Research On Domestic Single-phase Grid-connect Photovoltaic Inverter Control Technologies

Photovoltaic system as a kind of clean energy has developed fast in the recent years.A typical photovoltaic system consists of photovoltaic cell,inverter,and tracking system,it aims to make the photovoltaic system to convert the solar energy to the electrical energy at the highest efficiency.Therefore,the main task of this thesis is to research on the high-performance single-phase grid-connected photovoltaic-system control strategies,and verify the validity of the control strategies through simulation.At first,the equivalent circuit model of photovoltaic cell is established by its physical characteristics.Then the ideal mathematical model of photovoltaic cell can be deduced.Due to there are many characteristic parameters in the ideal mathematical model of photovoltaic cell,but these parameters are difficult to be obtained directly from the manufacturer,therefore,a practical mathematical model of photovoltaic cell is adopted to make the simulation model for photovoltaic cells.Next,in order to make the photovoltaic system work at the optimum condition,the maximum power point tracking control algorithms are reviewed,and an incremental inductance based global maximum power point search algorithm is proposed.The incremental inductance is treated as the input of the algorithm,and the output duty-cycle information is generated by an integrator.Comparing with the traditional perturb & observe algorithm,the proposed algorithm not only reduces the oscillation when the system working in the vicinity of the maximum power point,but also realizes the maximum power point tracking under partial shading condition.In order to feed the electrical energy,which is generated by photovoltaic system,to the grid with high-quality,grid-connected control algorithms are investigated.According to the comparison,the dead-beat control based grid-connected predictive current algorithm is adopted to control the output current of the grid-connected inverter.Theoretically,the proposed algorithm can achieve excellent current control performance,i.e.The current feedback can track its reference after two control periods.Finally,the reasons and consequences of islanding phenomenon are analyzed,according to the existing international islanding detection standards,a harmonic-injection based islanding detection algorithm is proposed in this thesis.In this algorithm,the order of the injected harmonic can be chosen according to the power quality requirements,and the magnitude of the harmonic voltage can be extracted by a novel Goertzel discrete Fourier transformation method.Theoretically,Goertzel algorithm can significantly reduce the calculation time for discrete Fourier transformation.