Modeling And Digital Control Of DC-DC Converter

With the rapid development of society and economy, the problem of energy has become strained increasingly. The new energy attracts international academic and engineering attention. DC-DC converter is the core component of the energy management in the field of hotovoltaic inverter, hybrid electric vehicles and so on. Traditional switching power supply is mainly used for maintaining constant input and output. However, it cannot realize very good control effect when the input or output changes in a large range. Based on the rapid development of power electronics technology, digital power supply has been used widely because of its outstanding advantages. It can be set flexibly, be adjusted online and monitor status. The existing digital power based on compensation network or PID control cannot consider the main citcuit topology and disturbance characteristics, so it’s difficult to improve the performance of digital power supply effectively. Therefore, it’s extremely necessary to do much deeper research of the digital control for DC-DC converter.This paper aims to design a digital control method based on LQR steady-state optimal control for DC-DC converter. The control method is designed to solve several problems in the practical application. The rapid prototype software and hardware platform of DC-DC converter are built integrally. Buck DC-DC converter and SEPIC DC-DC converter can be controlled efficiently and steadily. The work in this paper is summarized as follows:(1) The traditional compensation network control method cannot realize excellent control performance when load changes in a large range. In order to solve this problem, a new digital control method based on LQR steady-state optimal control for DC-DC converter is proposed. The output error is considered as an extension state variable to introduce integral action and eliminate the output tracking error. Steady-state LQR and steady-state Kalman are introduced to calculate feedback gain matrixes offline in order to meet ultra-high frequency control of DC-DC converter. Anti-windup and mode switching of controllers are also designed to handle saturation problem which is caused by constraints of duty cycle.(2) The digital control of Buck DC-DC converter and SEPIC DC-DC converter are realized. The large-signal model and small-signal model of Buck DC-DC converter and SEPIC DC-DC converter are built based on the average state space method. Based on the NI CompactRIO hardware experiment platform, the digital LQR steady-state optimal control method and the compensation network control method have been implemented for Buck DC-DC converter and SEPIC DC-DC converter. The selection method of the weighting parameters of the digital LQR steady-state optimal control method has been proposed based on plenty of experimental results. The experimental results also clearly show the better performance of the proposed method compared to the traditional method.(3) The rapid prototype software and hardware platform of DC-DC converter is built based on NI CompactRIO. The pc control programs and the lower computer programs of the digital LQR steady-state optimal control are completed by LabVIEW. The programs are also optimized by IP Builder to improve the precision and efficiency.