A Fast Photovoltaic Simulator Based On D/A Mixed Control

Due to the worldwide energy crisis, the development and utilization of renewable energy is imperative. Solar energy is a kind of renewable energy with attractive development prospects. In order to aid the research of photovoltaic (PV) inverters and PV optimizers, and to adapt the higher and higher MPPT disturbance frequency in recent years, this paper designs a PV simulator based on digital/analog mixed control, aiming to produce PV characterized outputs in place of actual PV modules.This paper firstly presents PV cells’equivalent circuit and the simplified PV characteristic equation for engineering application, then presents several PV characteristic curve fitting methods to simplify the computing and solving process. This PV simulator adopts the four-segment piecewise linear fitting method to build its PV characteristic equation.PV simulators use specific tracking strategies to adjust its output in order to operate on the pre-set PV characteristic curve. This paper researches PV characteristic tracking method thoroughly for different load types and different control strategies, and uses simulation experiments to verify the research results. Based on the above findings, for resistive load, the simulator uses the numerical solution method to track the PV characteristic curve.The simulator uses synchronous BUCK converter as the main power circuit topology, and adopts the voltage and current double-loop control. The inner current loop is realized using the peak current control method. Digital/analog control method is adopted to improve the simulator’s dynamic behavior. The analog control system consists of the peak current control system, the PI circuit, the sampling circuits, etc. The digital control system uses the PV characteristic equation to solve the simulator’s output voltage reference which is placed in PWM form. Digital-to-analog conversion is realized using a second-order filter circuit.Finally, this paper theoretically analyzes the simulator’s hardware and software response speed. The total response time is0.2ms.The results of simulation and experiments indicate that the theoretical analyses are accurate and the proposed simulator is feasible.