| In recent years,many countries have vigorously developed renewable energy to deal with energy crises and environmental problems.The source-grid-storage system has the advantages of smoothing output power fluctuations and ‘peak load shifting’,but the energy storage system has a dual nature of load and power source,making the relationship among the source,grid,and storage more complex.Currently,most research only considers the influence of single constraints,such as economic costs or voltage regulation,on the planning of photovoltaic-energy storage systems.Moreover,long-term planning models with economic targets are mostly studied,and research on short-term operation optimization is lesser.To address the shortcomings of previous research,we propose an investment-optimization two-layer planning model for the source-grid-storage system,which considers both long-term cost optimization and short-term operational benefit optimization.We aim to achieve the most optimal utilization of the source-grid-storage systems.The main research works and contributions of this paper include:(1)A simulation analysis of distributed photovoltaic grid connection with different connection locations and capacities.The results show that distributed photovoltaic grid connection can improve the node voltage level and reduce active power loss,which set a foundation for the research of distributed photovoltaic site selection and capacity planning in our following research.(2)Gray Wolf Optimization,as a new intelligent optimization algorithm,has been applied in reactive power optimization,load forecasting,and other power system fields with good results.But it also suffers from the shortcomings of premature convergence and the tendency of falling into local optima.To address these issues,we propose an improved Tent-Levy speedy gray wolf optimization(TLS-GWO)model,which is based on tent map and levy flight and use it to solve the two-stage distributed photovoltaic planning configuration problem.The simulation experiments are carried out on the IEEE33 node distribution network model and compared with those by the Adaptive Genetic Algorithm and the unimproved Wolf Optimization algorithm.The results show that our proposed algorithm has superiority in solving the distributed photovoltaic planning problem.At the same time,the planning results are also present that a reasonable configuration scheme can effectively reduce investment costs,active power loss,and voltage deviation,which is conducive to improving the stability of the distribution network system operation.a better distributed photovoltaic planning scheme and obtaining better distributed photovoltaic planning results.(3)We propose an investment-optimization two-layer planning model for the source-grid-storage system,which considers both the long-term(comprehensive cost optimization of the distribution network)and the short-term(daily operational benefit optimization of the energy storage)benefits.The comprehensive consideration of the long and short-term aspects can make the model more economically efficient.The TLS-GWO algorithm and the YALMIP toolbox are used to solve the model,and the simulation is carried out on the IEEE33 node distribution network model and compared with the Adaptive Genetic Algorithm.The results show that the proposed source-grid-storage planning model is optimal for investment benefits and can fully leverage the ‘peak load shifting’ effect of energy storage,which has a positive effect on the stable operation of the distribution network.The investment-oriented optimization of the source-grid-storage planning scheme studied in this paper can not only obtain the investment-optimal photovoltaic energy storage configuration scheme and the optimal operation state of the energy storage,but also has important theoretical significance and engineering application value for the distribution network company’s future large-scale planning and investment in photovoltaic energy storage systems. |
PDF Full Text Request