Research Of The Double Loop Non-ideal Proportional Resonant Control Inverter

As core component, inverters have been widely used in uninterruptible power supply, photovoltaic power system and so on. The inverter control algorithms determine the performance of the entire power supply system. Currently, the PID controller is unquestionably the most common way for inverter control techniques, but it can’t eliminate the steady-state static error of the controller. To solve this problem, this paper presents a double loop non-ideal proportional resonant control and designs the experimental prototype based on dsPIC to prove the control algorithm. The experiment results show that the control algorithm can achieve non-static error control. The main contents are as follows:Firstly, the mathematical model of the single-phase inverter is analyzed by using the average state model of unipolar SPWM. Based on a simplified model, the design of the inverter is divided into three sections:filter, inner current loop controller and outer voltage loop controller. On this basis, the design ideas and methods of the LC filter are analyzed; the performance of proportion resonant controller and non-ideal proportion resonant controller is compared. Non-ideal proportion resonant control algorithm is introduced to the double-loop control of inverter, which inner current loop and outer voltage loop are also presented in their open-loop, closed-loop frequency characteristics and non-ideal proportion resonant no static error control performance.Furthermore, the double closed-loop non-ideal proportion resonant inverter’s Matlab simulation model is established. When the inverter is in full load, half load, the resistance inductive load and nonlinear load cases, the simulation is running in steady-state and dynamic operation to verify the effect of double closed-loop non-ideal proportion resonant controller.Finally, the hardware circuit of inverter is designed, and the control circuit, the sampling circuit, the drive circuit, the sampling signal filter circuit are analyzed. Microcontroller dsPIC becomes the core of software design and the experimental prototype verifies the control algorithm.