Research On Control Strategy Of Photovoltaic Energy Storage Grid-Connected System Based On Artificial Neural Network

Since 2010,photovoltaic power generation has entered a stage of large-scale development,and the installed capacity of grid-connected has grown rapidly.However,the uncertainty and volatility of photovoltaic power generation are easy to cause great harm to the power grid,and the construction and absorption mechanism of the power grid lag behind,and the phenomenon of "abandonment of light" occurs,which peaks around 2016.New energy construction in areas with serious "abandonment of light" has come to a standstill,resulting in many photovoltaic power plants passively "basking in the sun".In order to improve the absorption capacity of the power grid,vigorously promote the development of energy storage.In this paper,the output power of photovoltaic power generation is smoothed by the battery energy storage system,and the LCL grid-connected inverter is used to convert the direct current generated by the photovoltaic into alternating current,and on the basis of the traditional control strategy,a Radial Basis Function Neural Network(RBF)adaptive control strategy is proposed,the main content is as follows:Firstly,the Perturbation observation method is selected as the algorithm for realizing Maximum Power Point Tracking(MPPT)in this paper,and the Boost circuit is used as the output circuit of the photovoltaic cell,and the parameters of the Boost circuit are designed.The topology of the battery energy storage system and the double closed-loop control block diagram of voltage and current are established,and the parameters of the voltage outer loop controller and the current inner loop controller are designed according to the corresponding constraints,and this parameter is used as the initial parameter of the RBF neural network adaptive control strategy,and the online adjustment of the controller parameters is realized through the online learning of the RBF neural network,so that the controller still has a good control effect when the operating state of the system changes.Then,the LCL filter is used to filter out a large number of harmonics generated in the process of inverter,and the active damping method of capacitor current proportional feedback and the grid voltage full feedforward strategy are used to offset the resonance spike of the LCL filter and the interference of the background harmonics of the power grid.For traditional controllers,there is a steady-state error of phase and amplitude when tracking sinusoidal quantities,so RBF neural network is introduced to tune controller parameters to improve the adaptive ability of the controller and realize no static difference tracking of grid-connected current.Finally,the simulation model of photovoltaic energy storage grid-connected system is built.Simulation experiments were carried out under different working conditions to verify the feasibility and effectiveness of the control strategy proposed in this paper,and the results show that when the photovoltaic cells are working normally,the LCL grid-connected inverter and energy storage system can achieve the purpose of smoothing the photovoltaic output power,stabilize the DC bus voltage,ensure the power balance in the system,and the total harmonic distortion rate is less than 5%,which meets the grid-connected requirements.When the photovoltaic power generation is stopped,the RBF neural network adaptive control strategy can still achieve smooth photovoltaic output power,stabilize the DC link voltage,ensure the power balance in the system,and the total harmonic distortion rate is less than 5%.