Robust Controller Design for DC-DC Converters Using Fuzzy Logic Abstract— DC-DC converters in DC system are equally important as that of transformers in AC system. Applications of DC-DC converters include integration of renewable sources, power supplies, controlled electric drives etc. Digital design of fuzzy logic controller (FLC) based robust controller for DC-DC converter is presented in this research paper. The FLC shows good performance where linear control techniques are unable to produce desired result. The power electronic converters suffer from non-linearity and therefore conventional linear controllers don’t provide desired response. The proposed FLC is inspected for practical operating conditions of DC-DC converters such as load changes, input voltage variations etc. A comparative simulation result is produced for practical operating conditions which shows superiority and robustness of controller. The proposed digital design of FLC can be implemented on DSP for generating switching duty cycle for converters. Keywords—fuzzy logic controller; robust control; DC-DC converter; buck-converter; FLC. I. INTRODUCTION DC-DC converters are extensively used today in power processing for regulated power supplies, electrical drives, renewable energy conversion system, electric traction etc. The dynamic performance of DC-DC converters can be enhanced by application of robust control schemes [1]. From many years, converters have been controlled using analog and linear control techniques. DC-DC converters shows non-linear peculiarity due to fluctuating operating points, which affects performance of converter. To achieve better performance, there are two solutions. One is to develop more exact model of converter and second is to use a non- linear controller. The exact model of system is extremely complex to use for controller design. A FLC is capable of adapting fluctuating operating points, so it is utilized as non- linear controller for DC-DC converters [2-4]. Design of a digital controller for buck converter by using state-space average (SSA) technique is presented in [5]. To obtain stable operation of buck and boost converter, classical PI and FLC based control strategy is presented in [6-8]. To rectify the problem of nonlinear nature of DC-DC converter, a FLC is used as it does not require precise mathematical model of plant. A general FLC design that can be used for wide variety of DC-DC converters is inspected in reference [9]. A FLC for full-bridge soft switching converter is designed and verified in [10]. A DC-DC converter is proposed in [11] for solar energy-hydrogen conversion system and as the electrolysis load is nonlinear, FLC is a choice in place of other linear controllers. In [2, 12] digital design and hardware implementation of FLC using DSP is demonstrated. The buck converter prototype considered in this paper is designed for 24V to 12V conversion. The complete designing and analysis of buck converter is described and verified using WEBENCH software tool in [13]. The implementation of linear controllers on DSP is simple and straightforward but implementation of FLC based non- linear controller is a typical task due to huge calculations and complexities involvement. FLC based controller is implemented on eZdsp F2812 module with TMS320F2812 DSP for Boost converter is conferred in [2]. Robustness of FLC is demonstrated in presented work by obtaining results for practical operating conditions of DC-DC converter such as, varying load resistance and varying input voltage. The proposed FLC can be used for other configurations of DC-DC converters without or small modifications in rule base. A digital design of FLC is demonstrated in this paper for implementation on digital signal processors. II. FUZZY LOGIC CONTROLLER Artificial intelligence techniques are adapted for solving complex control problems. Fuzzy logic is one of the artificial intelligence technique which provides expedient method for designing nonlinear controllers from heuristic information. Conventional controllers are designed on behalf of closed- loop control characteristics such as: stability, rise time, settling time, overshoot and steady state error. Frequency response measurement, root-locus plot, state space method etc. are used for linear conventional controller design, which needs in-depth knowledge of mathematical model of system. FLC is quite different from conventional control, as it is based on expert knowledge of system. FLC provides a simple methodology to represent and implement human’s expertise Kartik Sharma 1 * Department of Electrical Engineering Rajasthan Technical University, Kota Rajasthan, India 1 *kartiks.eemtech@rtu.ac.in Dheeraj Kumar Palwalia 2 Department of Electrical Engineering Rajasthan Technical University, Kota Rajasthan, India 2 dkpalwalia@rtu.ac.in Proceedings of the 4th International Conference on “Signal Processing, Computing and Control” ISPCC- 2k17; IEEE Conference ID: 40546, 21st-23rd September, 2017, Jaypee University of Information Technology, Waknaghat, India Copy Right © ISPCC 2K17 ; ISBN 978-1-5090-5837-2