Research On Control Strategies For Three-level PV Grid-connected Inverter

With the situation of global energy crisis and environment pollution becoming more and more serious, the whole world pays more attention on exploiting and utilizing the solar energy. The PV grid-connected system is the main utilization of the solar energy, so it will be developed rapidly. Under this background, the topology and control strategies of the PV grid-connected inverter are researched deeply in this paper.Traditional two-level inverter has disadvantages of large switching loss, low DC voltage utilization rate and high output current harmonics, so it couldn't fulfill high quality of the gird-connect system. Three-level NPC inverter is introduced for the gird-connect system in this paper. SVPWM modulation scheme of Three-level inverter is introduced in detail. Neutral-point potential unbalance is an inherent problem of three-level NPC inverter, so a neutral-point control method which is based on directions of the three-phase output current and the neutral-point voltage to regulate the action times of positive and negative small vector is adopted. Availability of the method is testified by the simulation results.The mathematic model in ABC three-phase static coordinate of the three-level inverter is set up. Using coordinate transformation, the mathematic model in dq two-phase rotating coordinate is established. To actualize the gird-connected current tracking control, this paper presents a deadbeat repetitive predictive current control scheme in dq two-phase rotating coordinate. The state observer is designed to obtain the predictive output current value and compensate the time delay of digital control system. At the same time, the current repetitive predictor is adopted to achieve better prediction of the grid-connected current with the same sampling frequency. This scheme has merits of reducing output current harmonics, improving the effect of time delay and the overall performance of the system. The feasibility and correctness of the proposed control strategy are verified by simulation results.