Research On Prediction Model And Algorithm Of Building Photovoltaic Power Generation

Photovoltaic power due to its broad development prospects,less constraints on development conditions,and the advantages of large-scale development.Combining intelligent photovoltaic systems with buildings can provide power to buildings in remote areas and enable tracking of photovoltaic power maximums.In practice,due to large fluctuations in photovoltaics,intermittent output power,and changes in the structure and operation of the main grid due to the increase in the penetration rate of building photovoltaic power generation systems in the distribution network,in order to reduce the impact on the main grid The purpose is to predict the power of the photovoltaic power generation system in advance,especially to optimize the capacity and location of the distributed photovoltaic power generation system in the distribution network.Therefore,this paper studies the photovoltaic power supply scheme for smart buildings,increases the friendliness of the power grid and the stability of the power supply mode,and realizes the intelligent dispatchability of photovoltaic inverters and the diversification of functions.,predict the collected samples and analyze the results of the error;in terms of optimal configuration,use the improved multi-objective particle swarm algorithm(MOPSO)to find the optimal configuration model of distributed photovoltaic.Aiming at the problem of mode switching under different solar radiation intensities,this paper proposes a scheme for building photovoltaic power generation systems.The working principle of building photovoltaic grid-connected power generation is that the rooftop solar panels receive light radiation,and the top-level photovoltaic outputs DC power,which is converted into AC power with the same frequency and phase as the city grid voltage through the inverter,while achieving the purpose of AC grid-connection..Considering that distributed photovoltaics in smart buildings are greatly affected by various problems,their power output has the characteristics of instability and large fluctuations,which will cause the system to be unstable when it is integrated into the power grid.Therefore,this paper proposes a random GPR Photovoltaic Power Generation Forecasting Method under Subspace Integration(SE-GPR).in this paper has greater advantages in prediction accuracy than the classic GPR prediction model and other prediction models.When distributed photovoltaics are integrated into the distribution network in smart buildings,it will have a relatively large impact on the power loss and power flow of the lines.Considering the importance of photovoltaic configuration,it will be important to optimize the distributed photovoltaic configuration.In this paper,a particle swarm optimization algorithm is improved and combined with the Pareto solution of the multi-objective optimization problem,an improved multi-objective particle swarm algorithm(MOPSO)is designed to solve the multi-objective optimization configuration model of photovoltaics.The effectiveness of this method is verified on grid nodes.