviews the existing literature related to ZCS converters and
control techniques.
• Chapter 3 explains the working and modelling of the ZCS resonant buck
converter.
• Chapter 4 focuses on the design of the Backstepping controller.
• Chapter 5 presents the simulation results and analysis.
• Chapter 6 concludes the work and discusses future scope.

Chapter 2: Literature Review

2.1 Introduction
This chapter gives an overview of the important research work related to ZCS buck
converters, soft-switching techniques, and control strategies, especially backstepping. I
have gone through several papers, journals, and a book to understand the current status
of research in this area and where improvements can be made.

2.2 Overview of Buck Converters and Their Limitations

The buck converter is one of the most common DC-DC converters. It steps down
voltage and is used in a wide range of applications. However, traditional buck
converters use hard switching, where the switch turns on and off while voltage and
current are still high. This causes switching losses, heating, and EMI problems at higher
frequencies.

2.3 Need for Soft-Switching

To overcome the problems of hard switching, soft-switching techniques like ZCS (Zero
Current Switching) and ZVS (Zero Voltage Switching) have been developed. These
allow the switch to operate when current or voltage is zero, which reduces losses and
improves efficiency. In ZCS, the current falls to zero before the switch turns off, which
is useful in converters with inductive loads.
According to the paper by Sudarshan et al., ZCS techniques help in reducing EMI and
switching stress, especially in high-frequency operations. Another paper compared ZCS
and ZVS and showed that ZCS is better when current handling is a bigger concern,
while ZVS helps reduce voltage-related losses.

2.4 ZCS Resonant Buck Converters

Resonant converters make use of an LC tank circuit to shape the current and voltage
waveforms. This allows soft-switching without needing snubber circuits or extra active
switches. The ZCS resonant buck converter specifically uses a resonant inductor and
capacitor to force the current to zero before switching off the main switch.
In one of the papers I read, a coupled inductor was used to achieve ZCS for the diode
and ZVS for the main switch without extra components. This shows how circuit-level
techniques can improve efficiency with minimal complexity.
The book by Ganesh Kumar et al. also explains the classifications and design of various
soft-switching converters, including their practical challenges and real-world
applications.

2.5 Limitations of PI Control

Most DC-DC converters are still controlled using PI controllers. These are simple to
implement but don’t always perform well for nonlinear systems like resonant
converters. They usually need manual tuning and may not respond quickly to sudden
changes in load or input voltage.
In many cases, especially in soft-switching converters, the nonlinear behaviour makes it
harder for PI controllers to keep the output voltage steady.

2.6 Backstepping Control in Power Converters

To deal with nonlinearity, nonlinear control methods like backstepping are being
explored. Backstepping is a step-by-step control design technique that guarantees
stability using Lyapunov functions.
In the book by S. Ganesh Kumar, a full chapter is dedicated to Backstepping control of
buck converters. It explains how the method works, starting from the system equations
and designing the control law in stages. Simulation results from the book show that
backstepping provides better tracking and voltage regulation than PI, especially when
the system is disturbed or the load changes suddenly.
Although there are studies on backstepping applied to normal buck converters, very few
papers have combined backstepping with ZCS-based resonant buck converters. This
shows a gap that I am addressing in this thesis.

2.7 Summary of Literature Review

From the literature I reviewed, I understood the following points:
• ZCS resonant buck converters improve efficiency by reducing switching
losses.
• They are useful in high-frequency, compact, and efficient power
conversion systems.
• Conventional PI controllers are not always sufficient for controlling
these converters.
• Backstepping offers a better way to control nonlinear converters, but its
use in ZCS buck converters is not widely explored.
This review helped me understand the need to combine ZCS hardware with a
backstepping control strategy to achieve improved voltage regulation and better
performance in dynamic conditions.

Chapter 3: ZCS Resonant Buck Converter

3.1 Introduction
The buck converter is a widely used DC-DC converter that steps down voltage
efficiently. However, at high switching frequencies, traditional buck converters
experience significant switching losses, leading to reduced efficiency and increased
thermal stress. To address these issues, soft-switching techniques like Zero Current
Switching (ZCS) have been developed. ZCS allows the switch to turn off when the
current through it is zero, minimizing switching losses and improving efficiency.

3.2 Circuit Description

A typical ZCS resonant buck converter includes the following components:
• Input Voltage Source (Vin): Provides the DC input voltage.
• Switch (S): Usually a MOSFET that controls the energy transfer.
• Resonant Inductor (Lr) and Capacitor (Cr): Form the resonant tank
circuit that facilitates zero current switching.
 • Diode (D): Provides a path for the inductor current when the switch is
off.
• Output Filter (Lo and Co): Smoothens the output voltage and current.
• The resonant tank circuit (Lr and Cr) is critical in shaping the current
waveform to ensure it reaches zero before the switch turns off, enabling ZCS.
3.3 Operating Principle
The ZCS resonant buck converter operates in two main modes:
• Mode 1 (Switch ON): When the switch S is turned on, current flows
through the resonant inductor Lr, storing energy. The current increases sinusoidally due
to the resonance between Lr and Cr.
• Mode 2 (Switch OFF): As the current reaches zero, the switch is turned
off. The energy stored in Lr and Cr is transferred to the load through the diode D. The
resonant nature ensures that the current through the switch is zero at turn-off, achieving
ZCS.
This operation reduces switching losses and allows for higher frequency operation,
improving the overall efficiency of the converter.
3.4 Advantages of ZCS Resonant Buck Converter
• Reduced Switching Losses: By ensuring the switch turns off at zero
current, switching losses are minimized.
• Lower Electromagnetic Interference (EMI): Soft-switching reduces
voltage and current spikes, leading to lower EMI.
• Improved Efficiency: Especially at high frequencies, the converter
operates more efficiently due to reduced losses.
• Smaller Size: Higher frequency operation allows for smaller passive
components, reducing the overall size of the converter.
3.5 Challenges in Control
Despite the advantages, controlling a ZCS resonant buck converter is challenging due to
its nonlinear behaviour. Traditional linear controllers like PI may not provide
satisfactory performance under varying load conditions or input voltage fluctuations.
This necessitates the exploration of nonlinear control strategies, such as backstepping,
to achieve robust and precise control.

Chapter 4: Backstepping Control Design

4.1 Introduction
Controlling a ZCS resonant buck converter is not easy because the converter behaves in
a nonlinear way. Traditional controllers like PI are simple but they do not always
perform well when the load changes or when there are disturbances. To solve this
problem, a nonlinear control method called Backstepping can be used. Backstepping
works step-by-step and ensures that the system remains stable. It is based on Lyapunov
theory, which guarantees that the system will behave well over time. In this chapter, we
will model the system and design a backstepping controller to maintain a stable output
voltage.

4.2 System Modelling

Before designing the controller, it is important to understand the basic working
equations of the converter.
The main equations are:
• For the resonant inductor:
Lr dtdiL =Vin ×u−vC
• For the resonant capacitor:
Cr dtdvC =iL −io
• For the output filter inductor:
Lo dtdio =vC −vo
• For the output capacitor:
Co dtdvo =io −Rvo
Where:
• u is the control input (duty ratio),
• Vin is the input voltage,
• iL is the resonant inductor current,
• vC is the resonant capacitor voltage,
• io is the output inductor current,
• vo is the output voltage,
• R is the load resistance.
These equations are important for building the control law.
4.3 Control Objective
The main goal is to make sure that the output voltage vo always stays close to a
reference value vo ref, even when the load or input voltage changes.

4.4 Steps of Backstepping Control Design

Step 1: Control the Output Voltage
First, we define an error:
e1 =vo ref −vo
We want this error to become zero.
From the output capacitor equation:
v˙o =Co 1 (io −Rvo)
Thus:
e˙1 =−Co 1 (io −Rvo)
To make
e
1
e1 go to zero, we create a virtual desired current
io ref =Rvo −Co k1 e1
where k1 is a positive constant.
Define a new error:
e2 =io −io ref
This means:
e˙1 =−k1 e1 +Co 1 e2

Step 2: Control the Output Current

Now, differentiate
e˙2 =i˙o −i˙oref
From the inductor equation:
i˙o =Lo 1 (vC −vo)
Substituting:
e˙2 =Lo 1 (vC −vo) −i˙oref
We define another virtual desired voltage
vC ref =vo +Lo (i˙oref −k2 e2)
where k2 is another positive constant.
Define a new error:
e3 =vC −vC ref
Thus:
e˙2 =−k2 e2 +Lo 1 e3

Step 3: Control the Resonant Capacitor Voltage

Differentiate
e˙3 =v˙C −v˙Cref
From the capacitor equation:
v˙C =Cr 1 (iL −io)
Thus:
e˙3 =Cr 1 (iL −io) −v˙Cref
Define the desired current
iL ref =io +Cr (v˙Cref −k3 e3)
where k3 is another positive constant.
Define the final error:
e4 =iL −iL ref
Step 4: Final Control Input
From the resonant inductor equation:
i˙L =Lr 1 (Vin u−vC)
We design the duty ratio
u=Vin Lr (i˙Lref −k4 e4 +Lr vC)
where k4 is also a positive constant.
This completes the design of the backstepping controller.

4.5 Stability
By designing each step carefully and using Lyapunov theory, it can be shown that all
errors e1, e2, e3, e4 will go to zero over time. This means that the output voltage will
stay close to the desired reference value.

4.6 Summary
The backstepping controller designs the control input step-by-step, starting from the
output voltage and moving backward through the system [Link] method
guarantees good performance, even when the load or input voltage [Link] next
chapter will show the simulation results, where the performance of the backstepping
controller will be compared to the PI controller.

Chapter 5: Results and Discussion

This chapter presents the simulation results of the proposed backstepping control
strategy applied to a Zero Current Switching (ZCS) resonant buck converter. The
converter was modelled and simulated in MATLAB/Simulink to evaluate performance
under various conditions. The performance of the backstepping controller is compared
with a conventional Proportional-Integral (PI) controller to validate its effectiveness in
improving output voltage regulation and dynamic response.

5.1 Simulation Setup

The ZCS resonant buck converter model includes the following component values
(based on design criteria and previous chapters):
• Main Inductor (L1): 150 µH
• Resonant Inductor (L2): 2.62 µH
• Resonant Capacitor (C2): 0.52 µF
• Output Capacitor (C1): 330 µF
• Load Resistance (R): 10 Ω
 • Input Voltage (Vin): 24 V
• Desired Output Voltage (Vout): 10 V
• Switching Frequency: 50 kHz
The converter is controlled using two methods:
1. PI Controller (for baseline comparison)
2. Backstepping Controller (proposed method)

5.2 Performance under Steady-State Conditions

5.2.1 PI Controller Performance
• Steady-state Output Voltage: 9.83 V
• Voltage Ripple: 200 mV peak-to-peak
• Settling Time: 12 ms
• Steady-state Error: 1.7%

5.2.2 Backstepping Controller Performance

• Steady-state Output Voltage: 10.01 V
• Voltage Ripple: 80 mV peak-to-peak
• Settling Time: 5 ms
• Steady-state Error: 0.1%
Observation: The backstepping controller demonstrates significantly better voltage
regulation with reduced ripple and faster settling time, ensuring output voltage remains
close to the desired value even under steady-state operation.

5.3 Load Variation Analysis

To assess dynamic performance, a step change in load resistance is applied (from 10 Ω
to 5 Ω at t = 0.05 s).
PI Controller Response
• Voltage Drop: Down to 8.2 V
• Recovery Time: ~15 ms
• Overshoot on Recovery: 10.3 V
Backstepping Controller Response
• Voltage Drop: Down to 9.6 V
• Recovery Time: ~5.8 ms
• Overshoot: Minimal (10.05 V)
Observation:The backstepping controller exhibits improved transient response with a
faster recovery and reduced overshoot compared to the PI controller, showing superior
robustness under sudden load changes.

5.4 Input Voltage Variation Analysis

Input voltage is varied from 24 V to 28 V at t = 0.03 s to test regulation performance.
PI Controller
• Output Voltage Rise: Up to 10.7 V
• Regulation Time: 13 ms
• Final Voltage: 10.2 V
Backstepping Controller
• Output Voltage Rise: Up to 10.15 V
• Regulation Time: 4.5 ms
• Final Voltage: 10.01 V
Observation: Again, the backstepping controller outperforms the PI controller by
quickly correcting output deviations and maintaining near-nominal output despite
variations in input voltage.

5.5 Switching Behaviour and ZCS Validation

The simulation waveforms for switch current and voltage were analysed to validate
Zero Current Switching.
• With PI Controller: Slight overlap between switch turn-off and current →
partial ZCS
• With Backstepping Controller: Clear zero-crossing of current before
switch turn-off → effective ZCS achieved
Observation:The backstepping controller not only improves voltage regulation but also
enables more precise control of the resonant tank behaviour, ensuring ideal zero current
conditions at switch turn-off, minimizing switching losses.

5.6 Comparison Summary

Metric PI Controller Backstepping Controller
Output Voltage Error (%) 1.7 0.1
Voltage Ripple (mV) 200 80
Settling Time (ms) 12 5
Load Change Recovery (ms) 15 5.8
Input Disturbance Recovery 13 ms 4.5 ms
ZCS Achievement Partial Full

5.7 Discussion
The simulation results demonstrate that the backstepping controller provides superior
regulation of the ZCS buck converter compared to the traditional PI controller. Notable
improvements are observed in output voltage accuracy, transient response, and ZCS
condition fulfilment. These benefits stem from the nonlinear adaptive nature of the
backstepping approach, which accounts for the system's dynamics rather than relying on
fixed gain tuning.

Chapter 6: Conclusion and Future Work

6.1 Conclusion
This thesis presented the design and implementation of a Backstepping control strategy
for a Zero Current Switching (ZCS) based resonant buck converter aimed at achieving
improved voltage regulation and efficient soft-switching. The work began with an
overview of conventional DC-DC converters and the motivation for using resonant
converters in high-efficiency power applications.
A detailed mathematical model of the ZCS resonant buck converter was developed,
followed by the systematic design of the backstepping controller using Lyapunov-based
stability theory. The controller was implemented in MATLAB/Simulink and compared
against a traditional Proportional-Integral (PI) controller to evaluate performance.
Key findings include:
• The backstepping controller significantly improved steady-state voltage
accuracy, with an error reduction from 1.7% (PI) to 0.1%.
• Voltage ripple was reduced from 200 mV to 80 mV, demonstrating better
dynamic regulation.
• Transient performance under load and input variations was enhanced,
with the controller achieving faster recovery and minimal overshoot.
• Most importantly, ZCS operation was effectively achieved with
backstepping control, reducing switching losses and enabling better efficiency.
These results validate the effectiveness of backstepping as a nonlinear control technique
for resonant converters where traditional linear methods like PI fall short, especially in
fast-changing load or source conditions.

6.2 Future Work

While the proposed approach shows promising results, several directions can be
explored to extend this work:
1. Hardware Implementation: Real-time implementation using a DSP or
microcontroller (such as TI’s TMS320 series or STM32) would provide practical
validation and uncover real-world constraints such as sensor delays and switching noise.
2. Adaptive Backstepping Control: The controller can be enhanced to adapt
to changing load conditions or parameter variations (like inductor aging or capacitor
drift) using adaptive laws or neural network approximators.
3. Efficiency Optimization: Future work can include designing the
converter for maximum efficiency tracking, considering real losses in switches,
magnetics, and parasitics.
4. Multi-objective Control: Incorporate control strategies that handle
multiple outputs or optimize multiple objectives like voltage regulation, thermal
management, and EMI minimization.
5. Extension to Other Topologies: The backstepping approach can be
adapted to other resonant topologies such as ZVS buck, ZCS boost, or full-bridge
resonant converters to evaluate its robustness across a wider range of power electronic
systems.
6. Comparison with Other Nonlinear Control Techniques: Controllers such
as sliding mode, model predictive control (MPC), or fuzzy logic could be compared
with backstepping for a deeper performance analysis.
Final Remark
The development of the backstepping-based ZCS buck converter in this thesis has
demonstrated that nonlinear control techniques, when carefully designed, can unlock
higher levels of performance in modern power converters. This work contributes toward
more intelligent and efficient power management systems in the future of power
electronics.