International Conference on Innovations in Power and Advanced Computing Technologies [i-PACT2017]
1
BBBC based Optimization of PI Controller
Parameters for Buck Converter
Akhilesh K. Gupta Bandi Mallikarjuna Reddy
Dept. of Electrical Engineering, MNNIT Dept. of Electrical Engineering, MNNIT
Utter Pradesh, Allahabad, INDIA Utter Pradesh, Allahabad, INDIA
akhileshgupta08@ gmail.com ree1505@mnnit.ac.in
Deepak Kumar Paulson Samuel
Dept. of Electrical Engineering, MNNIT Dept. of Electrical Engineering, MNNIT
Utter Pradesh, Allahabad, INDIA Utter Pradesh, Allahabad, INDIA
deepak_kumar@mnnit.ac.in paul@mnnit.ac.in
Abstract— the most widely used controllers in industries are PI or
PID controllers. The major concern with designing of such
controllers is the determination of controller parameters. An
intelligent method is discussed in this paper to determine the
controller parameter to control the dynamic performance of buck
converter by optimizing these parameters with the big bang big
crunch (BBBC) algorithm. Initially, the mathematical modeling is
developed and thereafter the weighted numerical values of
overshoot, peak time, rise time and settling time are summed to
make a fitness function which is to be minimize for the better
dynamic response. The performance of BBBC-PI controller is
analyzed by settling time, rise time and overshoot of the output
response. The disturbance rejection ability of optimized PI
controller is verified by three cases such as step change in input
voltage, output voltage and output load resistance. The closed loop
operation of buck converter is simulated and verified at the
MATLAB/Simulink platform.
Keywords– BBBC, Buck Converter, Fitness function, PI,
Optimization Algorithm
I. INTRODUCTION
ower electronics converters are gaining adamant
popularity in the field of generation and transmission of
electrical power. Conversion of one level of electrical
energy to the other level of electrical energy is possible by
these power converters in the field of SMPC. The energy is
first transmitted via electronic switches to energy storage
devices and afterward these devices act as a source to transmit
the stored energy into the load. MOSFETs and IGBTs are most
popular choice for the controlled switch application whereas
diodes act as an uncontrolled switch.
Inductors and capacitors are used as energy storage
elements. This process of energy transfer results in an output
voltage that is pertaining to the input voltage by the duty ratios
of the switches. The buck converter [1-2] is a ubiquitous DC-
DC converter that efficiently converts a high voltage to a low
voltage efficiently. Efficient power condition is very important
to enhance the battery life, and to reduce heat generated during
the application. The buck converter can be used in lots of
specific applications like microprocessor and embedded
systems. It offers the required local voltage from a higher
voltage bus that is common to several converters in the system.
The buck converter itself comprises of one active controlled
switch and an uncontrolled switch along with the filters
elements. This great simplicity allows for cost operative high
efficient power distribution throughout the application. The
buck converter has the filter inductor on the output side, which
provides a smooth continuous output current waveform to the
load. This could be reflected a qualitative profit but requires
special considerations for huge load transients.
Buck converter provides the regulated dc supply from
unregulated dc supply for various application which needs
regulated dc supply. A number of modern control techniques
such as optimal control [3] linear control [4], nonlinear control
[5], sliding mode control [6] and hybrid controllers [7] etc.
have been proposed to improve the dynamic performance of
power converters. Along with above explained techniques,
state space control [8] and LQR control [9] are also used as
output voltage control method for dc/dc converters. Although,
several methods are available for closed loop operation of
converter but PID and PI error control technique is still most
popular and powerful technique because of its simplicity in
hardware as well as software implementation. The main
challenge with the PID or PI controller implementation is
tuning of their parameters. These parameters are main
constraints for the effectiveness and robust transient
performance of the controller. Ziegler-Nichols [10] method is
applied to obtain the closed loop behavior of controller. The
complex mathematical model requirement and high overshoot
limits the use of this method.
Since last few years, a numerical approach for finding
optimum solution under certain constraints, called optimization
techniques are applied to find the parameters of PI or PID
controllers to improve the performance and robustness
behavior of controller. Genetic Algorithm [11-12], Particle
Swarm Optimization [13-14] and Bacterial Foraging
Optimization (BFO) [15-16] and Artificial Bee Colony (ABC)
[17] are successfully applied by the several researchers for
closed loop operation of buck converter. In this article, the
P
978-1-5090-5682-8 /17/$31.00 ©2017 IEEE