Chip‐implementation of a self‐tuning nonlinear function control for DC‐DC converters

Abstract

PurposeThe DC‐DC converters which convert one level of electrical voltage to the desired level are widely used in many electrical peripherals. During the past two decade, many different control laws have been developed. The proportional‐integral (PI) control and sliding‐mode control have been carried out for the DC‐DC converters since they are simple to implement and easy to design. However, its performance using PI control and sliding‐mode control is obviously quite limited. The purpose of this paper is to a self‐tuning nonlinear function control (STNFC) propose for the DC‐DC converters. The adaptation laws of the proposed STNFC system are derived in the sense of Lyapunov function, thus not only the controller parameters can be online tuned itself, but also the system's stability can be guaranteed.Design/methodology/approachIn general, the accurate mathematical models of the DC‐DC converters are difficult to derive. This paper proposes a model‐free STNFC design method. Since the proposed STNFC uses a simple fuzzy system with three fuzzy rules base to implement the control law, the computational loading of the fuzzy inference mechanism is slight. So the proposed STNFC system is suitable for the real‐time practical applications. The controller parameters of the proposed STNFC system can online tune in the Lyapunov sense, thus the stability of closed‐loop system can be guaranteed.FindingsThe proposed STNFC system is applied to a DC‐DC converter based on a field‐programmable gate array chip. The experimental results are provided to demonstrate the proposed STNFC system can cope with the input voltage and load resistance variations to ensure the stability while providing fast transient response.Originality/valueThe proposed STNFC approach is interesting for the design of an intelligent control scheme. The main contributions of this paper are: the successful development of STNFC system without heavy computational loading. The parameter‐learning algorithm is design based on the Lyapunov stability theorem to guarantee the system stability; the successful applications of the STNFC system to control the forward DC‐DC converter. And, the proposed STNFC methodology can be easily extended to other DC‐DC converters.